JP7807701B2 - Type discrimination device - Google Patents

Description

The present invention relates to a type discrimination device that automatically discriminates the type of medicine, a medicine sorting device that sorts medicines, etc.

Conventionally, returned medications of multiple types have been sorted by type by pharmacists or doctors. The returned medications are either prescribed to various patients or have been dispensed. Therefore, compared to the dispensing process of assembling (packaging) (one or more types of) medications (tablets) for each dosage period from groups of medications (medication cassettes) pre-organized by medication type in dispensing equipment, etc., based on prescription information for each patient, the number of types of medications returned together, which are prescribed for multiple patients, is significantly greater. Therefore, it would be highly useful to automatically sort and reuse returned medications. Note that the number of medications dispensed for each dosage period is generally around two to three, and at most around ten.

However, to avoid the time and effort required for sorting, or the risk of mis-administration due to incorrect sorting (putting the wrong medication back into the medication cassette), some pharmacies or hospitals (or more accurately, their in-house pharmacy departments) simply discard returned medications.

Patent Document 1 discloses a drug sorting device that automatically recognizes and stores returned ampoules or vials. This drug sorting device recognizes the orientation and posture of the ampule or vial, as well as the properties of the ampule or vial (e.g., shape, size, type, and expiration date). It then associates the identification information of each ampule or vial with the storage area set for that ampule or vial when it is stored based on the recognized size of the ampule or vial, and individually arranges the ampoules or vials, thereby storing them so that they can be removed.

International Publication No. 2015/170761 (Published November 12, 2015)

Patent Document 1 covers ampoules or vials as items for return, not tablets, capsules, or other medications that are not contained in a container, or medications themselves that are not packaged. Therefore, Patent Document 1 does not anticipate identifying such medications (e.g., tablets or capsules) themselves, or even automatically sorting such medications.

One aspect of the present invention aims to realize a type discrimination device that can identify the type of medicine that is not contained in a container, such as a tablet or capsule, and can improve the accuracy of such discrimination.

Another aspect of the present invention is to realize a medicine sorting device that can sort medicines that are not contained in containers, such as tablets or capsules, and can improve the efficiency of such sorting.

Another aspect of the present invention aims to realize a type discrimination device or the like that can discriminate the type of medication that is not contained in a container such as a tablet or capsule, and that can reduce the image data volume required to create the image used to discriminate the type of medication.

In order to solve the above problem, one aspect of the present invention provides a type discrimination device that discriminates the type of a target drug from an image of the target drug, and includes: a feature extraction unit that extracts, from the image, first feature information that indicates the external characteristics of the target drug, including at least reflection area information that indicates the reflection area of light irradiated onto the target drug; and a discrimination unit that compares the first feature information with second feature information that indicates the external characteristics of the reference drug, including at least reflection area information that indicates the reflection area of light irradiated onto a reference drug, and discriminates the type of the target drug based on the comparison result.

A type discrimination device according to one aspect of the present invention is a type discrimination device that discriminates the type of a target drug from an image of the target drug, and includes: a feature extraction unit that extracts third feature information indicating the external characteristics of the target drug, which includes at least reflection information indicating whether light irradiated onto the target drug is reflected by the target drug, from the image; and a discrimination unit that compares the third feature information with fourth feature information indicating the external characteristics of the reference drug, which includes at least reflection information indicating whether light irradiated onto a reference drug is reflected by the reference drug, and discriminates the type of the target drug based on the result of the comparison.

A type discrimination device according to one aspect of the present invention is a type discrimination device that discriminates the type of a target drug from an image of the target drug, and includes a discrimination unit that compares fifth feature information indicating the external characteristics of the target drug, including target color information indicating the color of the target drug, extracted from the image, with sixth feature information including reference color information indicating the color of a reference drug, and discriminates the type of the target drug based on the comparison result; and an update unit that, if the discrimination unit fails to discriminate the type of the target drug, updates the target color information used for the discrimination as new reference color information after receiving an operation from a user.

A type discrimination device according to one aspect of the present invention is a type discrimination device that discriminates the type of a target drug from an image of the target drug, and includes: a feature extraction unit that extracts seventh feature information indicating characteristics of the appearance of the target drug from the image, the seventh feature information including at least identification information displayed on the surface of the target drug and first position information indicating the relative position of a display area of the identification information with respect to the image of the target drug; and a discrimination unit that compares the seventh feature information with eighth feature information indicating characteristics of the appearance of the reference drug, the eighth feature information including at least specific drug information indicating whether a reference drug is a specific drug, the identification information displayed on the surface of the reference drug, and second position information indicating the relative position of a display area of the identification information with respect to the image of the reference drug in the image of the reference drug, and discriminates the type of the target drug based on the comparison result; and when the specific drug information indicates that the reference drug is a specific drug, the discrimination unit determines whether the discriminated target drug is the same type of drug as the reference drug without using the first position information and the second position information to discriminate the type of the target drug.

A drug sorting device according to one aspect of the present invention includes a first storage unit that stores multiple types of drugs, a second storage unit that stores the drugs by type, a temporary storage unit that temporarily stores drugs that are not stored in the second storage unit, an imaging unit that captures images of the target drugs removed from the first storage unit, a discrimination unit that determines the type of the target drugs from the captured image, a priority determination unit that determines a priority for each type of drug based on the frequency of discrimination in the discrimination history of the discrimination unit, and a sorting determination unit that determines to temporarily store the target drug in the temporary storage unit if the priority corresponding to the type of target drug discriminated by the discrimination unit is lower than a predetermined value.

In addition, a type discrimination device according to one aspect of the present invention is a type discrimination device that discriminates the type of a target drug from an image of the target drug, and includes an image processing unit that generates at least one cut-out image from an image including the target drug, the cut-out image including a first region that is the region of the image of the target drug and a second region that is the region of the image of the table on which the target drug is placed, in order to generate an image of a reference drug for discriminating the type of the target drug.

An image generation system according to one aspect of the present invention includes: (1) a type discrimination device that discriminates the type of a target drug from an image of the target drug; and (2) an information processing device that can be communicatively connected to the type discrimination device and generates an image of a reference drug for discriminating the type of the target drug. The type discrimination device generates a cut-out image by cutting out, from an image containing an image of the target drug, a first region that is the region of the image of the target drug or a second region that is the region of an image of a table on which the target drug is placed, including the first region. The information processing device generates the reference image based on the cut-out image generated by the type discrimination device.

An image generation system according to one aspect of the present invention includes: (1) a type discrimination device that discriminates the type of a target drug from an image of the target drug; and (2) an information processing device that can be communicatively connected to the type discrimination device and generates an image of a reference drug for discriminating the type of the target drug. The type discrimination device generates at least one of cut-out images obtained by cutting out a first region, which is the region of the image of the target drug, and a second region, which is the region of the image of a table on which the target drug is placed, including the first region, from an image containing the image of the target drug. The information processing device generates the reference image based on the cut-out image generated by the type discrimination device.

A type discrimination device according to one aspect of the present invention can improve the accuracy of drug discrimination.

Furthermore, a medicine sorting device according to one aspect of the present invention can improve the efficiency of medicine sorting.

Furthermore, with the type discrimination device according to one aspect of the present invention, it is possible to reduce the image data volume required to create images used to discriminate the type of medication.

Furthermore, with the image generation system according to one aspect of the present invention, it is possible to reduce the image data volume required to create the image used to identify the type of medication.

1 is a block diagram showing the overall configuration of a medicine sorting device. FIG. 1 is a diagram illustrating an example of the configuration of a medicine sorting device. 1A and 1B are perspective views showing the overall configuration of an imaging unit and an example of a medicine placing table, respectively. FIG. 10 is a diagram for explaining the rotation of the imaging unit. FIG. 2 is a diagram schematically illustrating the configuration of an imaging unit. FIG. 10 is a diagram showing an example of a reflection region of light reflected by a target drug in a captured image. 10 is a flowchart illustrating an example of a determination process performed by a determination unit. FIG. 10 is a diagram showing an example of the thickness of a target drug extracted by a feature extraction unit. 10 is a flowchart showing an example of an update process of a medicine database update unit. 10 is a flowchart illustrating an example of a determination process performed by a determination unit. FIG. 10 is a diagram showing an example of the relative position of a display area for identification information with respect to an image of a target drug or a reference drug. FIG. 10 is a diagram showing an example of a comparison between the relative position of a display area of identification information with respect to an image of a target drug and the relative position of a display area of identification information with respect to an image of a reference drug. 10 is a flowchart showing an example of a medicine sorting process of the sorting control unit. 10 is a flowchart illustrating an example of a print control process of a print output control unit. 10 is a flowchart showing an example of a malfunction determination process of the medicine sorting device. FIG. 1 illustrates an example of an image generation system. 10 is a flowchart illustrating an example of a cutout process performed by an image generating unit. FIG. 10 is a diagram illustrating an example of a cut-out image. 10A and 10B are diagrams illustrating an example of a process of cutting out an image from a captured image performed by an image processing unit 73. FIG. FIG. 2 is a block diagram showing an example of a configuration of a part of the medicine sorting device. 10 is a flowchart showing an example of a process for generating a shape model of a sorting cup performed by the medicine sorting device. 10A and 10B are diagrams illustrating an example of a captured image of a sorting cup and a shape model of the sorting cup. 10 is a flowchart showing an example of a medicine detection process for checking remaining medicines by the medicine sorting device. 10A and 10B are diagrams showing an example of a captured image of a sorting cup, an extracted image of the inside of the sorting cup, a shape model, and an image of a matching region in the extracted image. 10 is a flowchart showing an example of a printing process of the medicine sorting device. FIG. 1 is a diagram illustrating an example of a drug discrimination system. 10 is a flowchart showing an example of a discrimination process of the medicine sorting device. 10 is a diagram showing an example of identification information of a target drug, identification information of a reference drug in a user's adopted drug list, and identification information of a reference drug in a comprehensive drug database. FIG. FIG. 4 is a diagram illustrating an example of a first storage section. FIG. 10 is a diagram showing an example of an initial image (initial screen) on which the results of medicine sorting can be displayed. This figure shows an example of a setting image (setting screen) that allows the user to set the method of storing brought-in medications in the first storage section when the brought-in medication sorting mode is set. 10 is a diagram showing an example of a sorting image for displaying the sorting status of brought-by medications when the brought-by medication sorting mode is set. FIG.

[Embodiment 1]
(Overview of medicine sorting device 1)
First, an overview of the medicine sorting device 1 will be described using Figures 1 and 2. Figure 1 is a block diagram showing the overall configuration of the medicine sorting device 1. Figure 2 is a diagram showing an example of the configuration of the medicine sorting device 1, with reference numeral 2001 being a perspective view of the medicine sorting device 1 and reference numeral 2002 being a perspective view showing the basic configuration of the medicine sorting area 2. As shown in Figure 1 and 2001 and 2002 of Figure 2, the medicine sorting device 1 includes the medicine sorting area 2, a touch panel 3, a print output unit 4, and a packaging mechanism 6.

The medicine sorting device 1 takes an image of each of multiple types of medicine, determines the type of medicine based on the image obtained from the image, and sorts the medicines by type. Specifically, this process is performed in the medicine sorting area 2. Details of the medicine sorting area 2 (the internal configuration of the medicine sorting device 1) will be described later. After the medicines have been sorted by type, they are visually inspected by the user and then packaged or returned to the medicine shelf or packaging machine.

In this embodiment, the multiple types of medications are medications that are not contained in a container or are not packaged, and examples of these are tablets or capsules. The multiple types of medications are also described as returned medications. These medications are medications used by a pharmacy or hospital that are returned to the pharmacy or hospital as "returned medications." However, the multiple types of medications may also be "brought-in medications" that may include medications issued by other pharmacies or hospitals in addition to the medications used by the pharmacy or hospital. After the medications are returned, the medication sorting device 1 can automatically perform processes from imaging to sorting.

The touch panel 3 accepts various user inputs via the operation unit 31 and displays various images (e.g., images showing the progress of drug sorting, images for visual inspection) on the display unit 32.

The print output unit 4 prints a journal representing drug data (e.g., data indicating the drug name, manufacturer, or ingredients) related to the drug after visual inspection, in accordance with user input after visual inspection. The drug data may include image data representing a drug-specific image.

The packaging mechanism 6 packages the sorted medications. The packaging mechanism 6 is an optional mechanism. When the packaging mechanism 6 is provided in the medication sorting device 1, the medication sorting device 1 can perform all processes, from sorting returned medications to packaging after visual inspection, all at once. In particular, when medications are fed into the packaging mechanism 6 by the transport/sorting unit 12, the above processes from sorting to packaging can be performed automatically, except for visual inspection.

The packaging mechanism 6 can be a packaging unit of a conventional tablet packaging machine or powder packaging machine. In this case, for example, medications in a sorting cup 141 sorted by drug type can be packaged into one or more packets.

Please note that throughout this specification, the term "packaging," as used at least in the description of the medicine sorting device 1, encompasses both the meaning of "packaging medicines according to their dosing period based on prescription data" and the meaning of "simply packaging the medicines sorted into the second storage section 14 regardless of the prescription data."

The medicine sorting device 1 also includes a first RFID (Radio Frequency Identifier) reader/writer unit 5. The first RFID reader/writer unit 5 is provided on the medicine removal side of the base 19, as shown at 2002 in Figure 2.

The first RFID reader/writer unit 5 reads data related to the medications stored in each sorting cup 141, which is stored on an RFID tag (not shown) attached to the bottom of each sorting cup 141 in the second storage section 14. This data includes, for example, data indicating the number of medications stored, medication identification data (e.g., GS1 code) for identifying the medication, data indicating the type of sorting cup 141, and sorting cup identification information (identification information attached to the RFID tag) for identifying the sorting cup 141. The medication data (e.g., medication name) of the medications stored in the sorting cup 141 and image data acquired by the imaging unit 13 are linked to the sorting cup identification information and stored in the medication database 81.

The data may also include drug data determined by visual inspection (drug data after visual inspection). Furthermore, the drug data after visual inspection may be written to the RFID tag. The drug data after visual inspection is used when the drugs stored in the corresponding sorting cup 141 are (1) packaged by the packaging mechanism 6 or a packaging machine other than the drug sorting device 1, or (2) returned to the drug shelf. These drug data may also be linked to the sorting cup identification information and stored in the drug database.

In addition, as shown in 2001 in Figure 2, the medicine sorting device 1 is equipped with an opening/closing shutter 51 and an opening/closing door 52 that allow the medicine removal side to be opened and closed.

(Basic configuration of medicine sorting area 2)
Next, the basic configuration of the medicine sorting area 2 (internal configuration of the medicine sorting device 1) will be described with reference to Figures 1 and 2002 in Figure 2.

As shown in 2002 in Figures 1 and 2, the drug sorting area 2 mainly comprises, as hardware, a first storage section 11, a transport/sorting unit 12 (sorting section), an imaging unit 13, a second storage section 14, a waiting tray 15, a collection tray 16, a drug insertion port 17, and a second RFID reader/writer unit 18. Each component except for the transport/sorting unit 12 is mounted on a base 19. The main functions of the transport/sorting unit 12, the imaging unit 13, and the second RFID reader/writer unit 18 will be explained in detail in the explanations of each process below.

The first storage unit 11 stores a mixture of multiple types of medications returned by users. In this embodiment, the first storage unit 11 is divided into multiple storage units. In this case, for example, when all of the medications stored in one storage unit are transported by the transport/sorting unit 12, the medications stored in the storage unit adjacent to that storage unit become the next to be transported. The first storage unit 11 may also be rotatable about the Z axis (the center of the cylindrical shape). In this case, the control unit 60a of the computer (type determination device) 60 may rotate the first storage unit 11, for example, when one storage unit becomes empty, so that the transport/sorting unit 12 can easily retrieve the medications.

The second storage section 14 includes multiple sorting cups 141 that store medications sorted by type. The control section 60a determines the type of medication based on the image of the medication captured by the imaging unit 13, and determines the sorting cup 141 in which to store the medication based on the determination results. The medication is transported to and stored in the determined sorting cup 141 by the transport/sorting unit 12.

The waiting tray 15 is a temporary storage unit where medications are temporarily placed. For example, if all of the sorting cups 141 are filled with medications, medications determined by the control unit 60a to be of a type other than those stored in the waiting tray 15 are temporarily placed in the waiting tray 15. In this case, after the medications are removed from the sorting cup 141, they may be transported from the waiting tray 15 to that sorting cup 141.

In addition, in this embodiment, a suspected drug (described below) that is suspected to be a drug may be temporarily placed on the waiting tray 15. If a suspected drug is temporarily placed, the suspected drug is transported from the waiting tray 15 to a predetermined area in the second storage unit 14, for example, depending on the determination result of the control unit 60a.

The collection tray 16 is a storage unit that stores items whose type cannot be determined by the control unit 60a (e.g., foreign matter other than medication). Examples of foreign matter other than medication include pieces of PTP (Press Through Pack) sheets. These pieces of PTP sheets may become mixed into the first storage unit 11 when medication is returned. The control unit 60a also stores in the collection tray 16 medications that are registered in the medication database 81 as medications to be discarded, or medications that the user wishes to discard (e.g., medications with old manufacturing dates).

When the medicine sorting device 1 is equipped with a packaging mechanism 6, the medicine input port 17 is used to transport medicines stored in the second storage section 14 to the packaging mechanism 6 by the transport/sorting unit 12. Naturally, when the medicine sorting device 1 is not equipped with a packaging mechanism 6, the medicine input port 17 is not required.

As shown in FIG. 1, the drug sorting device 1 also includes a computer 60 that controls the above-mentioned components (hardware). The computer 60 includes a control unit 60a and a memory unit 80. The control unit 60a includes a transport control unit 61, a sorting control unit (sorting determination unit) 62, an imaging control unit 63, a discrimination unit 64, a priority determination unit 65, an operation input unit 66, a display control unit 67, an RFID control unit 68, a print output control unit 69, a registration unit 70, a packaging control unit 71, a drug adsorption control unit 72, an image processing unit 73, a drug database update unit (update unit) 74, a notification control unit 75, and a lighting failure determination unit 76.

Here, the basic processing of the operation input unit 66, display control unit 67, RFID control unit 68, print output control unit 69, registration unit 70, and packaging control unit 71 will be described. The basic processing of the transport control unit 61, sorting control unit 62, imaging control unit 63, and discrimination unit 64 will be described in detail in the sections below titled (Drug transport processing to imaging unit 13), (Drug sorting processing), (Drug imaging processing), and (Image processing/discrimination processing). Processing by other components included in the control unit 60a will be described in the respective embodiments below.

The operation input unit 66 and display control unit 67 control the operation unit 31 and display unit 32 of the touch panel 3, respectively. The RFID control unit 68 controls the first RFID reader/writer unit 5 and the second RFID reader/writer unit 18. The print output control unit 69 controls the print output unit 4 in accordance with user input received by the operation input unit 66.

The registration unit 70 registers drug data relating to drugs for which the discrimination unit 64 has determined that no drug data corresponding to the image data exists in the drug database 81. Specifically, for drugs for which the discrimination unit 64 has determined that no corresponding drug data exists, the registration unit 70 links the captured image 82 of the drug with the drug data specified by the user and registers the linked image in the drug database 81.

When the medicine sorting device 1 is equipped with a packaging mechanism 6, the control unit 60a is equipped with a packaging control unit 71 that controls the packaging mechanism 6. The packaging control unit 71 controls the operation of the packaging mechanism 6. The packaging control unit 71 also controls the transport and sorting unit 12 to transport the medicines stored in the sorting cup 141 to the medicine inlet 17.

The medicine sorting device 1 may also be equipped with a barcode reader 7. The barcode reader 7 reads, for example, a barcode indicating the medicine data of the medicine printed on the medicine packet (packaging paper) into which the packaging mechanism 6 has packaged the medicine after sorting. This allows the control unit 60a to perform processing based on the read barcode (e.g., displaying the medicine data indicated by the barcode). The packaging mechanism 6 may also be equipped with a barcode printing mechanism (not shown) that prints, on the medicine packet, a barcode indicating the medicine data of the medicine packaged in the medicine packet.

The barcode reader 7 may be any reading device capable of reading information indicating drug data printed on a medicine packet or the like. Furthermore, the drug sorting device 1 does not necessarily need to be equipped with a barcode reader 7. When the control unit 60a performs processing based on the above information, the control unit 60a may obtain the above information by communicating with an external device equipped with a barcode reader 7.

The medicine sorting device 1 may also be equipped with a notification unit 8. The notification unit 8 may include, for example, a PATLITE (registered trademark) that emits light, a speaker that outputs sound, etc.

