JP7447482B2 - Image processing device, system, method, and program - Google Patents

Description

The present disclosure relates to an image processing device, a system, a method, and a program, and in particular, it is possible to determine the format of a document form even if the document form has few character strings that are not to be recorded or the document form does not have an emblem image. The present invention relates to an image processing device, system, method, and program.

2. Description of the Related Art There is an image processing apparatus that performs OCR (Optical Character Recognition) processing on image data of a document form obtained using a scanner or the like to read a document. The image processing device disclosed in Patent Document 1 records the results of OCR processing, and generates learning data for each document form format from the recorded OCR processing results. The image processing device of Patent Document 1 uses the degree of matching of character strings that are not to be recorded (non-recorded character strings), the degree of matching of emblem images, the degree of matching of coordinate ranges of character strings that are not to be recorded, etc. , determines (determines) the format of the new document form. However, there is a problem in that it is difficult to distinguish the format of document forms that do not have or have few character strings that are not to be recorded, or document forms that do not have an emblem image (table form forms).

JP 2019-8775 Publication

As mentioned above, there is a problem in that it is difficult to distinguish the formats of document forms that do not have or have few character strings that are not to be recorded, or document forms that do not have an emblem image.

An object of the present disclosure is to provide an image processing device, system, method, and program that solve the above-mentioned problems.

The image processing device according to the present disclosure includes:
a feature quantity extraction generation unit that extracts an individual first feature quantity indicating a feature of a recorded character string included in a document form, and an individual second feature quantity indicating recognition information of a non-recorded character string included in the document form; ,
a group classification unit that groups the document form into a group first feature including a plurality of the individual first features and a group second feature including the plurality of the individual second features;
If the individual second feature of the new document form does not match any of the group second features recorded in advance, the new document form's individual second feature is selected from among the group first features recorded in advance. a control unit that identifies a specific first feature amount that matches the individual first feature amount, and sets a format indicated by the specific first feature amount as a format of the new document form;
Equipped with

The system according to the present disclosure is
Equipped with an image processing device, an image reading device, a recording device, and a database,
The image reading device includes:
a reading output unit that optically acquires image data of a document form and outputs the image data to the image processing device;
The image processing device includes:
The image data of the document form is character-recognized, and an individual first feature amount indicating a feature of a recorded character string included in the document form and an individual second feature amount indicating recognition information of a non-recorded character string included in the document form are provided. a feature amount extraction unit that extracts the feature amount;
a processing output unit that outputs the character recognition result of the character recognition to the recording device;
a group classification unit that groups the document form into a group first feature amount including a plurality of the individual first feature amounts and a group second feature amount including a plurality of the individual second feature amounts;
If the individual second feature of the new document form does not match any of the pre-recorded group second features, the new document form is selected from among the pre-recorded group first features. a control unit that specifies a specific first feature amount that matches the individual first feature amount, and sets a format indicated by the specific first feature amount as a format of the new document form;
The recording device includes:
a character recognition acquisition unit that acquires the character recognition result from the image processing device;
a character recognition recording unit that records the character recognition results in the database;
The database is
It has a recording table that records the character recognition results obtained from the recording device.

The method according to the present disclosure includes:
extracting an individual first feature amount indicating a feature of a recorded character string included in a document form, and an individual second feature amount indicating recognition information of a non-recorded character string included in the document form;
Grouping the document form into a group first feature amount including a plurality of the individual first feature amounts and a group second feature amount including a plurality of the individual second feature amounts;
If the individual second feature of the new document form does not match any of the group second features recorded in advance, the new document form's individual second feature is selected from among the group first features recorded in advance. identifying a specific first feature amount that matches the individual first feature amount, and setting a format indicated by the specific first feature amount as a format of the new document form;
Equipped with

The program related to this disclosure is
extracting an individual first feature amount indicating a feature of a recorded character string included in a document form, and an individual second feature amount indicating recognition information of a non-recorded character string included in the document form;
Grouping the document form into a group first feature amount including a plurality of the individual first feature amounts and a group second feature amount including a plurality of the individual second feature amounts;
If the individual second feature of the new document form does not match any of the group second features recorded in advance, the new document form's individual second feature is selected from among the group first features recorded in advance. identifying a specific first feature amount that matches the individual first feature amount, and setting a format indicated by the specific first feature amount as a format of the new document form;
have the computer execute it.