The computer 60 also includes a memory unit 80. The memory unit 80 pre-stores a drug database (drug master) 81 that manages drug data for multiple types of drugs, and stores captured images 82 and the like as sorting is performed by the drug sorting device 1. The captured images 82 are images captured by the first camera 131. The captured images 82 may also be images captured by the first camera 131 that have been subjected to image processing by the imaging control unit 63. Examples of the image processing include processing to generate an image to be analyzed by the discrimination unit 64 (an image that extracts drug characteristics and enables comparison with the characteristics of drugs (reference drugs) contained in the drug database 81). For example, processing to extract drug images from captured images may be included.

The various data stored in the memory unit 80 do not have to be managed by the memory unit 80 itself, but may instead be managed by, for example, an external device. In this case, the control unit 60a may obtain the various data from the external device via a communication line such as the Internet, as necessary. The drug database 81 may also be updated by adding new drug data.

The memory unit 80 may also include priority data 83, target drug color data 84, identification information change information 85, captured image discrimination information 86, and discrimination count information 88. Note that these components are not essential to the drug sorting device 1. Details of these components will be explained in each embodiment that requires them.

(Overview of processing in medicine sorting device 1)
In the medicine sorting device 1, the transporting/sorting unit 12 transports each medicine returned to the first storage unit 11 to the imaging unit 13. The imaging unit 13 sequentially captures images of each transported medicine. The control unit 60a determines the type of each medicine based on the captured images and determines a sorting position in the second storage unit 14 for each identified medicine. The transporting/sorting unit 12 transports each medicine to the determined sorting position. Information about the medicines stored in the second storage unit 14 is written to the RFID tag of the sorting cup 141, stored in the memory unit 80, or displayed on the touch panel 3. After or during the sorting of the medicines, the user operates the touch panel 3 to perform processes such as visual inspection and packaging. Each process will be described in detail below.

(Drug Delivery Process to Imaging Unit 13)
First, the process of transporting medicine from the first container 11 to the imaging unit 13 will be described with reference to 2001 in FIGS. 1 and 2. FIG.

Specifically, the transport and sorting unit 12 transports the medicines stored in the first storage section 11 to the receiving area Ar1 (see 3002 in Figure 3) where the imaging unit 13 receives the medicines. The transport control unit 61 controls the transport process performed by the transport and sorting unit 12.

The transport and sorting unit 12 includes a second camera 121, a suction and shutter mechanism 122, and a transport mechanism 123.

The second camera 121 sequentially captures images of the first storage unit 11 to identify the medication to be transported. The imaging control unit 63 controls the imaging process of the second camera 121. The second camera 121 is provided at the end of the transport/sorting unit 12 (specifically, of the housing including at least the suction/shutter mechanism 122) facing the base 19. The second camera 121 may also be provided at the tip of the suction mechanism described below. The imaging control unit 63 analyzes the captured image and determines whether the image contains medication. If it is determined that medication is included, the transport control unit 61, for example, brings the tip closer to the first storage unit 11 and identifies the medication included in the image captured at that time as the medication to be transported.

The suction/shutter mechanism 122 includes a suction mechanism that suctions the medicine identified as the delivery target, and a shutter mechanism that prevents the medicine suctioned by the suction mechanism from falling. The suction mechanism is provided so that it can move in the Z-axis direction. The shutter mechanism is provided in front of the end so that it can move approximately parallel to the XY plane.

When acquiring a drug, the suction mechanism extends from the end, adsorbs the identified drug at its tip, and then returns to the end. In this state, the transport control unit 61 moves the shutter mechanism to a position opposite the end and maintains the shutter mechanism's position (closed state) during drug transport. The transport control unit 61 moves the suction/shutter mechanism 122 to a position opposite the drug placement stage 133a (see 3002 in Figure 3) of the drug holding mechanism 133 arranged in the receiving area Ar1, and then moves the shutter mechanism to a position not facing the end (open state). Then, after extending the suction mechanism from the end, the suction state is released, placing the drug on the drug placement stage 133a.

Under the control of the transport control unit 61, the transport mechanism 123 moves the suction/shutter mechanism 122 in the X-axis and Y-axis directions. This transport mechanism 123 enables the movement of the suction/shutter mechanism 122 when searching for drugs to be transported in the first storage unit 11, or the transport of drugs from the first storage unit 11 to the drug placement table 133a. During the drug sorting process, drugs can also be transported from the drug placement table 133a to the second storage unit 14, waiting tray 15, or collection tray 16. During the drug sorting process, the sorting control unit 62 controls the transport/sorting unit 12 to transport drugs placed in the receiving area Ar1 to a specified sorting cup 141 in the second storage unit 14 or to the waiting tray 15, based on the discrimination result by the discrimination unit 64.

(Drug Imaging Processing)
Next, the medicine imaging process by the imaging unit 13 will be described with reference to Figures 1, 2002 in Figure 2, Figures 3, and 4. 3001 and 3002 in Figure 3 are perspective views showing the overall configuration of the imaging unit 13, and 3003 in Figure 3 is a perspective view showing an example of the medicine placing table 133a. 4001 and 4002 in Figure 4 are diagrams for explaining the rotation of the imaging unit 13. The medicine imaging process is mainly performed by the imaging unit 13 and the imaging control unit 63.

Specifically, the imaging unit 13 is placed on the medicine placing table 133a and captures an image of the medicine placed in the placement area Ar2 (imaging area) where the medicine to be imaged is placed, as shown at 3002 in Figure 3. The imaging control unit 63 controls the imaging process by the imaging unit 13, the pivoting movement of the first camera 131 and illuminator 134, and the movement of the medicine holding mechanism 133. As shown in Figures 1 and 3, the imaging unit 13 includes the first camera 131 (imaging unit), the rotation mechanism 132 (rotating unit), the medicine holding mechanism 133 (medicine placing table, movement mechanism), and the illuminator 134 (ultraviolet light irradiation unit, visible light irradiation unit).

The first camera 131 captures images of drugs placed in the placement area Ar2 facing the first camera 131 so that the discrimination unit 64, described below, can identify the type of drug. The drug holding mechanism 133 is a mechanism for holding drugs, and as shown at 3001 and 3002 in Figure 3, includes a drug placement table (petri dish) 133a, a swivel mechanism 133b (movement mechanism), and a shaft 133c connecting the drug placement table 133a and the swivel mechanism 133b. The drug placement table 133a is used to place drugs to be imaged. The swivel mechanism 133b moves the drug placement table 133a; specifically, it rotates the drug placement table 133a relative to the XY plane and rotates the shaft 133c in the circumferential direction of the shaft 133c.

When the medicine transported from the first container 11 is placed on the medicine placing table 133a, the imaging control unit 63 drives the swivel mechanism 133b to move the medicine placing table 133a from the receiving area Ar1 to the placement area Ar2. Then, the imaging control unit 63 controls at least the first camera 131 and the illuminator 134 to capture an image of the medicine placed in the placement area Ar2. The captured image is stored in the memory unit 80 as a captured image 82. For example, after the imaging is completed, the imaging control unit 63 drives the swivel mechanism 133b to move the medicine placing table 133a, on which the imaged medicine is placed, from the placement area Ar2 to the receiving area Ar1.

In this embodiment, two medicine placing tables 133a are provided at the tip (end) of the shaft 133c. By rotating the shaft 133c, the pivoting mechanism 133b positions one medicine placing table 133a in the placement area Ar2 and the other medicine placing table 133a in the receiving area Ar1. When imaging medicines in the placement area Ar2, the transport/sorting unit 12 transports medicines from the first storage unit 11 to the medicine placing table 133a in the receiving area Ar1, enabling continuous imaging of medicines. Note that it is assumed that no medicines are placed on the medicine placing table 133a, for example, after medicine sorting into the second storage unit 14.

In addition, in this embodiment, the medicine placing table 133a is transparent. Therefore, the first camera 131 can capture images of the medicines placed on the medicine placing table 133a from multiple directions through the medicine placing table 133a.

Also, as shown in 3003 of FIG. 3, the medicine mounting table 133a may have a generally V-shaped cross section with a recessed bottom. Furthermore, as shown in 3003 of FIG. 3 and FIG. 4, when the medicine mounting table 133a is placed in the receiving area Ar1 and the placement area Ar2, the groove direction of the generally V-shaped cross section (extension direction of the shaft portion 133c) is generally parallel to the rotation axis Ay of the imaging mechanism (described below) of the rotation mechanism 132. Furthermore, the bottom of the medicine mounting table 133a does not have to be an acute V-shape. As shown in 3003 of FIG. 3, the bottom may have a bottom surface portion 133aa and inclined surface portions 133ab inclined from two opposing points of the bottom surface portion 133aa. The shape of the bottom should be such that the information (engraved information or printed information) engraved or printed on the medicine can be recognized when viewed (photographed) from the back of the medicine placing table 133a, and the medicine can be secured in place.

If the medication is a capsule or deformed tablet (e.g., rugby ball-shaped), if the bottom of the medication mounting table 133a is flat, the medication will not be aligned on the XY plane, making it difficult to obtain a clear image of the medication (engraved or printed information). If the cross section is roughly V-shaped, the capsule or deformed tablet fits into the bottom end, fixing the medication in place. This makes it easier to obtain a clear image of the medication. In the case of tablets, for example, the axis 133c can be rotated circumferentially around the axis 133c to position the flat portion (inclined surface 133ab) of the medication mounting table 133a facing the first camera 131, thereby ensuring that the medication does not move.

In addition, the rotation mechanism 133b can also vibrate (slightly move or shake) the medicine placing table 133a. In this case, for example, by vibrating and rolling a capsule placed on the medicine placing table 133a, the printed part of the capsule can be oriented in a predetermined direction (e.g., this part can be made to face the first camera 131, which is positioned in its initial position, as described below). Furthermore, with this vibration, even if a cylindrical tablet (with a circular bottom) is placed upright on the flat surface, the tablet can be turned sideways (positioned so that the bottom of the tablet faces the flat surface).

Under the control of the imaging control unit 63, the illuminator 134 emits light that is irradiated onto the drug when imaging the drug. Figure 5 is a diagram schematically showing the internal configuration of the imaging unit 13. As shown in 3001 in Figure 3 and in Figure 5, the illuminator 134 includes a visible light irradiator (first irradiator 134a and second irradiator 134b) that irradiates the drug with visible light, and an ultraviolet light irradiator 134c that irradiates the drug with ultraviolet light.

The first irradiating unit 134a and the second irradiating unit 134b irradiate the medication with white light as visible light. The first irradiating unit 134a is a bar-shaped visible light source (bar illumination), and the second irradiating unit 134b is a ring-shaped visible light source (ring illumination). The first camera 131 captures an image based on visible light (visible light image) by receiving visible light emitted from the first irradiating unit 134a or the second irradiating unit 134b and reflected by the medication. The imaging control unit 63 stores image data representing the visible light image captured by the first camera 131 in the memory unit 80 as a captured image 82.

The ultraviolet light irradiation unit 134c irradiates the drug with ultraviolet light (e.g., light having a peak wavelength of 365 nm or more and 410 nm or less), thereby exciting the components contained in the drug. This extracts fluorescence (e.g., light having a peak wavelength of 410 nm or more and 800 nm or less) from the drug. The first camera 131 receives the fluorescence emitted from the drug and acquires an image based on the ultraviolet light (ultraviolet image). The imaging control unit 63 stores image data representing the ultraviolet image acquired by the first camera 131 in the memory unit 80 as an acquired image 82.

As shown in Figures 3 and 4, the rotation mechanism 132 rotates the first camera 131 so that it revolves around the placement area Ar2 (the medication placement table 133a placed at that position) where the medication to be imaged is placed. The first camera 131 captures images of the medication placed in the placement area Ar2 from multiple positions to which it is rotated by the rotation mechanism 132. Specifically, the imaging mechanism including the first camera 131 and illuminator 134 rotates so that it revolves around the placement area Ar2. Therefore, the first camera 131 can capture images of the medication from multiple directions while maintaining the positional relationship of the first camera 131 and illuminator 134 with respect to the placement area Ar2.

As shown in FIG. 3 at 3001, the rotation mechanism 132 includes an imaging mechanism drive unit 132a and a power transmission mechanism 132b. The imaging mechanism drive unit 132a generates power to rotate the imaging mechanism around the placement area Ar2. The power transmission mechanism 132b transmits the power generated by the imaging mechanism drive unit 132a to the imaging mechanism. The imaging mechanism drive unit 132a is driven under the control of the imaging control unit 63 to change the position of the imaging mechanism around the placement area Ar2.

The rotation mechanism 132 rotates the imaging mechanism between an initial position and a position opposite the initial position. The initial position is a position approximately perpendicular to the placement area Ar2 and above the placement area Ar2. The position opposite the initial position is a position approximately perpendicular to the placement area Ar2 and below the placement area Ar2. This position can also be said to be the position where the first camera 131 faces the bottom of the medicine placement table 133a located in the placement area Ar2.

As shown in FIG. 4, axis Ax0 is the axis that passes through the center of placement area Ar2 and is parallel to the Z axis, and axis Ax1 is the axis that passes through the center of placement area Ar2 and the center of the imaging mechanism. θ is the angle between axis Ax0 and axis Ax1. In this embodiment, the rotation mechanism 132 positions the imaging mechanism at one of θ = 0° (initial position), 45°, 135°, and 180°. 4001 in FIG. 4 shows the case where the imaging mechanism is at θ = 0°, and 4002 in FIG. 4 shows the case where the imaging mechanism has rotated from the initial position to a position of θ = 45°.

In this way, by rotating the imaging mechanism around the placement area Ar2, it is possible to capture images of the medication from multiple directions while the medication is fixed in the placement area Ar2. Furthermore, even if the medication (tablet) remains upright after shaking the medication placement table 133a, information indicated by markings or other markings on the medication can be obtained by capturing images from an oblique direction (θ = 45° or 135°).

In addition, the imaging mechanism may be fixed and the drug may be rotated to capture images of the drug from multiple directions.

(Imaging position control)
Next, an example of position control of the imaging mechanism will be described. The imaging control unit 63 first sets the imaging mechanism to an initial position and causes the first camera 131 to capture an image of the drug placed in the placement area Ar2 at the initial position. At this time, the first camera 131 acquires visible light images (two visible light images) based on visible light from the first irradiator 134a and the second irradiator 134b, and also acquires an ultraviolet light image based on ultraviolet light from the ultraviolet light irradiator 134c.

Next, the imaging control unit 63 sets the imaging mechanism to a position opposite to the initial position and causes the first camera 131 to capture an image of the medication placed in the placement area Ar2 at that position, acquiring two visible light images and an ultraviolet light image. The discrimination unit 64 analyzes these six images to determine the type of medication. If the medication type cannot be identified as one, the imaging control unit 63 causes the first irradiating unit 134a and the second irradiating unit 134b to emit visible light at positions θ = 45° and 135°, causing the first camera 131 to capture an image of the medication. The discrimination unit 64 analyzes the visible light image at this time to determine the type of medication.

The above is not the only possible method for controlling the position of the imaging mechanism, and various other methods can be used. For example, imaging may be performed from a position opposite the initial position, and then imaging may be performed again from the initial position. Alternatively, drug discrimination processing may be performed based on visible light images captured from a position of θ = 45°, and only if the drug type cannot be identified as a single type may a visible light image be acquired from a position of θ = 135°. Alternatively, only ultraviolet light images may be acquired at the initial position and a position opposite the initial position, drug discrimination processing may be performed based on the ultraviolet light images, and then a visible light image may be acquired at that position. Alternatively, visible light images and ultraviolet light images may be acquired at all positions.

(Image processing/discrimination processing)
Next, the image processing of the image captured by the imaging unit 13 and the drug discrimination processing based on the results of the image processing will be described with reference to Fig. 1. The image processing is mainly performed by the imaging control unit 63, and the discrimination processing is mainly performed by the discrimination unit 64.

The discrimination unit 64 discriminates the type of drug based on the image of the drug captured by the first camera 131. Specifically, the discrimination unit 64 discriminates the type of drug based on the imaging result (visible light image) of the drug captured while being irradiated with visible light from the first irradiation unit 134a or the second irradiation unit 134b. The discrimination unit 64 also discriminates the type of drug based on the imaging result (ultraviolet light image) of the drug captured while being irradiated with ultraviolet light.

The discrimination unit 64 performs image analysis on each of the visible light image and/or ultraviolet light image to extract characteristics of the drug contained in the image. Examples of drug characteristics include size, shape, markings, prints, secant lines, and representative color (the color of the area with the marking or print).

Examples of the size characteristics of the drug include the major and minor axes of the drug. Examples of the shape characteristics of the drug include the roundness and aspect ratio of the drug.

When OCR (Optical Character Recognition) or similar is performed, other information such as the engraved or printed name of the drug (e.g., identification code), identification information indicating the manufacturer (identification information that identifies the drug), and expiration date are extracted as drug characteristics. In the case of ultraviolet light images, the representative color of the drug in the image is also an example of a drug characteristic. The discrimination unit 64 associates the extracted information indicating the characteristics of each drug with the captured image 82 of the drug and stores it in the memory unit 80. Drug characteristics may be extracted using known techniques.

As mentioned above, the extracted drug features include the representative color of the drug in the image. Hereinafter, data indicating this representative color will be referred to as color data (data indicating the drug's color). Color data includes data generated from both visible light images and ultraviolet light images. In the following description, unless otherwise specified, the term "color data" refers to both color data generated from visible light images and color data generated from ultraviolet light images. Color data is data indicating the color of a specific region in an image of the drug (at least a portion of the area where the drug appears, typically an area with an inscription or print). For example, the color data may be the average RGB values of each pixel in that region. Alternatively, the color data may be a CIELab value, which is a value representing the color in that region as a combination of the L*, a*, and b* values in the CIELab color space.

The discrimination unit 64 discriminates the type of drug by comparing the characteristics of each drug with the drug database 81. For example, the discrimination unit 64 ranks candidate drug data (reference drug data) for the imaged drug from the drug database 81 based on color data. The discrimination unit 64 then discriminates the type based on other characteristics in accordance with the above ranking.

The discrimination unit 64 may also use pattern matching or the like to narrow down the drug data candidates related to the imaged drug from within the drug database 81 based on the extracted drug characteristics. In this case, for example, the drug data candidates are narrowed down using at least one of the above-mentioned size, shape, engraving, print, dividing line, and representative color. The discrimination unit 64 then performs OCR or the like to read identification information or the like indicated on the engraving or print, and further narrows down the drug types from the candidates using pattern matching or the like. The discrimination unit 64 may also rank the candidates based on the degree of match between the extracted drug characteristics and the characteristics of drugs (reference drugs) included in the drug database 81.

Furthermore, even if the drug characteristics (target features) extracted using pattern matching or the like are not found in the drug database 81, the discrimination unit 64 will determine the type of drug as a suspected drug if it is estimated to be a drug (tablet or capsule) based on at least a portion of the target features. In this case, the suspected drug can also be sorted into the second storage unit 14 or the waiting tray 15. In this embodiment, the suspected drug may first be temporarily placed on the waiting tray 15.

In this way, the discrimination unit 64 determines whether or not there is drug data corresponding to the captured image 82 captured by the first camera 131 in the drug data (drug database 81) related to multiple types of drugs that have been registered in advance.

The discrimination unit 64 outputs the discrimination result for the type of drug to the sorting control unit 62. For example, if the type of drug can be identified as one, or if the number of candidates has been narrowed down to a predetermined number or less, the discrimination unit 64 outputs the drug data for that drug as the discrimination result. In this case, the discrimination unit 64 stores the drug data for that drug in the memory unit 80, linking it to the captured image 82 of that drug.

When the discrimination unit 64 determines that the type of drug is a suspected drug, it outputs the characteristics of the drug (the characteristics of the object suspected to be a suspected drug) as the discrimination result. On the other hand, when the discrimination unit 64 determines that the drug is registered in the drug database 81 as a drug to be discarded, or when it determines that the object contained in the first container 11 is a foreign object other than a drug, it outputs the discrimination result indicating that the drug is not to be sorted.

As described above, the discrimination unit 64 analyzes the captured image 82 and performs discrimination processing. The discrimination unit 64 may be configured to perform the following processing when it is difficult to discriminate the type of drug in a single discrimination process (when the type of drug cannot be discriminated in a single discrimination process). For example, the discrimination unit 64 may be configured to perform image analysis of the captured image 82 again and attempt to discriminate the type of drug (discrimination retry processing) until a predetermined upper limit (e.g., three times) is reached or a timeout occurs.

(Medicine sorting process)
Next, the medicine sorting process based on the results of the above-mentioned discrimination process will be described with reference to Fig. 1. The medicine sorting process is mainly performed by the transport/sorting unit 12 and the sorting control unit 62.

The transport/sorting unit 12 sorts the medicines by type based on the discrimination results from the discrimination unit 64 and stores them in the second storage unit 14 or the waiting tray 15. The sorting control unit 62 controls the transport/sorting unit 12 to transport the medicines placed in the receiving area Ar1 after imaging and discrimination processing to a predetermined sorting cup 141 in the second storage unit 14 or the waiting tray 15 based on the discrimination results.

When the sorting control unit 62 receives the drug discrimination results, it determines a sorting position for storing the drug, and stores the discrimination results in the memory unit 80 in association with the determined sorting position. Specifically, the sorting control unit 62 determines whether a discrimination result identical to the above discrimination result is stored in the memory unit 80.

If a discrimination result identical to the above discrimination result is stored, the sorting cup 141 linked to the stored discrimination result is determined as the sorting position. Furthermore, if the sorting position linked to the stored discrimination result (e.g., estimated drug) is the waiting tray 15, the waiting tray 15 is determined as the sorting position. On the other hand, if a discrimination result identical to the above discrimination result is not stored, a sorting cup 141 that does not contain any drugs (a sorting cup 141 that has not been determined as a sorting position) is determined as the sorting position. If drugs are stored in all of the sorting cups 141, the waiting tray 15 is determined as the sorting position. Note that for estimated drugs, the sorting control unit 62 may determine one of the sorting cups 141 included in a specified area of the second storage unit 14 as the sorting position, rather than the waiting tray 15.

Once the sorting control unit 62 has determined the sorting position, it controls the transport mechanism 123, similar to the transport control unit 61, to move the transport/sorting unit 12 above the receiving area Ar1. Similarly to the transport control unit 61, the sorting control unit 62 controls the second camera 121 and the suction/shutter mechanism 122 to suction the medicines placed in the receiving area Ar1. The transport mechanism 123 then transports the medicines to the determined sorting cup 141 or waiting tray 15. As described above, the shutter mechanism is closed while the medicines are being transported, preventing the medicines from falling into areas other than the determined sorting position (e.g., sorting cups 141 other than the determined sorting cup 141). After transport, the suction is released, allowing the medicines to be stored in the sorting cup 141 or waiting tray 15. The sorting control unit 62 also counts the number of medicines stored in the sorting cup 141 and stores the count in the memory unit 80, linking it to the sorting position.

After the sorting control unit 62 transports the drugs (drugs after identification) from the drug placement table 133a arranged in the receiving area Ar1 to the sorting position, the transport control unit 61 controls the transport/sorting unit 12 to transport and place the drugs stored in the first storage unit 11 on the now empty drug placement table 133a. This allows the drug sorting device 1 to continuously identify the type of drug.

Furthermore, if the sorting control unit 62 receives a determination result indicating that the type of medicine could not be determined, the object placed in the receiving area Ar1 after the determination is a foreign object, and the foreign object is transported to the collection tray 16.

In this way, the sorting control unit 62 stores all items contained in the first storage unit 11 in either the second storage unit 14, the waiting tray 15, or the collection tray 16, regardless of the results of determining the type of medicine. Therefore, even if the type of medicine cannot be identified as a single type, or if a foreign object has been mixed into the first storage unit 11, the sorting process can be continued without being stopped for that reason.

In order to remove medications for which the discrimination process has been completed from the medication placement table 133a arranged in the receiving area Ar1, the sorting control unit 62 causes the second camera 121 to capture an image of the medication placement table 133a, thereby narrowing down the location of the medications. In addition, when transporting medications stored in the sorting cup 141 to the packaging mechanism 6, the sorting control unit 62 causes the second camera 121 to capture an image of the sorting cup 141, thereby narrowing down the medications to be transported, thereby removing the medications from the sorting cup 141.

In addition, when the upper limit of the number of medications that can be stored in a sorting cup 141 is reached, the sorting control unit 62 stores the medications to be sorted in an empty sorting cup 141 different from the sorting cup 141, even if the type of medication to be sorted is the same as the type of medication sorted in the sorting cup 141.

In addition, data about the medicines stored in the sorting cup 141 by the sorting control unit 62 is stored in an RFID tag attached to the sorting cup 141 by the second RFID reader/writer unit 18.

Like the first RFID reader/writer unit 5, the second RFID reader/writer unit 18 writes various data to RFID tags or reads various data stored in RFID tags. The sorting control unit 62 causes the RFID control unit 68 to write data related to the medication each time a medication is stored in a sorting cup 141. The second RFID reader/writer unit 18 is provided below the second storage unit 14. Specifically, it is provided so as to face the bottom of each sorting cup 141 when reading or writing data related to the medication stored in the RFID tag of each sorting cup 141.

[Embodiment 2]
Other embodiments of the present invention will be described below. For ease of explanation, the same reference numerals will be used to designate components having the same functions as those described in the above embodiment, and the description thereof will not be repeated. The same applies to the following embodiments.

(overview)
In this embodiment, a configuration for quickly and accurately distinguishing between coated tablets, such as sugar-coated tablets, whose surfaces are covered with a film, and other drugs (plain tablets) will be described using Figures 6 and 7.

The discrimination unit 64 in this embodiment extracts feature information indicating the external characteristics of the target drug, including at least reflection area information indicating the reflection area of light irradiated onto the target drug, from the captured image of the drug to be discriminated (target drug), and uses the reflection area information to discriminate the target drug.

For example, the reflection area of light irradiated onto a drug varies depending on whether or not the drug is sugar-coated. With the above configuration, the reflection area of light irradiated onto the drug is used to determine the type of drug in question, making it possible to quickly and accurately distinguish between coated tablets and other drugs.

(Control unit 60a)
The control unit 60a according to this embodiment has the main components described in the first embodiment.

(Discrimination unit 64)
The discrimination unit 64 according to this embodiment includes a feature extraction unit 64a and a matching unit 64b shown in Fig. 1. The discrimination unit 64 performs the following processes in addition to the processes described in the first embodiment.

(Feature extraction unit 64a)
The feature extraction unit 64a extracts feature information (first feature information) indicating the appearance characteristics of the target drug from the captured image of the target drug, the feature information including at least reflection area information indicating the reflection area of the light irradiated on the target drug. The light reflection area can also be referred to as, for example, an area affected by halation in the captured image 82.

In detail, the feature extraction unit 64a extracts features of the drug contained in the image by performing image analysis on the visible light image and/or ultraviolet light image (captured image 82). In addition to the drug features extracted by the discrimination unit 64 described in embodiment 1, the feature extraction unit 64a extracts the reflection area of light irradiated onto the target drug as a feature of the target drug. The feature extraction unit 64a outputs feature information indicating the extracted drug features, including reflection area information, to the matching unit 64b.

For example, the feature extraction unit 64a may extract, as the reflective region, an image region in the captured image 82 that has a saturation value higher than a predetermined value. The predetermined value may be set, for example, to a saturation level that allows the reflective region to be extracted from a captured image of a coated tablet so that coated tablets can be distinguished from other medications through experiments.

The feature extraction unit 64a may also extract the reflection area by performing filtering on the captured image 82, for example, using a bilateral filter.

Figure 6 shows an example of the reflection area of light reflected by the target drug in a captured image. D1 in Figure 6 is a visible light image of a sugar-coated tablet captured using the second irradiator 134b (ring illumination). D2 in Figure 6 is an image obtained by filtering the visible light image shown in D1 using a bilateral filter.

The image shown in D3 of Figure 6 is a visible light image of an uncoated tablet captured using the second irradiation unit 134b. The image shown in D4 of Figure 6 is an image obtained by filtering the visible light image shown in D3 using a bilateral filter.

As shown in D2 in Figure 6, in an image that has been filtered using a bilateral filter, a ring-shaped reflection area is extracted in the image of a sugar-coated tablet. On the other hand, as shown in D4 in Figure 6, in an image that has been filtered using a bilateral filter, no reflection area is extracted in the image of an uncoated tablet.

(Collation unit 64b)
The collation unit 64b compares characteristic information (second characteristic information) indicating the appearance characteristics of the reference drug with characteristic information of the target drug. The characteristic information of the reference drug includes at least reflection area information indicating the reflection area of light irradiated onto the reference drug. The characteristic information is drug data related to the reference drug contained in the drug database 81. The characteristic information of the target drug also includes at least reflection area information indicating the reflection area of light irradiated onto the target drug. The collation unit 64b determines the type of the target drug based on the result of the comparison.

The matching unit 64b may use pattern matching or the like to narrow down candidate drug data for the imaged target drug from drug data for reference drugs contained in the drug database 81, based on the features of the target drug extracted by the feature extraction unit 64a. Furthermore, the matching unit 64b may also narrow down candidate drug data using at least one of the drug features described in embodiment 1 in addition to the reflection region information. Furthermore, the matching unit 64b may rank the candidates based on the degree of agreement between the extracted features of the target drug and the features of the reference drug contained in the drug database 81.

For example, the drug database 81 includes, as drug data for each reference drug, reflection area information indicating the reflection area of light irradiated onto the reference drug in the captured image in which the reference drug is captured. The matching unit 64b matches the shape of the reflection area indicated by the reflection area information for the target drug with the shape of the reflection area indicated by the reflection area information for each reference drug, and narrows down the drug data for the multiple reference drugs to candidate drug data for the target drug (i.e., candidate types of target drug).

The collation unit 64b outputs the results of the determination of the type of target drug to the sorting control unit 62. For example, if the type of target drug can be identified as one, or if the number of candidates has been narrowed down to a predetermined number or less, the collation unit 64b outputs the drug data for the target drug as the determination result. In this case, the collation unit 64b stores the drug data for the target drug in the memory unit 80, linking it to the captured image 82 of the target drug.

(Processing example)
7 is a flowchart showing an example of the discrimination process of the discrimination unit 64 of this embodiment. As shown in FIG. 7, the feature extraction unit 64a extracts feature information of the target drug, including reflection region information, from the captured image 82 (S11). Next, the comparison unit 64b compares the feature information of the target drug with the feature information of the reference drug (drug data) contained in the drug database 81 (S12). Next, the comparison unit 64b determines the type of the target drug based on the comparison result (S13), outputs the drug type discrimination result to the sorting control unit 62, and the process ends.

(Modification)
(Summary of Modified Examples)
In this modification, the discrimination unit 64 extracts feature information indicating the external characteristics of the target drug from the captured image of the target drug. The feature information includes at least reflection information indicating whether light irradiated onto the target drug is reflected by the target drug. The discrimination unit 64 uses the reflection information to discriminate the target drug.

For example, whether or not the light irradiated onto the drug is reflected depends on whether or not the drug is coated (e.g., sugar coated). Therefore, by using whether or not light irradiated onto the drug is reflected to determine the type of drug in question, coated tablets can be quickly and accurately distinguished from other drugs.

Other processing by the discrimination unit 64 in this modified example is the same as that of the discrimination unit 64 described in embodiment 2 above, and therefore will not be described again.

(Feature extraction unit 64a)
The feature extraction unit 64a according to this modification extracts feature information (third feature information) of the target drug from the captured image of the target drug. The feature information includes at least reflection information indicating whether light irradiated onto the target drug is reflected by the target drug. The feature extraction unit 64a outputs the feature information of the target drug including the extracted reflection information to the comparison unit 64b.

For example, as described in embodiment 2, the feature extraction unit 64a extracts image regions in the captured image 82 that have a saturation value higher than a predetermined value. The feature extraction unit 64a determines whether or not light is reflected by the target drug depending on whether or not this extraction occurs. Furthermore, as described in embodiment 2, the feature extraction unit 64a determines whether or not light is reflected by the target drug using a captured image that has been subjected to bilateral filtering of the captured image 82. In response to this determination, the feature extraction unit 64a extracts reflection information indicating whether or not the irradiated light is reflected by the reference drug.

(Collation unit 64b)
The collation unit 64b in this modification collates the characteristic information of the reference drug (fourth characteristic information) with the characteristic information of the target drug (third characteristic information). The characteristic information of the reference drug includes at least reflection information indicating whether the irradiated light is reflected by the reference drug. The characteristic information of the target drug includes at least reflection information indicating whether the irradiated light is reflected by the target drug. The collation unit 64b determines the type of the target drug based on the result of the collation. The collation unit 64b outputs the determination result of the type of the target drug to the sorting control unit 62.

For example, the drug database 81 includes, as drug data for each reference drug, reflection information indicating whether light irradiated in an image of each reference drug is reflected by the reference drug. The matching unit 64b may match the reflection information for the target drug with the reflection information for each reference drug, and narrow down the drug data for the target drug from among the drug data for multiple reference drugs.

For example, when the matching unit 64b extracts reflection information from the captured image showing that light irradiated in the captured image of the target drug is reflected by the target drug, the matching unit 64b extracts characteristic information (drug data) including the reflection information from the drug database 81. This reflection information is information showing that light irradiated in the captured image of the reference drug is reflected. As a result, the matching unit 64b narrows down the characteristic information of the reference drug including reflection information showing that the light is reflected as candidate drug data for the target drug in the drug database 81.

Similarly, when the matching unit 64b extracts reflection information from the captured image indicating that the light irradiated in the captured image of the target drug is not reflected by the target drug, the matching unit 64b extracts feature information including the following reflection information. This reflection information is information indicating that the light irradiated in the captured image of the reference drug is not reflected. As a result, the matching unit 64b narrows down the feature information of the reference drug including reflection information indicating that the light is not reflected as candidates for drug data related to the target drug in the drug database 81.

In addition to the above-mentioned reflection information, the matching unit 64b may also narrow down the drug data candidates using at least one of the drug characteristics described in embodiments 1 and 2.

[Embodiment 3]
(overview)
In this embodiment, a configuration for quickly and accurately identifying medicines by utilizing differences in thickness of the medicines will be described.

(Control unit 60a)
The control unit 60a according to this embodiment has the main components described in the first embodiment.

(Discrimination unit 64)
The discrimination unit 64 according to this embodiment includes a feature extraction unit 64a and a matching unit 64b shown in Fig. 1. The discrimination unit 64 according to this modification performs the following processes in addition to the processes described in the first embodiment.

(Feature extraction unit 64a)
The feature extraction unit 64a according to this embodiment extracts the thickness of the target drug from a captured image of the target drug taken from an oblique direction (θ=45° or 135°). The feature extraction unit 64a outputs thickness information indicating the thickness to the collation unit 64b as one of the features of the target drug. For example, the feature extraction unit 64a extracts the thickness of the drug contained in the image by performing image analysis on the visible light image and/or ultraviolet light image (captured image 82). Note that the process of capturing an image of the target drug from an oblique direction has been described in detail in the "Drug Imaging Process" of the first embodiment, and therefore will not be described again here.

Figure 8 is a diagram showing an example of the thickness of a target drug extracted by the feature extraction unit 64a. In detail, the image shown in D5 of Figure 8 is a visible light image of the drug captured using the first irradiator 134a (bar illumination). For example, the feature extraction unit 64a extracts the thickness T1 of the target drug being captured using known image processing techniques. For example, the feature extraction unit 64a may extract the thickness T1 of the target drug from the captured image using techniques such as trapezoidal correction or panoramic imaging.

(Collation unit 64b)
The collation unit 64b collates the characteristic information of the reference drug with the characteristic information of the target drug. The characteristic information of the reference drug includes at least thickness information indicating the thickness of the reference drug. The characteristic information of the target drug includes at least thickness information indicating the thickness of the target drug.

For example, the drug database 81 includes thickness information indicating the thickness of each reference drug as drug data for each reference drug. The matching unit 64b may match the thickness information of the target drug with the thickness information of each reference drug to narrow down drug data candidates for the target drug from drug data for multiple reference drugs. Furthermore, the matching unit 64b may narrow down drug data candidates using at least one of the drug characteristics described in embodiments 1 and 2 in addition to the thickness information. Furthermore, the matching unit 64b may match the thickness information of the target drug with the thickness information of each reference drug in the "discrimination retry process" described in the "Image processing/discrimination process" of embodiment 1 to determine the type of target drug.

[Embodiment 4]
(overview)
In this embodiment, a configuration that can accurately determine the type of target medicine will be described with reference to FIG.

The drug database update unit 74 according to this embodiment performs the following processing when the discrimination unit 64 fails to discriminate the type of target drug. After receiving a user operation, the drug database update unit 74 updates the target drug color data 84 (target color information) indicating the color of the target drug used in the discrimination to color data (reference color information) indicating the color of the reference drug included in the drug database 81.

With the above configuration, for target drugs whose type has not been identified, the target drug color data 84 extracted from the captured image 82 taken in an environment where the type of the target drug is actually identified can be used as the color data of the reference drug. This allows for accurate identification of the type of the target drug.

The control unit 60a receives the drug database 81 created by the external device from the external device and stores it in the memory unit 80. The drug sorting device 1 used by the external device to create the drug database 81 is different from the drug sorting device 1 installed in a medical institution such as a hospital, and the drug imaging environment is also different. With the above configuration, it is possible to reduce the possibility that the type of target drug will not be properly identified due to color differences in the captured image caused by differences in the drug sorting device 1 or the imaging environment.

In addition, when creating the drug database 81, captured images of multiple drugs of the same or similar type are used to check whether the drug type can be properly identified using the drug data registered in the drug database 81. In the above configuration, if only the color data of the reference drug is updated, this check does not need to be performed. This reduces the time and effort required for this check.

(Control unit 60a)
The control unit 60a according to this embodiment includes the essential components described in the first embodiment, as well as a medicine database update unit 74 shown in FIG.

(Discrimination unit 64)
The discrimination unit 64 according to this embodiment performs the following processing in addition to the processing described in the first embodiment. The discrimination unit 64 compares the characteristic information (fifth characteristic information) of the target drug with characteristic information (sixth characteristic information) including color data (reference color information) indicating the color of a reference drug contained in the drug database 81, and discriminates the type of the target drug based on the comparison result. The characteristic information of the target drug includes target drug color data 84 (target color information) indicating the color of the target drug, extracted from the captured image of the target drug. The target drug color data 84 is the same as the color data described in the first embodiment, and therefore will not be described again here. The discrimination unit 64 also associates the target drug color data 84 with the captured image 82 of the target drug and stores it in the storage unit 80 as target drug color data 84.

(Operation unit 31)
The operation unit 31 of the touch panel 3 accepts user input. In particular, the operation unit 31 of this embodiment accepts an input operation for updating the target drug color data 84 linked to the captured image 82 in which the discrimination unit 64 failed to discriminate the type of the target drug, to the color data of the reference drug designated by the user.

For example, if the discrimination unit 64 fails to discriminate the type of target drug, the user checks the failed captured image 82 on the display unit 32 of the touch panel 3, etc. If the user is able to discriminate the target drug captured in the captured image 82, the user specifies the discriminated type of drug and performs an operation input to update the target drug color data 84 linked to the captured image 82 to the color data of the specified reference drug. That is, the above operation input includes the specification of the reference drug and the captured image 82. The operation unit 31 may also accept an operation input to update only the color data of the reference drug to the target drug color data 84 linked to the specified captured image 82. The operation unit 31 may also accept an operation input to update both the color data and the captured image of the reference drug to the specified captured image 82 and the target drug color data 84 linked to the specified captured image 82.

(Drug database update unit 74)
When the discrimination unit 64 fails to discriminate the type of target drug, the drug database update unit 74 accepts operation from the user and then updates the target drug color data 84 used for the discrimination as color data of a new reference drug.

In detail, the drug database update unit 74 updates the color data of the reference drug specified by the user in the drug database 81 to the target drug color data 84 linked to the captured image 82 specified by the user, in accordance with the user's operational input via the operation input unit 66.

Furthermore, when the drug database update unit 74 receives an operation input to update both the color data of the reference drug and the captured image of the reference drug, it performs the following processing. The drug database update unit 74 updates the color data and captured image of the reference drug specified by the user in the drug database 81 to the specified captured image 82 and the target drug color data 84 linked to the specified captured image 82.

(Processing example)
9 is a flowchart illustrating an example of the update process of the drug database update unit 74 according to this embodiment. As shown in FIG. 9 , the drug database update unit 74 determines whether or not an operation input for updating the color data of the reference drug stored in the drug database 81 has been received (S21). If an operation input for updating the color data of the reference drug has been received (YES in S21), the drug database update unit 74 performs the following process. The drug database update unit 74 updates the color data of the reference drug designated by the user in the drug database 81 to the target drug color data 84 associated with the designated captured image 82 (S22), and the process ends. Note that if an operation input for updating the color data of the reference drug has not been received (NO in S21), the process returns to S21.