According to the present disclosure, there is provided an image processing device, a system, a method, and a program capable of determining the format of a document form even if the document form has few character strings that are not to be recorded or the document form does not have an emblem image. can be provided.

FIG. 1 is a block diagram illustrating a system according to an embodiment. FIG. 1 is a block diagram illustrating an image processing device according to an embodiment. FIG. 1 is a block diagram illustrating an image processing device according to an embodiment. FIG. 2 is a schematic diagram illustrating a document form. FIG. 2 is a diagram illustrating an overview of a record table recorded by a database. 3 is a flowchart illustrating the operation of the image processing apparatus according to the embodiment. 3 is a flowchart illustrating the operation of the image processing apparatus according to the embodiment. 3 is a flowchart illustrating the operation of the image processing apparatus according to the embodiment.

Embodiments of the present invention will be described below with reference to the drawings. In each drawing, the same or corresponding elements are denoted by the same reference numerals, and for clarity of explanation, redundant explanation will be omitted as necessary.

[Embodiment]
<System>
A system according to an embodiment will be described.
FIG. 1 is a block diagram illustrating a system according to an embodiment.

As shown in FIG. 1, the system 10 includes an image processing device 1, an image reading device 2, a recording device 3, and a database 4. The image processing device 1 is connected to the image reading device 2 by a communication cable. The image processing device 1 is connected to the recording device 3. The image reading device 2 is connected to the recording device 3. The database 4 is connected to the image processing device 1 and the recording device 3.

The image reading device 2 has a reading output section (not shown) that optically acquires image data such as a document form and outputs it to the image processing device 1.

The image processing device 1 includes a feature amount extraction generation unit that performs OCR processing on image data of a document form to recognize characters, and a processing output unit (not shown) that outputs the character recognition result of character recognition to the recording device 3. . Details of the image processing device 1 will be described later.

The recording device 3 includes a character recognition acquisition section (not shown) and a character recognition recording section (not shown). The character recognition acquisition unit acquires character recognition results from the image processing device 1 . The character recognition recording unit records the acquired character recognition results in the database 4.

Database 4 has a record table (not shown). The recording table records the character recognition results obtained from the recording device 3. Specifically, the database 4 stores the correspondence between image data of a plurality of document forms registered in the past from the recording device 3 and recorded character strings indicating character strings to be recorded among character strings included in the image data. Record relationships. The character string indicated by the recorded character string is an important character string that should be recorded and saved in the database 4 among the character strings written in the document form. That is, the recorded character string indicates a character string to be recorded. Note that the non-recorded character string refers to a character string that is not to be recorded, such as a tabular form form.

An operator using the system 10 uses the recording device 3 to store in advance image data of a plurality of document forms registered in the past and recorded character strings among the character strings included in the image data in the database 4. Register in the record table. The recording device 3 stores the correspondence relationship between the image data of a document form and the recorded character string information indicating the character string to be recorded among the character string information included in the image data, for a plurality of document forms. shall be adequately recorded. The recording table will be described later.

<Image processing device>
An image processing device according to an embodiment will be described.
FIG. 2 is a block diagram illustrating an image processing apparatus according to an embodiment.
FIG. 2 is a block diagram illustrating the minimum configuration of the image processing device according to the embodiment.
FIG. 3 is a block diagram illustrating an image processing apparatus according to an embodiment.

As shown in FIG. 2, the image processing device 1 according to the embodiment includes a control section 101, a feature extraction generation section 100, a reading target feature generation section 104, and a group classification section 106. The feature extraction generation unit 100 includes a feature extraction unit 103 and a reading target feature generation unit 104. Further, as shown in FIG. 3, the image processing apparatus 1 according to the embodiment further includes an acquisition section 102, a recording section 105, and a group identification section 107.

The control unit 101 controls the acquisition unit 102 , the feature extraction unit 103 , the reading target feature generation unit 104 , the recording unit 105 , the group classification unit 106 , and the group identification unit 107 .

The acquisition unit 102 acquires image data of a document form and recorded character strings included in the document form from the database 4. The acquisition unit 102 also acquires image data of a new document form from the image reading device 2.

The feature amount extraction unit 103 extracts feature amounts indicating characteristics of character strings included in a document form. The reading target feature generating unit 104 generates an individual first feature indicating the feature of the recorded character string included in the document form and a feature of the non-recorded character string based on the feature extracted by the feature extracting unit 103. Generate an individual second feature quantity.

That is, the feature amount extraction generation unit 100 performs OCR processing on the image data of the document form, and extracts the characteristics of the recorded character string included in the document form based on the character recognition processing result (OCR processing result) of the image data of the document form. Extract the individual first feature shown. The feature quantity extraction generation unit 100 extracts an individual second feature quantity indicating recognition information of a non-recorded character string included in a document form.

Further, the feature quantity extraction generation unit 100 extracts an individual first feature quantity and an individual second feature quantity corresponding to image data of a plurality of document forms for each document form and for each character string. A specific method for extracting the individual first feature amount and the individual second feature amount by the feature amount extraction generation unit 100 will be described later.

The group classification unit 106 groups the document form into a group first feature amount that includes a plurality of individual first feature amounts and a group second feature amount that includes a plurality of individual second feature amounts.

The control unit 101 checks whether the individual second feature of the new document form acquired by the acquisition unit 102 matches any of the plurality of individual second features in the group second feature recorded in advance. . As a result of the confirmation, if there is no match, the control unit 101 selects a specific first feature that matches the individual first feature of the new document form from a plurality of individual first features in the group first feature recorded in advance. Identify the first feature. The control unit 101 sets the format indicated by the identified specific first feature amount as the format of a new document form.

The group specifying unit 107 specifies a group that includes the individual first feature amount of the new document form. The group identifying unit 107 identifies a group that includes the individual second feature amount of the new document form.

The recording unit 105 records the document form and the individual first feature amount of the document form in association with each other, and records the document form and the individual second feature amount of the document form in association with each other. Similarly, for the new document form, the recording unit 105 records the new document form and its individual first feature amount in association with each other, and records the new document form and its individual second feature amount in association with each other. do.

If the individual second feature of a new document form does not match any of the pre-recorded group second features, the image processing device 1 according to the embodiment selects the pre-recorded group first feature. , a specific first feature amount that matches the individual first feature amount of the new document form is specified, and the format indicated by the specific first feature amount is set as the format of the new document form.

As a result, according to the embodiment, an image processing device, system, and method are capable of determining the format of a document form even if the document form has only a few character strings that are not to be recorded or the document form does not have an emblem image. , and programs can be provided.

<Document forms and database>
The relationship between document forms and databases will be explained.
FIG. 4 is a schematic diagram illustrating a document form.
FIG. 5 is a diagram illustrating an outline of a record table recorded by the database.

As shown in FIG. 4, the document form 5 includes the mark of the company that created the document, the date of creation, the person in charge of creation, and the content of the document in a format specific to the document form 5. For example, if the document form 5 is an order form, the document content written in the document form 5 is a set of information such as the name of one or more ordered products and the number of items ordered. The recorded character string (character string to be recorded) is, for example, a date 51, a supplier 52, a product name 53, a quantity 54, and an amount 55 shown in FIG. Non-recorded character strings (character strings that are not to be recorded) include, for example, the name 501 of the orderer, the emblem image 502 of the orderer, the title 503 of the document form, and the greeting 504 shown in FIG.