[Embodiment 5]
(overview)
In this embodiment, a configuration that allows target drugs to be quickly and accurately identified will be described with reference to FIGS. 10 to 12. FIG.

In this embodiment, the drug database 81 includes specific drug information indicating whether each reference drug is a specific drug as drug data for each reference drug.

The discrimination unit 64 according to this embodiment discriminates the type of target drug using identification information displayed on the surface of the reference drug and the target drug, and position information indicating the relative position of the display area of the identification information with respect to the image of the drug in the captured image of the reference drug and the target drug. However, if the reference drug indicated by the specific drug information to be a specific drug is a candidate for the target drug, the discrimination unit 64 does not use the position information to discriminate the type of target drug, but instead determines whether the target drug is the same type of drug as the reference drug.

For example, in the case of spherical medicines or medicines for which the display position of identification information differs between production lots, even if the medicines are of the same type, the relative position in the reference image and the relative position in the target image may be different by more than a predetermined value. With the above configuration, when such medicines are the target medicine, it is possible to determine whether the target medicine is the same type as the reference medicine, regardless of whether the two relative positions are different by more than a predetermined value. On the other hand, for medicines other than the predetermined type of target medicine, if the two relative positions are different by more than a predetermined value, the type of the target medicine is not determined to be the same type as the reference medicine. For example, in the case of two different medicines that share the same character string as part of their identification information, the target medicine is not determined to be the same type as the reference medicine. Therefore, with the above configuration, the possibility of misidentifying medicines can be reduced.

(Control unit 60a)
The control unit 60a according to this embodiment has the main components described in the first embodiment.

(Discrimination unit 64)
The discrimination unit 64 according to this embodiment includes a feature extraction unit 64a and a matching unit 64b shown in Fig. 1. The discrimination unit 64 performs the following processes in addition to the processes described in the first embodiment.

(Feature extraction unit 64a)
The feature extraction unit 64a extracts feature information (seventh feature information) of the target drug from the captured image 82. The feature information includes at least (1) identification information displayed on the surface of the target drug, and (2) position information (first position information) indicating the relative position of the display area of the identification information with respect to the image of the target drug.

In detail, in addition to the features of the target drug extracted by the discrimination unit 64 described in embodiment 1, the feature extraction unit 64a extracts the relative position of the display area of the identification information with respect to the image of the target drug as a feature. The feature extraction unit 64a outputs feature information including the first position information to the matching unit 64b.

For example, as described in embodiment 1, the identification information may be an inscription or print affixed to the surface of the medication to identify the medication. Furthermore, the "relative position of the display area of the identification information relative to the image of the target medication" may be, for example, a predetermined position (e.g., the center of the display area) in the display area of the inscription or print on the surface of the medication relative to the image of the target medication.

(Collation unit 64b)
The collation unit 64b collates the characteristic information of the reference drug (eighth characteristic information) with the characteristic information of the target drug (seventh characteristic information). The characteristic information of the target drug includes at least identification information and first position information. Furthermore, the characteristic information of the reference drug (drug data related to the reference drug) includes at least specific drug information indicating whether the reference drug is a specific drug, identification information displayed on the surface of the reference drug, and position information (second position information) indicating the relative position of the display area of the identification information relative to the image of the reference drug. The "relative position of the display area of the identification information relative to the image of the reference drug" can be, for example, a position corresponding to a predetermined position on the target drug (e.g., the center of the display area). The relative positions on the target drug and the relative positions on the target drug can be any positions as long as they correspond to each other. Furthermore, examples of the specific drug indicated by the specific drug information include a drug with a shape that makes the reference drug roll easily (e.g., spherical), a drug whose display position of the identification information varies between production lots, etc.

The matching unit 64b determines the type of target drug based on the results of the matching. The matching unit 64b may use pattern matching or the like to narrow down candidate drug data for the imaged target drug from the drug data for reference drugs included in the drug database 81, based on the features of the target drug extracted by the feature extraction unit 64a. Furthermore, the matching unit 64b may also narrow down candidate drug data using at least one of the drug features described in embodiment 1 in addition to the above identification information.

When the specific drug information indicates that the reference drug is a specific drug, the matching unit 64b determines whether the target drug is the same type of drug as the reference drug without using the first location information and the second location information to determine the type of target drug.

In addition, when the specific drug information indicates that the reference drug is not a specific drug, the matching unit 64b uses the first location information and the second location information to determine the type of target drug and determines whether the target drug is the same type of drug as the reference drug.

In detail, the collation unit 64b determines whether the position indicated by the first position information and the position indicated by the second position information are separated by a predetermined value or more. In other words, the collation unit 64b determines whether the relative position of the target drug and the relative position of the reference drug are separated by a predetermined value or more. Depending on the result of this determination, the collation unit 64b outputs the result of drug type determination to the sorting control unit 62.

(Processing example)
FIG. 10 is a flowchart showing an example of the discrimination process of the discrimination unit 64 of this embodiment. As shown in FIG. 10 , the feature extraction unit 64a extracts feature information of the target drug from the captured image 82 (S1). The feature information includes, for example, shape, color data, identification information, and first position information. Next, the matching unit 64b matches part of the feature information of the target drug (shape, color data, etc.) with the feature information of the reference drug included in the drug database 81 to narrow down the drug data to be candidates for the target drug (S2). Next, the matching unit 64b begins matching the identification information of the reference drug data for each of the narrowed-down candidate target drugs with the identification information of the target drug (S3). In the process of matching the identification information, the matching unit 64b begins matching the entire identification information (S4). Specifically, the matching unit 64b determines whether the shape of the display area of the identification information of the candidate reference drug data matches the shape of the display area of the identification information of the target drug (S5).

If it is determined that the shape of the display area of the identification information of the candidate reference drug data matches the shape of the display area of the identification information of the target drug (YES in S5), the discrimination unit 64 begins pattern matching for each character of the identification information (S6). The feature extraction unit 64a performs OCR on the identification information of the target drug (S7). Note that the feature extraction unit 64a may perform OCR on the identification information of the target drug in step S1. Next, the matching unit 64b determines whether all characters included in the identification information of the target drug match the characters included in the identification information of the reference drug stored in the drug database 81 (S8). If it is determined that all characters included in the identification information of the target drug match the characters included in the identification information of the reference drug (YES in S8), the matching unit 64b performs the following processing. The matching unit 64b determines whether the specific drug information of the candidate reference drug data indicates that the drug is a specific drug (S31). Note that in this example, the specific drug is a drug with a shape that is easy to roll. If the specific drug information determines that the reference drug data for the candidate drug does not indicate a drug with a shape that is easy to roll (NO in S31), the matching unit 64b matches the first position information extracted by the feature extraction unit 64a with the second position information of the reference drug indicated by the drug data for the candidate drug contained in the drug database 81 (S32). Based on this matching, the matching unit 64b determines whether the position indicated by the first position information and the position indicated by the second position information are separated by a predetermined value or more (S33).

Figure 11 is a diagram showing an example of the relative position of a target drug or the relative position of a reference drug. Positions P1, P2, P3, and P4 are shown on the images of the target drug or reference drug shown at D6, D7, D8, and D9 in Figure 11. Positions P1, P2, P3, and P4 are the centers (predetermined positions) of the display area of the identification information (ABC or ABC 12), respectively. For example, the relative positions of the target drug or the relative positions of the reference drug are positions P1, P2, P3, and P4 with respect to the images of the target drug shown at D6, D7, D8, and D9. The feature extraction unit 64a extracts the display area of the identification information using, for example, known image processing techniques, and identifies the center of the display area.

Figure 12 shows an example of a comparison between the relative position of a target drug and the relative position of a reference drug.

D10 in Figure 12 shows an example of an image of a target drug. P5 shown in D10 in Figure 12 indicates the relative position of the identification information display area with respect to the image D10 of the target drug (the center of the identification information display area). D11 in Figure 12 also shows an example of an image of a reference drug. P6 shown in D11 in Figure 12 indicates the relative position of the identification information display area with respect to the image D11 of the reference drug (the center of the identification information display area). D12 in Figure 12 is a diagram in which the image D10 of the target drug and the image D11 of the reference drug are superimposed. Distance L1 shown in Figure 12 indicates the distance between the center P5 of the identification information display area in the image of the target drug and the center P6 of the identification information display area in the image of the reference drug.

The collation unit 64b determines whether the value of distance L1 is equal to or greater than a predetermined value. The predetermined value can be set as appropriate, for example, to 30 pixels. The predetermined value may be set to a level that allows drugs other than the specified drug to be identified as different drugs through experiments, for example.

If the collation unit 64b determines that the position indicated by the first position information and the position indicated by the second position information are not separated by more than a predetermined value (NO in S33), the collation unit 64b outputs the determined type of medicine to the sorting control unit 62 (S34), and the process ends.

If it is determined that the shape of the display area of the identification information of the candidate reference drug data does not match the shape of the display area of the identification information of the target drug (NO in S5), the discrimination unit 64 performs the following process. The discrimination unit 64 determines whether there is any unmatched candidate reference drug data (S35). If there is unmatched candidate reference drug data (YES in S35), the process returns to S3. If there is no unmatched candidate reference drug data (NO in S35), the process ends. If it is determined that all characters included in the identification information of the target drug do not match the characters included in the identification information of the reference drug (NO in S8), the process continues to S35. If the comparison unit 64b determines that the position indicated by the first position information and the position indicated by the second position information are separated by a predetermined value or more (YES in S33), the process continues to S35. If it is determined that the specific drug information indicates that the candidate reference drug data is a drug with a shape that is easy to roll (YES in S31), the process continues to S34.

Note that if the drug database 81 does not contain any drug data that could be a candidate for the target drug, i.e., if there is no reference drug data for a candidate that has not yet been matched (NO in S35), the following processing may be performed. If the matching unit 64b estimates that the target drug is a drug (tablet or capsule) based on at least some of the characteristics of the target drug, it determines the type of target drug as a suspected drug and ends the processing. Also, if the matching unit 64b cannot determine that the imaged object in the captured image 82 is a drug, it determines that the imaged object is a foreign object other than a drug and ends the processing.

[Embodiment 6]
(overview)
In this embodiment, a configuration that can improve the efficiency of the sorting process will be described with reference to FIG.

The priority determination unit 65 according to this embodiment assigns priorities to drug types based on the past sorting performance of the drug sorting device 1. For example, the priority determination unit 65 may determine a priority for each drug type based on the frequency of discrimination in the discrimination history of the discrimination unit 64. Furthermore, the sorting control unit 62 according to this embodiment determines that drug types with a priority lower than a predetermined value should be temporarily placed in the waiting tray (temporary storage unit) 15.

Generally, while the medicine sorting device 1 sorts hundreds to thousands of medicines, there is a limit to the number of sorting cups 141 that can be placed in the second storage unit 14 (in this embodiment, 40 sorting cups 141 can be placed). Therefore, when medicines are sorted in sorting order, there is a possibility that medicines with a low discrimination frequency (medicines that are held in small numbers in the sorting cups 141) will be sorted into the sorting cups 141, and medicines with a high discrimination frequency (medicines that are held in large numbers in the sorting cups 141) will be sorted into the waiting tray 15.

With the above configuration, medications with a low frequency in the classification history are temporarily placed on the waiting tray 15. Therefore, medications with a high frequency in the classification history can be prioritized and sorted into the sorting cups 141 of the second storage unit 14. In other words, more medications can be sorted (targets for sorting into the sorting cups 141).

The priority determination unit 65 may also determine priorities for each type of medication based on prescription frequency.

The priority determination unit 65 also uses a genetic algorithm to identify a combination of drug types to be sorted into the sorting cups 141 that will maximize the rate at which the sorting work can be completed by sorting into the sorting cups 141 in a single sorting operation. In identifying these combinations, the priority determination unit 65 may use past sorting results. The priority determination unit 65 calculates an index for each drug type from the identified combination of drug types. The priority determination unit 65 assigns a priority to each drug type based on the calculated index.

The process according to this embodiment may also be configured to be performed when the amount of medication contained in the first storage unit 11 is greater than a predetermined amount (for example, when the bottom of the first storage unit 11 is not visible due to the amount of medication contained therein).

(Control unit 60a)
The control unit 60a according to this embodiment includes a priority determination unit 65 shown in FIG. 1 in addition to the main components described in the first embodiment.

(Priority determination unit 65)
The priority determination unit 65 prioritizes drug types based on the past sorting performance of the drug sorting device 1. The priority determination unit 65 stores priority data 83 indicating the determined priority of each drug type in the storage unit 80. Specifically, the priority determination unit 65 determines a priority for each drug type (reference drug) based on the frequency of discrimination in the discrimination history of the discrimination unit 64. More specifically, when the discrimination unit 64 has identified a single drug type or narrowed down the number of candidates to a predetermined number or less, the discrimination unit 64 associates drug data (discriminated type) related to the target drug with a captured image 82 of the target drug and stores the data in the storage unit 80. For example, the priority determination unit 65 may use the drug data (discriminated type) related to the target drug as the discrimination history. Furthermore, the priority determination unit 65 may set a priority for each drug type such that the drug type that is most frequently discriminated in the discrimination history has a higher priority.

The priority determination unit 65 may update the priority data 83 each time the discrimination unit 64 discriminates the target drug. Furthermore, the priority determination unit 65 may update the priority data 83 when the discrimination unit 64 has discriminated a predetermined number of times since the previous update of the priority data 83. Furthermore, the priority determination unit 65 may update the priority data 83 at predetermined intervals.

(Sorting control unit 62)
The sorting control unit 62 according to this embodiment performs the following processes in addition to the processes described in the first embodiment.

If the priority corresponding to the type of target drug identified by the discrimination unit 64 is lower than a predetermined value, the sorting control unit 62 determines to temporarily place the target drug on the waiting tray 15. In detail, the sorting control unit 62 acquires the priority corresponding to the type of target drug identified by the discrimination unit 64 from the priority data 83 in the memory unit 80. If the acquired priority is lower than a predetermined value, the sorting control unit 62 determines to temporarily place the target drug identified by the discrimination unit 64 on the waiting tray 15. The predetermined value may be set to a value that allows the user to sort the number of types of drugs desired when using the drug sorting device 1, for example.

(Processing example)
FIG. 13 is a flowchart showing an example of a medicine sorting process of the sorting control unit 62 of this embodiment.

As shown in FIG. 13, when the sorting control unit 62 receives the identified type of target drug, it determines whether the priority corresponding to that type of target drug is lower than a predetermined value (S41). If the priority is lower than the predetermined value (YES in S41), it decides to temporarily place the target drug on the waiting tray 15 (S42), and the process ends.

If the priority is equal to or greater than a predetermined value (NO in S41), the sorting control unit 62 determines whether a discrimination result identical to the above discrimination result is stored in the memory unit 80 (S43). That is, it determines whether the type of the target drug has already been discriminated. If a discrimination result identical to the above discrimination result is stored (YES in S43), the sorting cup 141 linked to the stored discrimination result is determined as the sorting position (S44), and the process ends.

Also, if the same determination result is not stored (NO in S43), a sorting cup 141 that does not contain any medication (a sorting cup 141 whose sorting position has not been determined) is determined as the sorting position (S45). Next, the sorting control unit 62 associates the determined type of medication with the determined sorting position and stores them in the memory unit 80 (S46), and the process ends.

[Embodiment 7]
(overview)
In this embodiment, a configuration that can improve the sorting work will be described with reference to FIG.

If the identification information shown on a drug is changed by a pharmaceutical manufacturer or other entity, drugs with the previous identification information and drugs with the new identification information will be mixed in the pharmacy. In particular, if the identification information of a drug has been changed, but the drug identified before and after the change is managed using the same YJ code, etc., the drugs with the identification information before and after the change may be managed in a mixed state.

In this embodiment, the print output control unit 69 performs the following processing when a drug whose identification information has changed has been identified multiple times by the discrimination unit 64. The print output control unit 69 controls the print output unit 4 to print an alert stating "Identification information has changed" on the medicine packaging paper used to package the drug, or in a journal that displays drug data related to the drug.

The above configuration makes it possible for users to be aware of the possibility that medications before and after the identification information change may be mixed and sorted. For example, when filling or reusing medications in a packaging machine, users can be warned not to mix medications before and after the identification information change.

(Control unit 60a)
The control unit 60a according to this embodiment has the main components described in the first embodiment.

(Discrimination unit 64)
The discrimination unit 64 according to this embodiment performs the following process in addition to the processes described in the first embodiment. The discrimination unit 64 stores the captured image 82 used to discriminate the target drug and captured image discrimination information 86 in the storage unit 80. In detail, the discrimination unit 64 discriminates the type of the target drug from the captured image 82, and stores a code (e.g., a YJ code) that specifies the discriminated type of the target drug in the storage unit 80 as the captured image discrimination information 86. The captured image discrimination information 86 and the captured image 82 used to discriminate the target drug are associated with each other and stored in the storage unit 80.

The memory unit 80 according to this embodiment also stores identification information change information 85. The identification information change information 85 indicates a code (e.g., a YJ code) that identifies the type of drug whose identification information has been changed. The identification information change information 85 may also indicate the date and time when the identification information change occurred.

(Print output control unit 69)
The printout control unit 69 references the identification information change information 85 and acquires a code that identifies the type of drug whose identification information has changed. The printout control unit 69 compares the captured image discrimination information 86 with the identification information change information 85 to determine whether there are multiple captured images 82 linked to the drug whose identification information has changed. Note that, when a change in the identification information of a certain drug occurs, the printout control unit 69 may be configured to perform the above processing only for a predetermined period after the change (e.g., 30 days after the change).

If it is determined that there are multiple captured images 82 associated with a drug whose identification information has changed, the printout control unit 69 performs the following process. The printout control unit 69 controls the printout unit 4 to print an alert stating "Identification information has changed" on the medicine packaging paper or journal. In the above configuration, the packaging mechanism 6 is equipped with a printing mechanism (not shown) that prints drug data for the drugs packaged in the medicine packaging on the medicine packaging, and the printing of the printing mechanism may be controlled by the printout control unit 69.

In addition, for drugs whose identification information has changed, if the discrimination unit 64 discriminates only drugs that show either the identification information before the change or the identification information after the change within a specified period, the print output control unit 69 may discontinue control to print the alert.

(Processing example)
FIG. 14 is a flowchart showing an example of a print control process of the print output control unit 69 according to this embodiment.

As shown in FIG. 14, the printout control unit 69 references the identification information change information 85 and determines whether there are any drugs whose identification information has changed (S51). For example, the printout control unit 69 may determine whether there are any drugs whose identification information has changed within a predetermined period of time.

If it is determined that a drug with a changed identification information exists (YES in S51), the printout control unit 69 determines whether there are multiple captured images 82 linked to the code that identifies the drug with a changed identification information (S52). If it is determined that multiple captured images 82 exist (YES in S52), the printout control unit 69 controls the printout unit 4 to print an alert stating "Identification information changed" (S53), and the process ends.

If the printout control unit 69 determines that there are no drugs whose identification information has changed (YES in S51), the process ends. Also, if the printout control unit 69 determines that there are no multiple captured images 82 (NO in S52), the process ends.

[Embodiment 8]
(overview)
In this embodiment, a configuration that can improve the sorting work will be described.

The medicine sorting device 1 according to this embodiment accepts operational input for setting unmanned times, such as overnight, and restricts processes that require a heat source, emit light, output alarm sounds, etc.

For example, if a medicine sorting device 1 is installed in an environment where a security system is activated during unmanned hours, such as at night, and is operated at night, the security system may detect light, alarm sounds, heat, etc. emitted by the medicine sorting device 1. In this case, the medicine sorting device 1 may stop processing. Alternatively, the medicine sorting device 1 may not be usable at night. The above configuration makes it possible to operate the medicine sorting device 1 during unmanned hours, such as at night, when the security system is activated.

(Operation unit 31)
The operation unit 31 of the touch panel 3 accepts user input. In particular, the operation unit 31 according to this embodiment accepts an input operation for setting the time period specified by the user.

(Notification section 8)
The medicine sorting device 1 according to this embodiment includes a notification unit 8 shown in Fig. 1. The notification unit 8 includes, for example, a PATLITE (registered trademark) that turns on a light, a speaker that outputs sound, etc. The notification unit 8 notifies the completion or interruption of each process of the medicine sorting device 1 by turning on the PATLITE (registered trademark), outputting sound, etc.

(Saving mechanism 6)
In addition to the configuration described in the first embodiment, the packaging mechanism 6 according to this embodiment is equipped with a heater roller (not shown) for sealing the packaging paper.

(Control unit 60a)
The control unit 60a according to this embodiment includes a notification control unit 75 shown in FIG. 1 in addition to the main components described in the first embodiment.

(Notification control unit 75)
The notification control unit 75 receives instructions to notify the completion or interruption of processing from each component included in the control unit 60a. The notification control unit 75 controls the notification unit 8 in accordance with the received instructions.

In this embodiment, the notification control unit 75 particularly accepts the setting input of the above-mentioned time period specified by the user via the operation input unit 66. The notification control unit 75 controls the notification unit 8 so that the notification unit 8 does not issue a notification during the set time period.

(Packaging control unit 71)
The packaging control unit 71 controls the operation of the packaging mechanism 6. In this embodiment, in particular, the packaging control unit 71 accepts setting input of the time period specified by the user via the operation input unit 66. The packaging control unit 71 controls the packaging mechanism 6 so that the heater roller of the packaging mechanism 6 does not start during the set time period.

(Processing example)
Next, an example of the function restriction process of the medicine sorting device 1 according to this embodiment will be described.

The notification control unit 75 and packaging control unit 71 determine whether or not they have accepted an input operation to set the above-mentioned time period. If they have accepted an input operation to set the above-mentioned time period, the notification control unit 75 and packaging control unit 71 perform the following processing. The notification control unit 75 controls the notification unit 8 so that the notification unit 8 does not issue a notification during the set time period. The packaging control unit 71 also controls the packaging mechanism 6 so that the heater roller of the packaging mechanism 6 is not started during the set time period. The processing then ends. Note that if they have not accepted an input operation to set the above-mentioned time period, the processing returns to the processing where the packaging control unit 71 determines whether or not they have accepted an input operation to set the above-mentioned time period.

[Embodiment 9]
(overview)
In this embodiment, a configuration that can accurately discriminate target drugs will be described with reference to FIG.

The illumination failure determination unit 76 according to this embodiment determines whether the illuminator 134 is malfunctioning when the medicine sorting device 1 is started up or after the determination process has been performed a predetermined number of times. The above configuration makes it possible to detect a malfunction in the illuminator 134, thereby reducing the possibility that the determination process will be performed using an image 82 captured when the illuminator 134 is malfunctioning.

(Control unit 60a)
The control unit 60a according to this embodiment includes a lighting failure determination unit 76 in addition to the main components described in the first embodiment.

(Lighting failure determination unit 76)
As shown in FIG. 1, the lighting failure determination unit 76 includes a luminance calculation unit 76a and a luminance comparison unit 76b.

(Brightness calculation unit 76a)
The luminance calculation unit 76a instructs the imaging control unit 63 to capture an image, calculates the average luminance value in a predetermined area (an imaging area such as the frame of the medicine placing table 133a) in the acquired captured image, and outputs the calculated average luminance value to the luminance comparison unit 76b.

For example, the brightness calculation unit 76a may calculate the average value when the number of times the discrimination unit 64 has performed the discrimination process to discriminate the type of target drug reaches a predetermined number since the previous failure determination.

The storage unit 80 may store discrimination count information 88. The discrimination count information 88 indicates the cumulative number of discrimination processes performed by the discrimination unit 64. The discrimination unit 64 updates the cumulative number indicated by the discrimination count information 88 each time the discrimination process for one captured image 82 is completed. The brightness calculation unit 76a references the discrimination count information 88 and determines whether the number of discrimination processes performed by the discrimination unit 64 has reached a predetermined number since the previous failure determination. For example, the brightness calculation unit 76a determines whether the number of discrimination processes performed by the discrimination unit 64 has reached 100 since the previous failure determination. The predetermined number may be set, for example, through experiments, to a number that is lower than the number of times at which a failure is likely to occur.

In addition, the brightness calculation unit 76a may instruct the imaging control unit 63 to capture an image when the medicine sorting device 1 is started, and calculate the average brightness of the captured image.

(Brightness comparison unit 76b)
The brightness comparison unit 76b determines whether the average brightness of the captured image calculated by the brightness calculation unit 76a is within a predetermined range with respect to a reference value. For example, the brightness comparison unit 76b determines whether the average brightness of the captured image calculated by the brightness calculation unit 76a is greater than or equal to 80% and less than or equal to 120% of the reference value. The predetermined range may be set appropriately, for example, through experiments or the like.

The reference value is, for example, the average brightness of a specific area (such as the imaged area of the medicine placement table 133a) of an image captured when the illuminator 134 is not malfunctioning. The reference value may be obtained when the medicine sorting device 1 is started up (e.g., during teaching).

The brightness may be, for example, the V value of an HSV (H: hue, S: saturation, V: brightness) value that can be converted from the RGB values of each pixel included in the specified area in the captured image.

Furthermore, the captured image used to determine whether the ultraviolet light irradiation unit 134c has a malfunction may be, for example, an image captured using the maximum allowable exposure time for the ultraviolet light irradiation unit 134c. For example, if the exposure time during normal imaging is 8 to 10 msec, the maximum time may be set to, for example, 1 sec. The maximum time may be set as appropriate, for example, through experiments, etc.

The brightness comparison unit 76b determines whether the illuminator 134 is malfunctioning based on the above determination result. The display control unit 67 displays the malfunction determination result on the display unit 32.

(Processing example)
15 is a flowchart showing an example of a malfunction determination process of the medicine sorting device 1 according to the present embodiment. As shown in FIG. 15, the brightness calculation unit 76a determines whether the number of determination processes performed by the determination unit 64 has reached a predetermined number since the previous malfunction determination (S71).

If the brightness calculation unit 76a determines that the number of discrimination processes has reached a predetermined number since the previous failure determination (YES in S71), the imaging control unit 63 controls the first camera 131 to capture an image (S72). Next, the brightness calculation unit 76a calculates the average brightness of the captured image captured in S72 (S73). Next, the brightness comparison unit 76b determines whether the average brightness of the captured image calculated by the brightness calculation unit 76a is within a predetermined range with respect to a reference value (S74). If the brightness comparison unit 76b determines that the average value is not within the predetermined range (NO in S74), the brightness comparison unit 76b determines that the illuminator 134 is malfunctioning (S75). Next, the display unit 32 displays the determination result (S76), and the process ends.

If the brightness calculation unit 76a determines that the number of determination processes has not reached the predetermined number since the previous failure determination (NO in S71), the process returns to S71. Also, if the brightness comparison unit 76b determines that the average value is within the predetermined range (YES in S74), the brightness comparison unit 76b determines that the illuminator 134 is normal (S77), and the process continues to S76.

[Embodiment 10]
(overview)
In this embodiment, a configuration for reducing the data volume of an image for creating an image used to distinguish the type of medicine will be described mainly with reference to FIGS.

The image processing unit 73 provided in the medicine sorting device 1 according to this embodiment generates a cut-out image by cutting out either a first region, which is the region of the image of the target medicine, or a second region, which is the region of the image of the medicine placing table 133a on which the target medicine including the first region is placed, from a captured image including the image of the target medicine, in order to generate an image of a reference medicine for determining the type of target medicine.

For example, with the above configuration, it is possible to reduce the amount of image data required to generate an image of the reference drug. When the medicine sorting device 1 transmits a captured image to an external device, the medicine sorting device 1 transmits a cropped image to the external device as the captured image, thereby reducing the amount of data transmitted compared to a configuration in which cropped images are not generated. This makes it easier to transfer data from the medicine sorting device 1 to an external device.

In addition, the transfer time of captured image data can be reduced. This makes it possible to transfer captured image data using a VPN line, etc. Furthermore, even when transferring the data to a HDD, the cost of the HDD can be reduced by reducing the data volume. This reduces data transfer costs. Furthermore, the running costs of the medicine sorting device 1 can be reduced.

(Image Generation System 1000)
An image generation system 1000 according to this embodiment will be described with reference to Fig. 16. Fig. 16 is a diagram showing an example of the image generation system 1000.

As shown in FIG. 16 , the image generation system 1000 includes a drug sorting device 1 and an information processing device 200 that can be communicatively connected to the drug sorting device 1. The information processing device 200 generates an image of a reference drug for determining the type of target drug. More specifically, the information processing device 200 generates a reference image based on a cut-out image generated by the drug sorting device 1. The reference image is an image for determining the type of target drug to be determined, and is an image (master image) that is pre-registered in the drug database 81. The drug sorting device 1 and the information processing device 200 send and receive data via an internet line or a dedicated VPN (Virtual Private Network) line. Note that the image generation system 1000 may include multiple drug sorting devices 1, and each of the drug sorting devices 1 may be communicatively connected to the information processing device 200.

Furthermore, data transmission and reception between the medicine sorting device 1 and the information processing device 200 does not have to be performed using the above-mentioned line. For example, data may be transmitted and received with the information processing device 200 using a communication device (e.g., a PC (Personal Computer)) different from the medicine sorting device 1. Furthermore, data to be transmitted to the information processing device 200 may be stored in a storage medium (e.g., an HDD (Hard Disk Drive) or a DVD (Digital Versatile Disc)), and the data may be input to the information processing device 200. When a communication device or storage medium is used, the medicine sorting device 1 does not have to be equipped with the transceiver unit 100. Similarly, the information processing device 200 does not have to be equipped with the transceiver unit 201.

(Medicine sorting device 1)
16 , the medicine sorting device 1 includes an imaging unit 13, a touch panel 3, and a computer 60. The computer 60 also includes a control unit 60b, a storage unit 80b, and a transmitting/receiving unit 100 that transmits and receives data to and from the information processing device 200.

Note that Figure 16 shows only the components of the medicine sorting device 1 that are related to the extraction process and transmission process to the information processing device 200 according to this embodiment. The medicine sorting device 1 according to this embodiment may include all of the components provided in the medicine sorting device 1 shown in Figure 1.

(Operation unit 31)
The operation unit 31 of the touch panel 3 accepts input operations by the user. In particular, the operation unit 31 according to this embodiment accepts input operations by the user for transmitting the cropped image 821, the region information file 822, and the compressed image 823 to the information processing device 200.

(Control unit 60b)
The control unit 60b according to this embodiment includes a transmission control unit 77b in addition to the main components of the control unit 60a described in the first embodiment.

(Imaging control unit 63)
The imaging control unit 63 causes the first camera 131 to capture an image of the medicine placed on the medicine placing table 133 a. The imaging control unit 63 acquires the captured image from the first camera 131 and transmits the image to the image processing unit 73.

(Operation input unit 66)
The operation input unit 66 accepts input by the user via the operation unit 31. In this embodiment, the operation input unit 66 particularly accepts an operation for transmitting the cropped image 821, the region information file 822, and the compressed image 823 to the information processing device 200. When the operation is accepted, the operation input unit 66 transmits a signal indicating that the operation has been accepted to the transmission control unit 77b.

(Image processing unit 73)
The image processing unit 73 receives the captured image from the imaging control unit 63. The image processing unit 73 generates a cutout image 821, a region information file 822, and a compressed image 823 from the received captured image. The cutout image 821 may be a cutout image obtained by cutting out a region of the image of the target drug (first region) from a captured image including the image of the target drug. The cutout image 821 may also be a region of the image of the drug placement table 133a on which the target drug is placed (second region), including the image of the target drug. The region information file 822 is region information indicating the cutout region in the captured image (the position and size of the drug image in the captured image). The compressed image 823 is an image obtained by compressing the captured image. The image processing unit 73 stores the generated cutout image, region information file 822, and compressed image 823 in the memory unit 80b.

(Transmission control unit 77b)
The transmission control unit 77b transmits the cut-out image 821, the area information file 822, and the compressed image 823 stored in the storage unit 80b to the information processing device 200 via the transmission/reception unit 100 in accordance with the received user operation input.

(Storage unit 80b)
The storage unit 80b according to this embodiment stores a cut-out image 821, a region information file 822, and a compressed image 823 in addition to the data stored in the storage unit 80 described in the first embodiment.

(Information processing device 200)
The information processing device 200 can be communicatively connected to the medicine sorting device 1 and generates an image of the reference medicine. As shown in Fig. 16, the information processing device 200 includes a transmitting/receiving unit 201, a control unit 202, a storage unit 203, and a touch panel 204. The touch panel 204 includes an operation unit 241 and a display unit 242.

(Transmitter/Receiver 201)
The transmitting/receiving unit 201 transmits and receives data to and from the medicine sorting device 1. In this embodiment, in particular, the transmitting/receiving unit 201 receives a cut-out image 821, an area information file 822, and a compressed image 823 from the transmitting/receiving unit 100 of the medicine sorting device 1.

(Control unit 202)
The control unit 202 comprehensively controls the information processing device 200. The control unit 202 mainly includes an acquisition unit (receiving unit) 221 and an image generation unit 222. The acquisition unit 221 acquires the cropped image 821, the region information file 822, and the compressed image 823 from the transmission/reception unit 201 and stores them in the storage unit 203. The storage unit 203 stores the cropped image 821, the region information file 822, and the compressed image 823. The image generation unit 222 generates a reference image by superimposing the cropped image 821 on the compressed image 823 based on the region information file 822. For example, the image generation unit 222 may resize the reduced compressed image 823 to its original size, and then superimpose the cropped image 821 at the position of the drug image in the captured image indicated by the region information file 822 to generate a reference image. For example, the image generating unit 222 may generate a reference image in response to an operation input by a user via the operation unit 241 of the touch panel 204 .

With the above configuration, the drug sorting device 1 transfers the compressed image 823, the cropped image 821, and the region information file 822 to the external device, the information processing device 200, thereby reducing the amount of data transferred from the drug sorting device 1 to the information processing device 200. Furthermore, even if the amount of data transferred from the drug sorting device 1 to the information processing device 200 is reduced, the information processing device 200 can still generate a reference image in which the resolution of the drug image is at least equivalent to the resolution of the captured image.

(Processing example)
Fig. 17 is a flowchart showing an example of the cutout process of the image processing unit 73 of this embodiment. As shown in Fig. 17, the imaging control unit 63 acquires a captured image from the first camera 131 (S101).

Next, the image processing unit 73 identifies a cutout area in the received captured image (S102). For example, the image processing unit 73 converts the captured image into a binary (black and white) image. The image processing unit 73 may identify a white area in the binary image as the cutout area. Alternatively, the image processing unit 73 may identify the cutout area by pattern matching the shape of the white area in the binary image with a predetermined shape. The predetermined shape may be, for example, the shape of the image of the drug placed on the drug placing table 133a in an image captured of the drug placing table 133a. For example, when generating a cutout image using the shape of the image of the drug placing table 133a in an image captured of the drug placing table 133a, the image processing unit 73 may generate each cutout image from the following two captured images: one is an image of the front side of the tablet, and the other is an image of the back side of the tablet. Here, the front surface refers to one side of the tablet on which the identification information is displayed, and the back surface refers to the other side of the tablet that is different from the one side on which the identification information is displayed.

Next, the image processing unit 73 generates a cut-out image 821 by cutting out the region identified in S102 from the captured image D100 (S103). The image processing unit 73 also generates a region information file 822 (S104). Next, it generates a compressed image 823 (S105). Figure 18 shows an example of the captured image D100, cut-out image D101, region information file D102, and compressed image D103 of the target drug.

For example, the cropped image D101 may be generated as PNG format data. Furthermore, as shown in FIG. 18, the region information file D102 indicates the position of the cropped image D101 in the captured image D100. Furthermore, the compressed image D103 may be generated as JPEG format data obtained by compressing the captured image D100. For example, an example of the compressed image D103 is JPEG format image data obtained by resizing the captured image D100 to 1/4 its original size. In other words, by resizing the captured image D100 to 1/4 its original size, the amount of data can be compressed.

Next, the image processing unit 73 stores the cropped image 821, the region information file 822, and the compressed image 823 in the storage unit 80b (S106), and the processing ends.

Then, the medicine sorting device 1 transmits the cut-out image 821, the region information file 822, and the compressed image 823 to the information processing device 200. This allows the information processing device 200 to restore the captured image D100 based on, for example, the cut-out image D101, the region information file D102, and the compressed image D103.

Note that multiple captured images may be used to determine the type of a single target drug. For example, if the target drug is a tablet, the determination may be made using 10 captured images. If the target drug is a capsule, the determination may be made using up to 10 additional captured images in addition to the 10 captured images. For example, the image processing unit 73 may determine whether the target drug captured in the received captured images is a tablet or a capsule. Note that the image processing unit 73 may perform this determination by pattern matching between the image of the target drug captured in the received captured images and a predetermined shape. If the image processing unit 73 determines that the target drug is a tablet, it may not process any captured images other than the first 10 received images and may not store those captured images in the memory unit 80b. This is because if the target drug is a tablet, the type can be determined using 10 captured images. On the other hand, if the target drug is a capsule, the type may not be determined using 10 captured images. Therefore, if the target drug is a capsule, the image processing unit 73 may store in the memory unit 80b more than 10 captured images acquired up until the time it has been determined that the target drug is a capsule.

(Modification)
A modification of this embodiment will be described with reference to Fig. 19. For ease of explanation, members having the same functions as those described in the above embodiment will be denoted by the same reference numerals, and description thereof will not be repeated.

The image processing unit 73 in this modification generates serialized data including the cut-out image, cut-out area image, area information file, information indicating the color of the target drug, information indicating the shape of the target drug, etc., and consolidates them into a single file. The image processing unit 73 stores this file on a specified backup disk. For example, serialized data is data in which specific data is arranged in a regular pattern. In this modification, an image cut out from the range of the image of the target drug in the captured image is called a cut-out image. Here, the "range of the image of the target drug" is the same as the "area of the image of the target drug (first area)" in embodiment 10 above. For ease of explanation, in this modification, it is called the "range of the image of the target drug."

In this modified example, an image obtained by cutting out the image of the medicine placing table 133a including the image of the target medicine in the captured image is referred to as a cut-out area image. In addition, the "range of the image of the medicine placing table 133a on which the target medicine including the image of the target medicine is placed" in this modified example is the same as the "area (second area) of the image of the medicine placing table 133a on which the target medicine including the image of the target medicine is placed" in the above-mentioned embodiment 10. For ease of explanation, in this modified example, it is referred to as the "range of the image of the medicine placing table 133a on which the target medicine including the image of the target medicine is placed." In addition, in this modified example, a cut-out image of a predetermined range centered on the image of the target medicine in the captured image is also referred to as a cut-out area image.

In addition, the information processing device 200 according to this modified example deserializes (restores serialization) serialized data and returns it to the file format before serialization. The serialized data includes, for a discrimination process for one target drug type, the number of discrimination processes, data indicating the drug's color and shape, a cut-out image generated from the captured image, and aggregated cut-out data.

For example, the information processing device 200 creates a black background of the same size as the captured image from which the cropping has been performed. The information processing device 200 generates a reference image by pasting the cropped image into the area of the black background that corresponds to the area from which the cropping has been performed in the captured image based on the region information file.

(Generating image set classes and lightweight classes)
The image processing unit 73 according to this modification generates an image set class for each discrimination process for a single target drug type. The image processing unit 73 prepares in advance, as an image set class, a data storage area capable of storing the maximum number of cutout images and cutout region images that can be generated in one discrimination process. In this embodiment, the image set class can store cutout images and cutout region images generated from a maximum of 20 captured images.

For example, if the matching unit 64b is able to determine the type from a predetermined number of captured images (e.g., 10 images) in one discrimination process, the image processing unit 73 will collect the cut-out images and cut-out area images generated from the predetermined number of captured images (e.g., 10 images) and store them in an image set class prepared in advance for that discrimination process.

If the matching unit 64b is unable to determine the type from a predetermined number of captured images (e.g., 10), the imaging control unit 63 controls the first camera 131 to acquire additional captured images. The imaging control unit 63 continues to acquire captured images until the matching unit 64b is able to determine the type from the acquired captured images. The image processing unit 73 also stores the cropped images and cropped area images generated from the captured images acquired in excess of the predetermined number in the above image set class.

For example, when the matching unit 64b completes the identification of one target drug, the image processing unit 73 adds the image set class corresponding to each identification process to the image set class list for the lighter class. The drug sorting device 1 sends the image set class list for the lighter class to the information processing device 200, which can then deserialize the data included in the image set class for each target image to generate a reference image.

The discrimination unit 64 may be configured to perform image analysis on the captured image 82 again to attempt to discriminate the type of drug (perform discrimination retry processing) if the type of drug cannot be determined in a single discrimination process. When a discrimination retry processing occurs, the image processing unit 73 according to this modified example generates an image set class corresponding to the discrimination retry processing.

(Example of cutout processing)
Next, an example of the process of cutting out from the captured image by the image processing unit 73 according to this modification will be described with reference to Fig. 19. Fig. 19 is a diagram showing an example of the process of cutting out from the captured image performed by the image processing unit 73.

First, two patterns of captured images will be explained. There are two patterns of captured images: a horizontal image (N image) and a vertical image (V image), which correspond to the inclination of the medicine placing table 133a when the captured image is taken. N in Figure 19 is a diagram showing the inclination of the medicine placing table 133a when the captured image is taken as an N image. As shown in N in Figure 19, when the captured image is taken as an N image, the N image is taken with one of the inclined surface portions 133ab of the medicine placing table 133a horizontal.

V in Figure 19 shows the inclination of the medicine placing table 133a when the captured image is a vertical image (V image). As shown in V in Figure 19, when the captured image is a V image, the V image is captured with the surface of the bottom surface 133aa of the medicine placing table 133a horizontal.

The image processing unit 73 performs cropping processing on the captured image based on the pattern of the captured image (V image, N image), the type of lighting used when capturing the captured image, etc.

(1) An example of the cutout process by the image processing unit 73 for an N image (bar illumination N image) captured using the first irradiator 134a (bar illumination) during shooting will be described using captured image D104 in Figure 19. Captured image D104 in Figure 19 is a bar illumination N image. Captured image D104 in Figure 19 shows range J104 of the image of the target drug, which is the range within which cutout will be performed, and range R104 of the image of the drug placing table 133a including the image of the target drug.

The image processing unit 73 performs binarization processing on the captured image. The image processing unit 73 detects the white area of the binarized captured image as the area to be cut out. As shown in captured image D104 in Figure 19, for example, the cut-out areas detected by the image processing unit 73 are the area J104 of the image of the target drug and the area R104 of the image of the drug placement table 133a including the image of the target drug. For example, the image processing unit 73 may use captured images binarized using different thresholds to detect the area of the image of the target drug and the area of the image of the drug placement table 133a including the image of the target drug. The image processing unit 73 cuts out the area of the image of the target drug from the captured image to generate a cut-out image. The image processing unit 73 also cuts out the area of the image of the drug placement table 133a including the image of the target drug to generate a cut-out area image.

(2) An example of the cropping process by the image processing unit 73 for a captured image (bar illumination V image) captured using the first illuminator 134a (bar illumination) during shooting is described using captured image D105 in Figure 19.

The captured image D105 in Figure 19 is a bar-illuminated V image. The captured image D105 in Figure 19 shows the range J105 of the image of the target drug, which is the range for cutting out, and a predetermined range R105 centered on the image of the target drug.

As shown in captured image D105 in Figure 19, for example, the image processing unit 73 detects the range J105 of the image of the target drug by pattern matching. The above-mentioned predetermined shape may be, for example, the shape of the image of the drug in an image captured after placing the drug on the drug placement table 133a. Next, the image processing unit 73 determines a range R105, which is a predetermined range centered on the range J105 of the image of the target drug. For example, range R105 is a range obtained by enlarging range J105 of the image of the target drug by a predetermined magnification, with range J105 of the image of the target drug at its center. The above-mentioned predetermined magnification may be determined, for example, from the degree of reduction in the volume of the image data or the results of verifying the quality of the restored image.

The image processing unit 73 cuts out the area of the image of the target drug from the captured image to generate a cut-out image. The image processing unit 73 also cuts out a predetermined area centered on the image of the target drug to generate a cut-out area image.

(3) An example of the cropping process by the image processing unit 73 for an N image (ring illumination N image) captured using the second illuminator 134b (ring illumination) during shooting will be described using captured image D106 in Figure 19.

The image processing unit 73 references the range of the image of the target drug detected in the bar illumination N image (the relative position of the range of the image of the target drug in the captured image) and identifies the position of the range of the image of the target drug in the ring illumination N image. In particular, the image processing unit 73 identifies the position of the range of the image of the target drug in the ring illumination N image so that the relative position of the range of the image of the target drug in the ring illumination N image is the same as the relative position of the range of the image of the target drug in the bar illumination N image.

The captured image D106 in Figure 19 is a ring illumination N image. The captured image D106 in Figure 19 shows the range J106 of the image of the target drug, which is the range from which extraction will be performed.

For example, in the example shown in Figure 19, the image processing unit 73 refers to the relative position of the range J104 of the image of the target drug in the captured image D104, which is the bar illumination N image, and identifies the position of the range J106 of the image of the target drug in the captured image D106.

(4) This section describes an example of cropping processing by the image processing unit 73 for a captured image (ring illumination V image) captured using the second illumination unit 134b (ring illumination) during shooting.

The image processing unit 73 references the range of the image of the target drug detected in the bar illumination V image (the relative position of the range of the image of the target drug in the captured image) and identifies the position of the range of the image of the target drug in the ring illumination V image. In detail, the image processing unit 73 identifies the position of the range of the image of the target drug in the ring illumination V image so that the relative position of the range of the image of the target drug in the ring illumination V image is the same as the relative position of the range of the image of the target drug in the bar illumination V image.

For example, in the example shown in Figure 19, the image processing unit 73 refers to the relative position of the range J105 of the image of the target drug in the captured image D105, which is the bar illumination V image, and identifies the position of the range of the image of the target drug in the ring illumination V image (not shown).

(5) This section describes an example of cropping processing by the image processing unit 73 for an N image (UV illumination N image) captured using the ultraviolet light irradiation unit 134c (UV illumination) during shooting.

The image processing unit 73 references the range of the image of the target drug detected in the bar illumination N image (the relative position of the range of the image of the target drug in the captured image) and identifies the position of the range of the image of the target drug in the UV illumination N image. In particular, the image processing unit 73 identifies the position of the range of the image of the target drug in the UV illumination N image so that the relative position of the range of the image of the target drug in the UV illumination N image is the same as the relative position of the range of the image of the target drug in the bar illumination N image.

For example, in the example shown in Figure 19, the image processing unit 73 refers to the relative position of the range J104 of the image of the target drug in the captured image D104, which is the bar illumination N image, and identifies the position of the range of the image of the target drug in the UV illumination N image (not shown).

Note that the image processing unit 73 does not need to generate a crop area image for images captured using ring illumination or UV illumination during shooting.

The processes (1) to (5) above may be performed on the predetermined number of captured images. That is, the predetermined number of captured images are a bar illumination N image, a bar illumination V image, a ring illumination N image, a ring illumination V image, and a UV illumination N image. The bar illumination N image, bar illumination V image, ring illumination N image, ring illumination V image, and UV illumination N image may be captured by positioning the imaging mechanism so that the angle θ between the axis Ax0 and the axis Ax1 shown in FIG. 4 is 0° and 180°. In this case, the predetermined number of captured images is a total of 10 captured images: the bar illumination N image, bar illumination V image, ring illumination N image, ring illumination V image, and UV illumination N image, captured by positioning the imaging mechanism so that θ is 0° and 180°. On the other hand, the captured images acquired in excess of the predetermined number are bar illumination V images captured while changing the imaging angle. Therefore, the process (2) above is performed on these images.

When comparing the data capacity of a captured image without the above-mentioned processing with the data capacity of the serialized data after the above-mentioned processing, the data capacity of the serialized data relative to the data capacity of a captured image without the cutout processing was as follows.
- For 39 engraved tablets, the data volume of the captured images without cutting out was 23.2%.
- For 39 printed tablets, the data volume of the captured image without cutting out is 21.5%.
For a 20-tablet capsule, the data volume was 34.5% of the data volume of a captured image without cutting out. Furthermore, it was confirmed that when the serialized data was zip-compressed, the data volume was approximately 10% of the data volume of a captured image without cutting out.

This modified example can be rephrased as follows:

In order to generate an image of a reference drug for determining the type of target drug, the image processing unit 73 of the drug sorting device 1 in this modified example generates at least one cut-out image by cutting out the range of the image of the target drug (first region) and the range of the image of the drug placement table 133a on which the target drug is placed, including the image of the target drug, from a captured image including the image of the target drug.

The image processing unit 73 generates at least one of a region information file indicating a first region in the captured image and a region information file indicating a second region.

The information processing device 200 in this modified example can be communicatively connected to the medicine sorting device 1 and generates an image of the reference medicine. The acquisition unit 221 of the information processing device 200 receives a cut-out image and a region information file corresponding to the cut-out image from the medicine sorting device 1. In addition, the image generation unit (generation unit) 222 generates a reference image based on the region information file.

The image generation system 1000 of this modified example includes a medicine sorting device 1 and an information processing device 200.

[Embodiment 11]
(overview)
In this embodiment, a configuration for reducing the replacement cost of the sorting cup 141 will be described mainly with reference to FIGS.

The medicine sorting device 1 according to this embodiment generates a shape model 800 that represents the shape of a scratch on the sorting cup 141 and stores the generated shape model 800. When detecting medicines from a captured image of the sorting cup 141 to check for remaining medicines, the medicine sorting device 1 uses the shape model 800 to prevent the scratch on the sorting cup 141 from being detected as medicine. With the above configuration, a scratched sorting cup 141 can continue to be used. This reduces the cost of replacing the sorting cup 141.

(Medicine sorting device 1)
The medicine sorting device 1 according to this embodiment will be described with reference to Fig. 20. Fig. 20 is a block diagram showing an example of a partial configuration of the medicine sorting device 1 according to this embodiment. Note that Fig. 20 shows only the configuration of the medicine sorting device 1 that is related to the shape model generation process of the sorting cup 141 and the medicine detection process in the sorting cup 141 according to this embodiment. The medicine sorting device 1 according to this embodiment may include all of the configuration provided in the medicine sorting device 1 shown in Fig. 1.

(Operation unit 31)
The operation unit 31 of the touch panel 3 accepts input operations by the user. In particular, the operation unit 31 according to this embodiment accepts a user input instructing the generation of a shape model 800 of the sorting cup 141. The operation unit 31 also accepts a user input instructing the checking of remaining medicines in the sorting cup 141. The checking of remaining medicines refers to the medicine sorting device 1 determining whether or not the sorting cup 141 contains any medicines based on a captured image of the sorting cup 141.

(Control unit 60c)
The control unit 60c according to this embodiment includes, in addition to the main components of the control unit 60a described in the first embodiment, a shape model generation unit 791c, a drug detection unit 792c, and a matching determination unit 793c.

(Imaging control unit 63)
The imaging control unit 63 causes the second camera 121 to capture an image of the inside of the sorting cup 141 in accordance with instructions received from the operation input unit 66. The imaging control unit 63 acquires the captured image from the second camera 121. When the operation input unit 66 accepts a user input instructing the generation of a shape model 800 of the sorting cup 141, the imaging control unit 63 transmits the acquired captured image to the shape model generation unit 791c. Furthermore, when the operation input unit 66 accepts a user input instructing the confirmation of remaining medicine in the sorting cup 141, the imaging control unit 63 transmits the acquired captured image to the medicine detection unit 792c.

(Operation input unit 66)
The operation input unit 66 accepts input by the user on the operation unit 31. In this embodiment, in particular, the operation input unit 66 accepts a user input instructing the generation of a shape model 800 of the sorting cup 141 and a user input instructing the confirmation of remaining medicine in the sorting cup 141.

When a user input instructing the generation of a shape model 800 of the sorting cup 141 and a user input instructing the confirmation of remaining medication in the sorting cup 141 are received, the operation input unit 66 instructs the imaging control unit 63 to capture an image of the sorting cup 141 using the second camera. Note that the user input instructing the generation of the shape model 800 of the sorting cup 141 may also include an input specifying the sorting cup 141.

(Display control unit 67)
The display control unit 67 receives a signal indicating the detection result of the medicine image from the medicine detection unit 792c. The display control unit 67 displays a remaining medicine confirmation image on the display unit 32. For example, the remaining medicine confirmation image may display an image of each sorting cup 141 and also display "remaining medicine" or "no remaining medicine" for that sorting cup 141.

(Shape model generation unit 791c)
The shape model generation unit 791c receives the captured image from the imaging control unit 63. The shape model generation unit 791c generates a shape model 800 of the sorting cup 141 from the captured image. The process of generating the shape model 800 will be described in detail below. The shape model generation unit 791c associates the shape model 800 with the UID (unique identification) of the RFID tag of the corresponding sorting cup 141 and stores it in the storage unit 80c. The shape model 800 may include area information indicating the area of the scratch in the sorting cup 141 in addition to the shape model indicating the shape of the scratch.

(Drug detection unit 792c)
The drug detection unit 792c receives the captured image from the imaging control unit 63. The drug detection unit 792c detects the image of the drug from the captured image. When the drug detection unit 792c detects the image of the drug from the captured image, it transmits the captured image to the matching determination unit 793c.

For example, the drug detection unit 792c converts the captured image into a binarized image. The drug detection unit 792 may identify white areas in the binarized image as images of drugs. The drug detection unit 792c may also identify images of drugs by pattern matching the shape of the white areas in the binarized image with a predetermined shape. The predetermined shape may be, for example, the shape of an image of a drug in an image captured in advance.

The medicine detection unit 792c also receives a signal indicating a matching area from the matching determination unit 793c. The matching area will be described in detail below. The medicine detection unit 792c excludes the matching area indicated by the signal from the area to be detected and detects an image of the medicine. If a medicine image is detected, the medicine detection unit 792c sends a signal to the display control unit 67 indicating that there is remaining medicine in the corresponding sorting cup 141. If the medicine detection unit 792c also receives a signal from the matching determination unit 793c indicating that no matching area has been detected, it sends a signal to the display control unit 67 indicating that there is remaining medicine in the corresponding sorting cup 141. If the medicine detection unit 792c does not detect an image of the medicine from the captured image excluding the matching area or from the captured image received from the imaging control unit 63, the medicine detection unit 792c sends a signal to the display control unit 67 indicating that there is no remaining medicine in the corresponding sorting cup 141.

(Matching determination unit 793c)
The matching determination unit 793c receives the captured image from the drug detection unit 792c. The matching determination unit 793c matches the shape model 800 in the storage unit 80c with the image detected as the drug image in the captured image and calculates a matching score. The matching determination unit 793c determines whether the matching score is greater than a predetermined value. If the matching score is greater than the predetermined value, the matching determination unit 793c identifies the area of the image detected as the drug image as a matching area. The matching determination unit 793c transmits a signal indicating the matching area to the drug detection unit 792c. If the matching score is equal to or less than the predetermined value, the matching determination unit 793c transmits a signal indicating that a matching area has not been detected to the drug detection unit 792c.

As described above, a matching region is a region where the matching score is greater than a predetermined value. In other words, a matching region is a region of the captured image in which an image of a shape similar to the shape of the flaw shown in shape model 800 is captured. Furthermore, a value greater than the above-mentioned predetermined value (threshold value) is a value that can be considered to indicate a high degree of similarity between the shape of the flaw shown in shape model 800 and the image in the captured image. A specific example of the above-mentioned predetermined value is 0.75 (75 points). For example, the matching score is a value where a perfect match is 1 (100 points) and dissimilarity is 0 (0 points).

(Example of generation process of shape model 800)
21 is a flowchart showing an example of a shape model generation process for the sorting cup 141 performed by the medicine sorting device 1 of this embodiment. As shown in FIG. 21 , the operation input unit 66 receives an instruction to generate a shape model 800 for the sorting cup 141 (S201). For example, when checking the remaining medicines after removing medicines from the sorting cup 141, if the user determines that the sorting cup 141 has a scratch, the user inputs an instruction to generate the shape model 800 for the sorting cup 141 from the operation unit 31.

Next, the imaging control unit 63 acquires an image of the interior of the sorting cup 141 from the second camera 121 (S202). Next, the shape model generation unit 791c generates a shape model 800 of the sorting cup 141 from the captured image (S203). Figure 22 is a diagram showing an example of an image of the sorting cup 141 and a shape model 800 of the sorting cup 141. The captured image D200 shown in Figure 22 is an example of an image of the interior of the sorting cup 141 captured by the second camera 121. The captured image D200 shows an area R20, in which one scratch is present. The image D201 shown in Figure 22 is an enlarged image of the area R20. The shape model 800E1 shown in Figure 22 shows the shape model 800 corresponding to the scratch detected in image D201.

For example, shape model 800E1 is the shape of a scratch on the sorting cup 141 extracted from image D201. Note that shape model 800 may show only a portion of the shape of the scratch on the sorting cup 141. Image D202 shown in Figure 22 is an example of an enlarged image of a captured image of the inside of another sorting cup 141. There are multiple scratches on the sorting cup 141 captured in image D202. If multiple scratches are detected in the captured image of the sorting cup 141, the shape model generation unit 791c may generate a shape model 800 corresponding to each of the detected scratches. Shape models 800E2 to 800E4 shown in Figure 22 show shape models 800 corresponding to each of the multiple scratches detected in image D202.

Next, the shape model generation unit 791c stores the generated shape model 800 in the storage unit 80c (S204).

(Example of drug detection process for checking remaining medication)
23 is a flowchart showing an example of a medicine detection process for checking remaining medicines by the medicine sorting device 1 of this embodiment. As shown in Fig. 23, the operation input unit 66 receives an instruction to check remaining medicines in the sorting cup 141 (S301). Next, the imaging control unit 63 acquires an image of the inside of the sorting cup 141 from the second camera 121 (S302).

Next, the drug detection unit 792c detects drug images from the captured image (S303). If the drug detection unit 792c detects drug images from the captured image (YES in S303), the matching determination unit 793c matches the shape model 800 stored in the memory unit 80c with the image detected as a drug in the captured image (S304). Next, the matching determination unit 793c determines whether the matching score is greater than a predetermined value (S305).

Figure 24 shows an example of a captured image of the sorting cup 141, an extracted image in which an image of the inside of the sorting cup 141 has been extracted, a shape model 800, and an image of a matching area in the extracted image. The captured image D203 shown in Figure 24 is an example of a captured image of the sorting cup 141. The matching determination unit 793c may generate an extracted image in which an image of the inside of the sorting cup 141 has been extracted from the captured image of the sorting cup 141. The extracted image D204 shown in Figure 24 is an extracted image in which an image of the inside of the sorting cup 141 has been extracted from the captured image D203. The matching determination unit 793c may perform matching using the extracted image as the captured image. The shape model 800F1 shown in Figure 24 is an example of a shape model of a scratch in the sorting cup 141. Furthermore, image D205 shown in Figure 24 is an example of an image of a matching area corresponding to the shape model 800F1 in the extracted image D204. As shown in FIG. 24, the matching determination unit 793c searches the extracted image D204 for an image pattern similar to the shape model 800F1, and calculates a matching score that indicates how similar the searched image pattern is to the shape model 800F1. FIG. 24 shows an example in which the matching score calculated by the matching determination unit 793c is 0.93.

If the matching determination unit 793c determines that the matching score is greater than a predetermined value (threshold value) (YES in S305), the medication detection unit 792c excludes the matching area from the area in the captured image where medication is detected (S306) and detects medication images (S307). If the medication detection unit 792c detects medication images (YES in S307), the display control unit 67 displays a remaining medication confirmation image on the display unit 32 that displays "Remaining medication" on the corresponding sorting cup 141 (S308), and the process ends.

Image D206 in Figure 24 shows an example of the drug detection results of the drug detection unit 792c when the matching area is excluded from the detection target area. As shown in image D206 in Figure 24, the drug detection unit 792c detects only the drug image J200 and does not detect the image of the wound as a drug.

If the drug detection unit 792c does not detect an image of a drug from the captured image (NO in S303), the display control unit 67 displays a remaining drug confirmation image on the display unit 32 that displays "No remaining drug" in the corresponding sorting cup 141 (S309), and the process ends. Also, if the matching determination unit 793c determines that the matching score is below a predetermined value (NO in S305), the process continues to 308. Also, if the drug detection unit 792c does not detect an image of a drug (NO in S307), the process continues to 309.

[Embodiment 12]
In this embodiment, a configuration that can assist the user in excluding medicines in poor condition from the sorted medicines will be described with reference to Fig. 25. For example, medicines in poor condition include medicines that are partially missing, soiled, discolored, etc.

The medicine sorting device 1 according to this embodiment displays buttons that allow the user to select the status of each medicine on an image (image for visual inspection) that allows the user to visually inspect the sorted medicines. The medicine sorting device 1 prints the status of each medicine on at least one of the journal and packaging paper in accordance with the received user input.

With the above configuration, if the sorted medications include medications in poor condition, the user can easily determine the number of medications in poor condition. Therefore, the medication sorting device 1 can assist the user in removing medications in poor condition from the sorted medications. Furthermore, by making the user aware that the sorted medications include medications in poor condition, it is possible to reduce the risk of missing medications in poor condition when being removed.

In this embodiment, the operation unit 31, control unit 60a, and print output unit 4 perform the following processes in addition to the processes described in embodiment 1.

(Operation unit 31)
The operation unit 31 accepts input operations by the user. In particular, the operation unit 31 according to this embodiment accepts user input for displaying an image for visual inspection. The operation unit 31 also accepts user input entered using the buttons displayed on the image for visual inspection. The operation unit 31 transmits a signal indicating the accepted user input to the operation input unit 66.

(Imaging control unit 63)
The imaging control unit 63 receives an instruction from the operation input unit 66 and causes the second camera 121 to capture an image of the sorting cup 141. The imaging control unit 63 acquires the captured image and transmits it to the display control unit 67.

(Operation input unit 66)
When the operation input unit 66 receives a signal indicating a user input for displaying an image for visual inspection, it instructs the imaging control unit 63 to capture an image of the sorting cup 141 with the second camera 121. The operation input unit 66 also receives a signal indicating a user input entered using the button displayed on the image for visual inspection. The operation input unit 66 transmits the signal indicating the user input to the print output control unit 69.

(Display control unit 67)
The display control unit 67 receives the captured image from the imaging control unit 63, generates an image for visual inspection using the captured image, and displays it on the display unit 32. The image for visual inspection includes a button that allows the user to select the condition of each medicine. The button allows the user to input the condition of each medicine sorted in the sorting cup 141. For example, the button allows the user to select the condition of the medicine, such as whether the medicine is dirty, whether there are any chips in the medicine, or whether the medicine has discolored.

(Print output control unit 69)
The printout control unit 69 controls the printout unit 4 in accordance with the user input received by the operation input unit 66 .

(Print output unit 4)
The printing output unit 4 prints the status of each medicine on at least one of a journal and a packaging paper in accordance with user input.

(Processing example)
FIG. 25 is a flowchart illustrating an example of the printing process of the medicine sorting device according to this embodiment. As shown in FIG. 25 , the operation input unit 66 accepts user input for displaying an image for visual inspection (S401). Next, the imaging control unit 63 controls the second camera 121 to capture an image of the sorting cup 141 (S402). Next, the display control unit 67 displays an image for visual inspection, including buttons that allow the user to select the status of each medicine (S403). Next, the operation input unit 66 accepts user input for the buttons regarding the status of each medicine (S404). The user visually checks the images for visual inspection and visually inspects the medicines shown in each image. If the user finds any problems with the medicines during visual inspection, the user selects the button indicating "Yes" for the corresponding problem. If no problems are found, the user selects the button indicating "No" for each problem. In response to this selection, the operation input unit 66 accepts user input for the buttons.

The printout control unit 69 then controls the printout unit 4 to print the status of each medication on the journal (S405), and the process ends. For example, if the user performs a visual inspection before the sorted medications are packaged, the status of each medication may be printed on both the journal paper and the packaging paper. Alternatively, if the user performs a visual inspection after the sorted medications are packaged, the status of each medication may be printed only on the journal paper. For example, the status of the medication and the number of medications in that state may be printed for each sorting cup 141, such as "Dirty: 1 tablet present, Missing: 2 tablets present."

[Embodiment 13]
(overview)
In this embodiment, mainly using Figures 26 to 28, a configuration that can increase the probability of successful identification of the type of target drug and can increase the accuracy of identification of the type of target drug will be described.

When the identification information of a target drug contains unidentifiable characters, the drug sorting device 1 according to this embodiment generates a converted string in which the unidentifiable characters in the character string of the identification information are converted into wildcards that match all patterns. The drug sorting device 1 determines the type of target drug from the search results using the converted string against the user's adopted drug list 810 and the comprehensive drug database 541.

With the above configuration, if the identification information of the target drug contains unidentifiable characters, the drug sorting device 1 determines the type of the target drug from the search results using the user's adopted drug list 810 and the comprehensive drug database 541. As a result, the drug sorting device 1 can increase the probability that the target drug will be one of the drugs.

Here, Figure 26 is a diagram showing an example of a drug discrimination system 2000 according to this embodiment. As shown in Figure 26, a user-adopted drug list 810 is stored in the memory unit 80d of each drug sorting device 1. The user-adopted drug list 810 is a list containing characteristics (drug data) of reference drugs corresponding to drugs managed by the hospital, ward, pharmacy, etc. where the drug sorting device 1 is used. The drug data is associated with identification information of the reference drug (e.g., information indicating the type of drug, such as the drug name). The drug data may be image data including the characteristics of the reference drug.

As shown in FIG. 26, the comprehensive drug database 541 is stored in the memory unit 504 of the data management device 500. The data management device 500 manages drug data relating to multiple types of drugs that can be used in any hospital, ward, pharmacy, etc. The data management device 500 can be communicatively connected to the drug sorting device 1, and comprehensively and centrally manages multiple types of drugs in the comprehensive drug database 541.

Because the amount of drug data managed in the comprehensive drug database 541 is enormous, it is not realistic to include all of it in the user's adopted drug list 810. Therefore, the drug data included in the user's adopted drug list 810 is drug data extracted from the drug data included in the comprehensive drug database, which is drug data expected to be used by each drug sorting device 1, or each hospital, ward, pharmacy, etc.

(Drug Identification System 2000)
A drug discrimination system 2000 according to this embodiment will be described with reference to FIG. 26. As shown in FIG. 26, the drug discrimination system 2000 includes a plurality of drug sorting devices 1 and a data management device 500 that can be communicatively connected to each drug sorting device 1. The drug sorting devices 1 and the data management device 500 transmit and receive data via an internet line or a dedicated VPN (Virtual Private Network) line. Note that the drug discrimination system 2000 may include only one drug sorting device 1.

(Data management device 500)
The data management device 500 is a device used, for example, by the manufacturer of the drug sorting device 1. The data management device 500 includes a communication unit 501, a control unit 502, a touch panel 503, and a storage unit 504. The storage unit 504 stores a comprehensive drug database 541. The communication unit 501 transmits and receives data to and from the drug sorting device 1. The communication unit 501 receives, for example, a conversion character string from the communication unit 101 of the drug sorting device 1. The control unit 502 searches the comprehensive drug database 541 using the conversion character string. The control unit 502 transmits drug data of the reference drug searched for in the comprehensive drug database 541 to the drug sorting device 1 via the communication unit 501. The touch panel 503 includes an operation unit and a display unit (not shown). For example, the drug data included in the comprehensive drug database 541 may be updated by a user operation via the touch panel 503.

(Medicine sorting device 1)
As shown in FIG. 26, the medicine sorting device 1 includes a touch panel 3, a control unit 60a, a memory unit 80d, and a communication unit 101.

Note that Figure 26 only shows the components of the medicine sorting device 1 that are related to the discrimination process of this embodiment. The medicine sorting device 1 of this embodiment may include all of the components of the medicine sorting device 1 shown in Figure 1.

(Touch panel 3)
An operation unit (not shown) of the touch panel 3 accepts input operations by the user. In particular, the operation unit according to this embodiment accepts a user operation to instruct adding drug data of a reference drug to the user adopted drug list 810.

For example, the drug data of a reference drug added to the user's adopted drug list 810 may be drug data of a reference drug that is not found in the user's adopted drug list 810 but is found in the comprehensive drug database 541 in a search using a conversion string. For example, when the operation input unit 66 receives a user operation instructing the addition of drug data of a reference drug, the drug database update unit 74 shown in FIG. 1 may add the drug data of the reference drug to the user's adopted drug list 810. In addition, the user may request the manufacturer of the drug sorting device 1 to add drug data of a reference drug that is not found in the user's adopted drug list 810 but is found in the comprehensive drug database 541 to the user's adopted drug list 810.

The control unit 60a in this embodiment has the following functions in addition to the functions described in embodiment 1.

(Feature extraction unit 64a)
The feature extraction unit 64a performs OCR processing on the identification information of the target drug. The feature extraction unit 64a also determines whether or not the character string of the identification information after OCR includes an unidentifiable character.

The feature extraction unit 64a also determines whether the character string of the identification information after OCR contains a predetermined number of characters equal to or greater than a first number. For example, the feature extraction unit 64a determines whether the total number of identifiable characters and unidentifiable characters contained in the character string of the identification information is equal to or greater than a first number. The first number may be a number set by the user.

The feature extraction unit 64a also determines whether the character string of the identification information after OCR contains more identifiable characters than a predetermined second number. For example, the feature extraction unit 64a may determine whether the number of identifiable characters is greater than the number of unidentifiable characters. If the number of characters included in the character string of the identification information after OCR is 8, the feature extraction unit 64a may determine whether the number of identifiable characters is 5 or more (greater than 4).

If the character string of the identification information after OCR contains a first number of characters or more and the character string of the identification information after OCR contains more than a second number of identifiable characters, the feature extraction unit 64a generates a converted character string. In particular, the feature extraction unit 64a generates a converted character string by converting unidentifiable characters in the character string of the identification information after OCR into wildcards. The feature extraction unit 64a transmits a signal indicating the generated converted character string to the matching unit 64b.

(Collation unit 64b)
The collating unit 64b searches for a reference drug of the same type as the target drug by collating the received converted character string with the character string of the identification information of the reference drug included in the user adopted drug list 810.

The matching unit 64b also transmits a signal indicating the conversion string to the data management device 500 via the communication unit 101. The matching unit 64b then acquires drug data for the reference drug searched for using the conversion string from the comprehensive drug database 541 from the data management device 500 via the communication unit 101. The matching unit 64b determines the type of target drug based on the search results. The matching unit 64b also updates the determination result information 820 stored in the memory unit 80d. The determination result information 820 includes the type of target drug determined by the matching unit 64b, drug data for the reference drug used in the determination, and all candidate drugs if there are multiple candidate target drugs. The determination result information 820 may be linked to the captured image 82 of the target drug and stored in the memory unit 80.

(Processing example)
27 is a flowchart showing an example of the discrimination process of the medicine sorting device 1 of this embodiment. As shown in FIG. 27, the feature extraction unit 64a performs OCR processing on the identification information of the target medicine (S501). Next, the feature extraction unit 64a determines whether the character string of the identification information after OCR contains an unidentifiable character (S502).

Figure 28 shows an example of the identification information for a target drug, the drug data for a reference drug in the user's adopted drug list 810, and the drug data for a reference drug in the comprehensive drug database 541. D301 in Figure 28 shows an example of the identification information for the image of the upper (front) side of the target drug. D302 in Figure 28 shows an example of the identification information for the image of the lower (back) side of the target drug. In the example shown in D301 in Figure 28, the identification information for the upper side of the target drug is displayed as "ABC" in the upper row and "123" in the lower row. As shown in C1 in D301 in Figure 28, after OCR processing, the letter "C" in the identification information in the upper row on the upper side of the target drug is an unidentifiable character. Furthermore, as shown in C2 in D301 in Figure 28, after OCR processing, the letter "2" in the lower row on the upper side of the target drug is an unidentifiable character. In the example shown in D302 of FIG. 28, the identification information "20" is displayed on the lower side of the target drug. As shown in C3 of D302 of FIG. 28, after OCR processing, the character "2" in the identification information on the lower side of the target drug is an unidentifiable character. In the examples shown in D301 and D302 of FIG. 28, the characters shown in C1, C2, and C3 are unidentifiable characters, so the feature extraction unit 64a determines that the character string of the identification information after OCR includes an unidentifiable character.

If the character string of the identification information contains an unidentifiable character (YES in S502), the feature extraction unit 64a determines whether the character string of the identification information after OCR contains a first number or more of characters (S503). For example, in the example shown in D301 and D302 of FIG. 28, regardless of whether the characters are identifiable or unidentifiable, six-character identification information, "ABC" and "123," is displayed on the upper side, and two-character identification information, "20," is displayed on the lower side. That is, in the example shown in D301 and D302 of FIG. 28, a total of eight characters of identification information are displayed. If the first number is set to a number equal to or less than eight (e.g., 5), in the example shown in D301 and D302 of FIG. 28, the feature extraction unit 64a determines that the character string of the identification information after OCR contains a first number or more of characters.

If the character string of the identification information after OCR contains more than the first number of characters (YES in S503), the feature extraction unit 64a determines whether the character string of the identification information after OCR contains more than the second number of identifiable characters (S504). For example, in the example shown in D301 and D302 of FIG. 28, the "A" and "B" in the upper row on the upper side of the target drug, the "1" and "3" in the lower row, and the "0" on the lower side are identifiable characters, meaning that the character string of the identification information contains five identifiable characters. For example, the second number may be the number of unidentifiable characters included in the character string of the identification information displayed on the target drug. In this case, in the example shown in D301 and D302 of FIG. 28, the number of unidentifiable characters is three, and the number of identifiable characters (5) is greater than the number of unidentifiable characters (3). Therefore, the feature extraction unit 64a determines that the character string of the identification information after OCR contains more identifiable characters than the second number (the number of unidentifiable characters).

If the character string of the identification information after OCR contains more than the second number of identifiable characters (YES in S504), the feature extraction unit 64a converts the unidentifiable characters in the character string of the identification information after OCR to wildcards (S505) and generates a converted character string. For example, in the example shown in D301 and D302 in Figure 28, the feature extraction unit 64a may generate a converted character string as follows:

First, the feature extraction unit 64a may divide the string of identification information into display areas using delimiters ("space", ":"). The display areas are the upper row on the upper side, the lower row on the upper side, the lower side, etc. In the example shown in D301 and D302 of FIG. 28, the feature extraction unit 64a generates "ABx 1x3:x0" (x is an unidentifiable character). Next, the feature extraction unit 64a converts the unidentifiable character to a wildcard (*). The feature extraction unit 64a also adds a wildcard (*) to the beginning and end of the string for each display area. That is, the string on the upper row on the upper side becomes "*AB**", the string on the lower row on the upper side becomes "*1*3*", and the string on the lower side becomes "**0*". Note that the wildcard (*) indicates any string of zero or more characters.

The process of adding wildcards (*) to the beginning and end of a character string is a process to prepare for the case where there is a character before the beginning or after the end of the character string that the feature extraction unit 64a is unable to recognize as a character. This process makes it possible to properly search for a reference drug when there is a character before the beginning or after the end of the character string that the feature extraction unit 64a is unable to recognize as a character.

The matching unit 64b then searches for a reference drug of the same type as the target drug using the user's adopted drug list 810 (S506). D303 in Figure 28 shows an example of drug data for a reference drug searched using the user's adopted drug list 810. As shown in D303 in Figure 28, the reference drug (adopted drug MB master) has identification information "ABC" displayed in the upper row (Upper 1) on the upper side and "123" displayed in the lower row (Upper 2) on the upper side. Furthermore, the reference drug has identification information "30" displayed on the lower side. In this example, the adopted drug MB master searched using the user's adopted drug list 810 is a different type of drug from the target drug.

The collation unit 64b then obtains the search results using the comprehensive drug database 541 (S507). D304 in FIG. 28 shows an example of drug data for a reference drug searched using the comprehensive drug database 541. As shown in D304 in FIG. 28, the reference drug (drug MA master) has identification information "ABC" displayed in the upper row (Upper 1) on the upper side and "123" displayed in the lower row (Upper 2) on the upper side. Furthermore, the reference drug has identification information "20" displayed on the lower side. In this example, drug A searched using the comprehensive drug database 541 is the same type of drug as the target drug.

The collation unit 64b then determines whether any reference drugs have been found (S508). Specifically, the collation unit 64b determines whether a reference drug of the same type as the target drug has been found in either the search using the user's adopted drug list 810 or the search using the comprehensive drug database 541.

If there is a reference drug found (YES in S508), the collation unit 64b determines whether there are multiple reference drugs found (S509). For example, the collation unit 64b determines that there are multiple reference drugs found in the following cases:

(1) When one reference drug is found in a search using the user's adopted drug list 810 and one reference drug is found in a search using the comprehensive drug database 541. (2) When multiple reference drugs are found in at least one of a search using the user's adopted drug list 810 and a search using the comprehensive drug database 541.

If multiple reference drugs are found (YES in S509), the collation unit 64b determines the type of the target drug as one of multiple candidate drugs (S510). For example, if a reference drug is found in both a search using the user's adopted drug list 810 and a search using the comprehensive drug database 541, the collation unit 64b may determine the reference drug found in the search using the user's adopted drug list 810 as the type of target drug. Furthermore, a discrimination priority indicating the discrimination priority may be set for the reference drugs included in the user's adopted drug list 810 and the comprehensive drug database 541. The collation unit 64b may determine the type of target drug based on the discrimination priority. Next, the collation unit 64b stores all of the multiple candidate drugs that are found as reference drugs in the storage unit 80d as discrimination result information 820 (S511), and the processing ends.

Note that if the string of identification information does not contain any unidentifiable characters (NO in S502), the processing ends. For example, the determination processing described in embodiment 1 may be performed without converting unidentifiable characters in the string of identification information into wildcards.

If the character string of the identification information after OCR does not contain more than the first number of characters (NO in S503), the matching unit 64b determines that the target drug is a suspected drug (S513), and processing ends. If the character string of the identification information after OCR does not contain more than the second number of identifiable characters (NO in S504), processing continues to S513. If no reference drug was found (NO in S508), processing continues to S513.

Also, if the searched reference drug is not multiple (single) (NO in S509), the matching unit 64b determines the type of the target drug as the searched reference drug (S512), and processing ends.

For example, after the above-mentioned discrimination process of the medicine sorting device 1 is completed, if the operation input unit 66 receives user input via the operation unit 31 to display an image for visual inspection, the following process is performed. The display control unit 67 causes the display unit 32 to display a display image indicating a warning for target drugs that have multiple searched reference drugs. The image for visual inspection includes a warning indicating that there are multiple candidates, drug data for the multiple candidate drugs, etc. In the example shown in Figure 28, the display image includes drug data from the adopted drug MB master and drug MA master as candidate drugs for the target drug.

Furthermore, by checking this display image, if the drug data for the drug whose type has been determined is not included in the user's adopted drug list 810, the user inputs an instruction from the operation unit 31 of the touch panel 3 to add the drug data to the user's adopted drug list 810. When the operation input unit 66 accepts this instruction, the drug database update unit 74 adds the drug data for the reference drug to the user's adopted drug list 810.

[Embodiment 14]
(overview)
As described above, the medicine sorting device 1 is capable of sorting medicines stored in the first storage unit 11 by type. Medicines stored in the first storage unit 11 include returned medicines and patient-brought medicines. This embodiment mainly describes an example configuration of the first storage unit 11A assuming that the medicine sorting device 1 will sort patient-brought medicines, an example process when the medicine sorting device 1 sorts patient-brought medicines, and an example display image (display screen) that can be displayed when the medicine sorting device 1 sorts patient-brought medicines. Patient-brought medicines are medicines (adopted medicines) that were once prescribed to a patient at a medical institution, and then brought to another medical institution due to hospitalization or other reasons, and are currently being taken by the patient.

In this embodiment, the patient is prescribed medication for multiple days, with multiple dosing times set for each day. The same type of medication is taken at the same dosing time on each day. A specific day out of the multiple dosing days is referred to as a "predetermined dosing day." The specified dosing day is, for example, the start date of dosing, but it may also be any other specific day.

(Example of the first storage section that is effective for sorting medications brought with you)
Fig. 29 is a diagram showing an example of the first storage unit 11A. Unlike the first storage unit 11 shown in Fig. 2, the first storage unit 11A shown in Fig. 29 has an internal structure that enables the medicine sorting device 1 to efficiently sort brought-in medications. The first storage unit 11A can be effectively used mainly when sorting brought-in medications, but may also be used when sorting returned medications.

In the example of Figure 29, the first storage section 11A has four storage sections 111-114 with approximately the same volume. Each of the four storage sections 111-114 is further divided into four sections. Storage section 111 has three small storage sections 111a-111c and one medium storage section 111d. Storage section 112 has three small storage sections 112a-112c and one medium storage section 112d. Storage section 113 has three small storage sections 113a-113c and one medium storage section 113d. Storage section 114 has three small storage sections 114a-114c and one medium storage section 114d.

When storing patient-brought medications in the first storage unit 11A, the user classifies the medications into first and second medications. A first medication is a medication prescribed to a patient PA by a medical institution for administration at each of multiple administration times on a specified administration date (specified administration time), and subsequently brought to a different medical institution. A second medication is a medication prescribed to a patient PA by a medical institution, and subsequently brought to a different medical institution, the remaining medications other than the first medication. In the storage unit 111, each of the small storage units 111a to 111c functions to store the patient PA's first medication. The medium storage unit 111d functions to store the patient PA's second medication. For example, if the patient PA is prescribed to take three doses each day, the three doses taken on the first day are the first medications, and the medications taken on the second day and thereafter are the second medications.

Since the number of first brought-in medications stored in small storage sections 111a-111c is equal to or less than the number of second brought-in medications stored in medium storage section 111d, the volume of small storage sections 111a-111c is smaller than the volume of medium storage section 111d. Taking into account the number and volume of small and medium storage sections, in the example of Figure 29, small storage sections 111a-111c are arranged circumferentially in the outer region of storage section 111, and medium storage section 111d is arranged in the remaining region (inner region) of storage section 111. Storage sections 112-114 have the same function and configuration as storage section 111.

Storage units 111-114 may each store, for example, first and second medications brought by different patients PA-PD. In this case, first storage unit 11A can store up to four people's worth of medications at one time. However, in this case, the medications brought by patients PA-PD whose dosing times on a given dosing day are set to three or fewer times (e.g., at least one of morning, afternoon, and evening) will be stored in storage units 111-114, respectively. If a patient's dosing time on a given dosing day is set to four or more times, two or more of storage units 111-114 will be used as storage locations for that patient's medications. Note that for the second medication, at least one middle storage unit out of the two or more middle storage units will be used. If the dosing time is set to four or more times, first storage unit 11A can store up to three people's worth of medications at one time.

For example, if a patient's medication schedule on a given medication day is set to four to six times, two storage sections (e.g., storage sections 111 and 112) are used to store the patient's medications. For example, if the patient's medication schedule on a given medication day is set to six times, small storage sections 111a-111c and 112a-112c are used to store the patient's first medication. Meanwhile, medium storage sections 111d and/or 112d are used to store the patient's second medication.

Indicators 111e to 114e are also provided in storage sections 111 to 114, respectively. Indicators 111e to 114e function as markers for identifying storage sections 111 to 114. Indicators 111e to 114e have different colors. Specifically, indicators 111e to 114e may be areas each colored. For example, indicator 111e may be red, indicator 112e may be blue, indicator 113e may be yellow, and indicator 114e may be green, but this is not limited to this. In addition, indicators 111e to 114e do not necessarily have to have different colors as long as they are distinguishable from one another; for example, they may have different shapes or patterns.

Indicator 111e is located in two positions at the boundary between small storage sections 111a-111c and medium storage section 111d. Indicator 112e is located in two positions at the boundary between small storage sections 112a-112c and medium storage section 112d. Indicator 113e is located in two positions at the boundary between small storage sections 113a-113c and medium storage section 113d. Indicator 114e is located in two positions at the boundary between small storage sections 114a-114c and medium storage section 114d. However, the positions of indicators 111e-114e are not limited to the above example, as long as there is no inconvenience in identifying storage sections 111-114.

(Example of sorting and processing of medications brought in by the patient)
The medicine sorting device 1 of this embodiment can be set to a normal sorting mode (returned medicine sorting mode) in which returned medicines are automatically sorted, and a brought-in medicine sorting mode in which brought-in medicines are automatically sorted.

The processing and operation of the medicine sorting device 1 when set to normal sorting mode is as described above. That is, in normal sorting mode, the medicine sorting device 1 removes multiple types of medicine stored in the first storage unit 11A one by one and determines the type of medicine based on an image of the removed medicine. Based on the determination result, the medicine sorting device 1 transports the medicine to either the second storage unit 14, the waiting tray 15, or the collection tray 16. In normal sorting mode, the user may store the medicines to be sorted in any of the small and/or medium storage units.

When the user selects the medication sorting mode, the user sorts the first and second medications and stores them in the small and/or medium storage sections, as described above. The control unit 60a sorts the first medication before the second medication. The order in which medications are removed from storage sections 111-114 and the order in which medications are stored in the sorting cup 141 may be preset. For example, the order in which medications are removed from storage sections 111-114 is set to storage sections 111, 112, 113, and 114. Furthermore, the order in which medications are stored in the sorting cup 141 is set to, for example, columns A to F in the first row, columns A to F in the second row, etc., if the sorting cup 141 is arranged in rows 1 to 7 and columns A to F. For example, consider a case where patient PA has five types of medications stored in storage unit 111 and patient PB has five types of medications stored in storage unit 112. In this case, the control unit 60a stores patient PA's medications in sorting cups 141 arranged in columns A to E of the first row, and then stores patient PB's medications in column F of the first row and columns A to D of the second row. In other words, the control unit 60a stores patient PA's medications in the sorting cups 141 in a predetermined order for each patient and each type of medication.

In addition, the discrimination unit 64 in the medicine sorting device 1 of this embodiment has the function of determining reference data for determining the type of second medication brought in, for example, by one of the following two methods. The discrimination unit 64 stores the determined reference data in the memory unit 80.

(Method 1) The discrimination unit 64 determines reference data indicating information based on at least one characteristic of the first brought-over drug (e.g., the size, shape, or representative color described above) extracted from an image of the first brought-over drug. The discrimination unit 64 determines, as reference data, multiple characteristics managed as drug data for the first brought-over drug, obtained by the comparison unit 64b comparing at least one characteristic of the first brought-over drug with a drug database. In this case, the comparison unit 64b can compare the multiple characteristics of the first brought-over drug with at least one characteristic of the second brought-over drug.

(Method 2) The discrimination unit 64 determines at least one characteristic of the first brought-on medication as reference data. For example, the discrimination unit 64 determines the size, shape, and representative color of the first brought-on medication as reference data.

After sorting the first brought-by medication, the control unit 60a sorts the second brought-by medication. At this time, the control unit 60a is only required to use the reference data to sort the second brought-by medication based on the characteristics of the first brought-by medication. Therefore, even if the type of the first brought-by medication is not accurately determined or narrowed down, it is sufficient that the first brought-by medication and the second brought-by medication that are considered to be the same type are placed in the same sorting cup 141. In the above method 1, the type of the first brought-by medication can be accurately determined or narrowed down by comparing the characteristics of the first brought-by medication with the medication database, but the amount of medication data managed in the medication database is enormous, so this takes time. In the above method 2, the discrimination unit 64 determines the characteristics of the first brought-by medication as reference data, thereby shortening the processing time of the discrimination unit 64 (the time required to create the reference data).

When the same small container contains multiple first brought-in medications, the matching unit 64b compares the reference data for each first brought-in medication to determine whether the reference data can be considered identical. The matching unit 64b may consider the reference data to be identical if the degree of match between at least one feature of the two first brought-in medications being compared is equal to or greater than a predetermined value. The number of features used to determine whether the degree of match is equal to or greater than a predetermined value, and the predetermined value, may be set through experiments, etc. The matching unit 64b may also perform similar processing when comparing at least one feature of the second brought-in medication with the reference data.

The matching unit 64b also functions as a first matching unit that matches at least one feature of the second brought-over drug extracted from an image of the second brought-over drug with the reference data determined by the discrimination unit 64. The matching unit 64b does not match the features of the second brought-over drug with the drug database, at least for the second brought-over drug. Furthermore, only the number of first brought-over drugs is stored in the memory unit 80 as reference data. Therefore, the matching time can be reduced compared to matching the features of the second brought-over drug with all drug data managed in the drug database. Furthermore, by narrowing down the features of the first brought-over drug included in the reference data (e.g., the size, shape, and representative color of the first brought-over drug), the matching time can be further reduced.

Furthermore, when the discrimination unit 64 determines reference data using Method 1 above, the collation unit 64b functions as a second collation unit that narrows down the type of the first brought-on medication by comparing at least one characteristic of the first brought-on medication with the medication database. In this case, the collation unit 64b can accurately determine or narrow down the type of the first brought-on medication, and ultimately accurately determine or narrow down the type of the second brought-on medication.

Note that while the matching unit 64b is described as having the functions of both the first matching unit and the second matching unit, the control unit 60a may also have the first matching unit and the second matching unit as separate functions.

(Screen description)
FIG. 30 is a diagram showing an example of an initial image (initial screen) on which the results of drug sorting can be displayed. The display control unit 67 normally displays the initial image on the display unit 32. The initial image shown in FIG. 30 includes a button B11 for displaying a menu related to the operation of the drug sorting device 1. When the user operates button B11, the display control unit 67 displays a menu including a button B12 for displaying a submenu for switching the sorting mode of the drug sorting device 1. When the user operates button B12, the display control unit 67 displays a submenu including a button B13 for setting the sorting mode to the normal sorting mode and a button B14 for setting the sorting mode to the brought-drug sorting mode. When the user operates button B14, the control unit 60a sets the sorting mode to the brought-drug sorting mode.

Figure 31 shows an example of a setting image (setting screen) that allows the user to set the storage method for brought-medications in the first storage unit 11A when the brought-medication sorting mode is set. When the control unit 60a sets the sorting mode to the brought-medication sorting mode, the display control unit 67 changes the image displayed on the display unit 32 from the initial image to the setting image. The setting image shown in Figure 31 includes an input area PIR that accepts input of patient information. The user inputs patient information such as the patient's name, gender, age, and medical department in the input area PIR. The control unit 60a stores the input patient information in the memory unit 80.

The settings image shown in FIG. 31 further includes a button B21 for selecting a storage unit for storing brought-in medications. Button B21 includes areas corresponding to the colors of indicators 111e to 114e. By operating button B21, the user can select a storage unit having an indicator corresponding to the color of the operated area as the storage unit for storing brought-in medications.

The settings image shown in FIG. 31 further includes a button B22 for selecting another storage section for storing patient-brought medications. As described above, if a patient's medication schedule is set to four or more times on a given medication date, two or more of storage sections 111-114 will be used as storage locations for the patient's medications. By operating button B22, the user can select the storage section with an indicator corresponding to the color of the area operated on as the second storage section for storing patient-brought medications.

Button B22, like button B21, includes areas corresponding to the colors of indicators 111e-114e, as well as an area for clearing the selected storage unit. If selecting a second storage unit is not necessary, the user can simply operate the area of button B22 that clears the storage unit for medications brought with them.

In the following description, it is assumed that storage section 111 is selected with button B21 and storage section 112 is selected with button B22.

The setting image shown in FIG. 31 further includes buttons B3 corresponding to each of the small containers 111a-111c and 112a-112c. Button B3 allows the user to set the usage of the medications stored in each of the small containers 111a-111c and 112a-112c. When any of the buttons B3 is operated, the display control unit 67 displays a submenu (not shown) for selecting the usage of the medication stored in the small container corresponding to that button B3. In the submenu, the user can select usage such as "Morning," "Daytime," "Evening," "Every," "Upon waking up," "Before going to bed," "Before meals," "Between meals," "After meals," "Just before meals," and "Just after meals." If there is no appropriate usage in the submenu (for example, "When in pain"), the user can directly input the usage in text. However, the control unit 60a may not display the submenu and instead accept text input for all usage.

The setting image shown in FIG. 31 further includes image IA, which indicates storage section 111, and image IB, which indicates storage section 112. In other words, the display control unit 67 includes images IA and IB, which correspond to storage sections 111 and 112, respectively, selected by the user by operating buttons B21 and B22, in the setting image. Image IA includes image IAe, which corresponds to indicator 111e. Image IB includes image IBe, which corresponds to indicator 112e. By referring to images IAe and IBe, the user can confirm that the selected storage sections are storage sections 111 and 112.

Image IA includes images IAa to IAd corresponding to the small storage sections 111a to 111c and the medium storage section 111d of storage section 111, respectively. Image IA also includes image IAf showing the use of medications brought by the patient to be stored in each of the small storage sections 111a to 111c. Similarly, image IB includes images IBa to IBd corresponding to the small storage sections 112a to 112c and the medium storage section 112d of storage section 112, respectively. Image IB also includes image IBf showing the use of medications brought by the patient to be stored in each of the small storage sections 112a to 112c.

The display control unit 67 causes images IAf and IBf to reflect the usage of the medications stored in small storage compartments 111a-111c and 112a-112c, respectively, selected by the user using button B3. The user refers to images IAf and IBf and places the first medication into small storage compartments 111a-111c and 112a-112c. The user also places the second medication into medium storage compartments 111d and 112d.

The settings image shown in Figure 31 further includes a button B41 for adding patient information, etc. for patients who bring their own medications. By operating button B41, a tab for displaying a settings image for setting patient information, etc. for a new patient is added next to button B41. By selecting a tab, the user can register patient information for each patient and set which storage section will store the patient's own medications, and which small storage section will store the first medication with which usage.

The settings image shown in FIG. 31 further includes a button B42 for deleting patient information, etc. When sorting of a patient's medications is no longer necessary, the user operates button B42 to delete the patient information, etc. for that patient. When the control unit 60a receives a user's operation of button B42, it deletes the patient information, etc. for the corresponding patient from the memory unit 80.

The setting image shown in Figure 31 further includes a button B5 for starting the sorting of medications brought by the user. After the user has finished putting the medications brought by the user into the first storage section 11A, the user operates button B5 to cause the medication sorting device 1 to start sorting the medications brought by the user. Note that, as shown in Figure 30, button B5 is also included in the initial image.

On the other hand, when the user operates button B13 in the setting image shown in Figure 30, the control unit 60a sets the sorting mode to normal sorting mode. As described above, in normal sorting mode, first and second brought-by medications are not sorted for each patient, and brought-by medications or returned medications may be stored in any small and/or medium storage unit. In this case, the display control unit 67 does not change the image displayed on the display unit 32 from the initial image to the setting image. After the user places the medications to be sorted into the first storage unit 11, the user operates button B5 included in the initial image. The control unit 60a starts sorting when button B5 is operated.

Figure 32 shows an example of a sorting image for displaying the sorting status of brought-by medications when the brought-by medication sorting mode is set. After the control unit 60a starts sorting the brought-by medications, the display control unit 67 changes the image displayed on the display unit 32 from the initial image to a sorting image. The display control unit 67 updates the sorting image according to the progress of the control unit 60a's sorting of the brought-by medications. The sorting image shown in Figure 32 includes an image IC showing the sorting status of the brought-by medications. The image IC includes an image showing the number of brought-by medications contained in each area corresponding to the sorting cups 141 of the second storage unit 14.

The display control unit 67 corresponds the color of the area corresponding to the sorting cup 141 to the color of the indicator of the storage unit in which the patient's own medications sorted into that sorting cup 141 were stored. In the example shown in Figure 32, the display control unit 67 corresponds the colors of columns A to D in the first row to indicator 111e of storage unit 111. The display control unit 67 also corresponds the colors of columns E and F in the first row to indicator 112e of storage unit 112. The display control unit 67 also corresponds the colors of columns C to E in the second row to indicator 113e of storage unit 113. The display control unit 67 also corresponds column F in the second row and column A in the third row to indicator 114e of storage unit 114.

The image IC also includes an image indicating whether a visual inspection of the medications stored in each sorting cup 141 in the second storage section 14 has been completed. In the example shown in Figure 32, a visual inspection has not been completed for any of the sorting cups 141. Therefore, the display control unit 67 displays the text "Not visually inspected" in the area corresponding to each sorting cup 141.

When the type of medication contained in each sorting cup 141 of the second storage unit 14 can be determined at the time of sorting, the image IC includes an image indicating the type. In the example shown in FIG. 32, the display control unit 67 indicates the type with multiple "*"s below the string "not visually inspected" in columns A and C of the first row, column E of the second row, and column A of the third row. Furthermore, when the type of medication contained in each sorting cup 141 of the second storage unit 14 cannot be determined at the time of sorting, the image IC includes an image indicating that the type of medication contained in that sorting cup 141 is in a provisionally sorted state, with the type undetermined. In the example shown in FIG. 32, the display control unit 67 displays "provisionally sorted" instead of a string indicating the type in the area corresponding to the sorting cup 141 containing undetermined medications. The user can visually inspect the type of medication in the provisionally sorted state and input the type into the medication sorting device 1.

The sorting image shown in Figure 32 further includes a button B6 for starting packaging of the sorted brought-by medications. After the user has completed sorting and visually inspecting the brought-by medications, they operate button B6 to cause the medication sorting device 1 to start packaging the brought-by medications. When the user operates button B6, the print output control unit 69 controls the print output unit 4 to print a journal showing medication data related to the brought-by medications to be packaged. The packaging control unit 71 also controls the packaging mechanism 6 to package the brought-by medications sorted into each sorting cup 141.

As described above, in the medicine sorting device 1, the display control unit 67 can easily identify the storage unit into which the user should place their own medication by referring to the image displayed on the display unit 32. This reduces the possibility of placing medication in the wrong storage unit. Furthermore, by sorting and packaging medications using the medicine sorting device 1, labor costs can be reduced and economic productivity can be improved. This can contribute to achieving the Sustainable Development Goals (SDGs).

[Software implementation example]
The drug sorting device 1, the information processing device 200, and the data management device 500 each include a computer that executes instructions from a software program that realizes each function. This computer includes, for example, one or more processors and a computer-readable recording medium storing the program. The object of the present invention is achieved by the computer having the processor read and execute the program from the recording medium. The processor may be, for example, a central processing unit (CPU). The recording medium may be a "non-transitory tangible medium," such as a read-only memory (ROM), tape, disk, card, semiconductor memory, or programmable logic circuit. The device may also include a random access memory (RAM) for expanding the program. The program may be supplied to the computer via any transmission medium capable of transmitting the program (such as a communication network or broadcast waves). One aspect of the present invention may also be realized in the form of a data signal embedded in a carrier wave, in which the program is embodied by electronic transmission.

[Additional Notes]
The present invention is not limited to the above-described embodiments, and various modifications are possible within the scope of the claims. Embodiments obtained by appropriately combining the technical means disclosed in different embodiments are also included in the technical scope of the present invention.

Another aspect of the present invention can be described as follows:

A type discrimination device according to one aspect of the present invention is a type discrimination device that discriminates the type of a target drug from an image of the target drug. The type discrimination device includes a feature extraction unit that extracts, from the image, first feature information that indicates the external characteristics of the target drug, which is reflection area information that indicates the reflection area of light irradiated onto the target drug; and a discrimination unit that compares the first feature information with second feature information that indicates the external characteristics of the reference drug, which is reflection area information that indicates the reflection area of light irradiated onto a reference drug, and discriminates that the target drug is a coated tablet as a type of the target drug based on the comparison result. The reflection area is an area located inside the outer edge of the target drug in the image.

In a type determination device according to one aspect of the present invention, the determination unit may compare the shape of the reflection area indicated by the reflection area information included in the first feature information with the shape of the reflection area indicated by the reflection area information included in the second feature information.

In one aspect of the type determination device of the present invention, the reflective region is an image region in the captured image having a saturation value higher than a predetermined value, and if the target drug is a coated tablet, may be an annular region formed on the surface of the coated tablet.

1 medicine sorting device 11 first storage section 13 imaging unit (imaging section)
14 Second storage section 15 Waiting tray (temporary storage section)
60 Computer (type determination device)
62 Sorting control unit (sorting determination unit)
64a Feature extraction unit 64b Matching unit (discrimination unit)
65 Priority determination unit 73 Image processing unit 74 Drug database update unit (update unit)
82 Captured image 200 Information processing device 221 Acquisition unit (reception unit)
222 Image generation unit (generation unit)
1000 Image Generation System

Claims (2)

A type discrimination device that discriminates the type of a target drug from an image of the target drug,
a discrimination unit that compares fifth feature information indicating the characteristics of the appearance of the target drug, including target color information indicating the color of the target drug, extracted from the captured image, with sixth feature information including reference color information indicating the color of a reference drug, and discriminates the type of the target drug based on the comparison result;
and an update unit that, when the discrimination unit fails to discriminate the type of the target drug , accepts a user operation to specify the type of the target drug that has failed to be discriminated, and then updates the reference color information of the reference drug corresponding to the type of the target drug specified by the user using the target color information of the target drug that has failed to be discriminated. A feature extraction unit extracts third feature information indicating an appearance feature of the target drug from the captured image, the third feature information including at least reflection information indicating whether light irradiated onto the target drug is reflected by the target drug;
The type determination device of claim 1, wherein the determination unit compares the third characteristic information with fourth characteristic information indicating the appearance characteristics of the reference drug, which includes at least reflection information indicating whether light irradiated onto the reference drug is reflected by the reference drug, and determines the type of the target drug based on the result of the comparison.