As shown in FIG. 5, in the record table of the database 4, the correspondence relationship (combination) between image data for a plurality of different document forms and specific record character strings written in each document form is stored in the record table of the database 4. A large number of data are recorded in advance. Specifically, in the record table of the database 4, specific recorded character string information for each of a plurality of different document forms is recorded for a plurality of sheets. In the record table of the database 4, a form ID for identifying a document form, image data of the document form, and a specific record character string to be recorded among the character strings written in the document form are associated with each other. recorded. There may be a plurality of recorded character strings corresponding to one document form. In this way, in the recording table, a form ID corresponding to a document form, image data corresponding to a document form, and one or more record character strings corresponding to a document form are recorded for each document form. There is.

<Operation of image processing device>
The operation of the image processing device according to the embodiment will be described.
FIG. 6 is a flowchart illustrating the operation of the image processing apparatus according to the embodiment.
FIG. 7A is a flowchart illustrating the operation of the image processing apparatus according to the embodiment.
FIG. 7B is a flowchart illustrating the operation of the image processing apparatus according to the embodiment.

The operator starts up the image processing device 1, with image data and recorded character strings etc. recorded in advance for each document form (see FIG. 4) in the database 4 (see FIG. 5), and transfers data to the image processing device 1. Instructs to start processing.

When the operator instructs to start processing, as shown in FIG. 6, the acquisition unit 102 of the image processing device 1 obtains the image data of the document form 5 from the database 4, information on the recorded character string corresponding to the image data, and It is determined whether all have been read (step S101).

If NO in step S101, the acquisition unit 102 reads the image data of the document form 5 and the recorded character string information corresponding to the image data from the database 4 (step S102). The acquisition unit 102 outputs the image data and the recorded character string to the feature amount extraction unit 103.

The feature extraction unit 103 performs OCR processing on the image data to detect all character strings in the image data and coordinates in the image data indicating the range of the character strings (step S103). Note that a character string is a group of characters composed of a plurality of characters. The feature extracting unit 103 analyzes the range of one group based on the spacing with other characters, extracts one or more characters included in the range as a character string, and extracts the characters in the image data. Detects coordinates that indicate the range of a string. The characters included in the character string may include symbols such as ideograms and phonetic characters, marks, icon images, and the like.

The feature extraction unit 103 compares the character string extracted from the image data by OCR processing and the recorded character string read from the database 4 together with the image data. The feature extraction unit 103 extracts character strings in the image data that match the character information of the recorded character strings, the attributes of the characters included in the character strings, and the ranges thereof, among the character strings extracted from the image data by OCR processing. (step S104).

Character attributes are information expressed by numbers, alphabets, hiragana, kanji, number of characters, character height, font, etc. Further, the coordinates of the range of a character string are information indicating the coordinates of the first character, the coordinates of the end character, etc. included in the character string. The attributes of a character may be referred to as attribute information of a recorded character string, and the coordinates of the range of a character string may be referred to as coordinate information of a recorded character string.

The feature amount extraction unit 103 generates a feature amount including the identified information for one document form 5 (step S105). The feature amount is generated for each recorded character string in the document form 5 by the feature amount extraction unit 103. That is, the feature amount extraction unit 103 generates a feature amount for each document form and each recorded character string. This feature amount for each document form and each recorded character string is referred to as an individual first feature amount. The individual first feature amount may include one or both of character attributes and coordinates indicating a range of character strings.

The feature amount extraction unit 103 associates the individual first feature amounts of one or more recorded character strings in each document form 5 with the identifier (form ID) of the document form 5 and the identifier of the recorded character string, and stores them in a database. 4 (step S106).

As the identifier of the recorded character string, for example, coordinate values indicating the position of the recorded character string can be used. The feature extraction unit 103 indicates the character attributes and character string ranges of the date 51, supplier 52, product name 53, quantity 54, and amount 55, which are recorded character strings included in the document form 5 shown in FIG. Each individual first feature indicating coordinates etc. is recorded in the database 4 in association with the identifier of the document form 5 and the identifier of the recorded character string.

In addition, the feature extraction unit 103 extracts non-recorded character strings in the image data that do not match the character information included in the recorded character strings, the attributes of the characters included in the non-recorded character strings, and the coordinates of the range. Specify (step S107).

The feature amount extraction unit 103 generates a feature amount including the identified information for one document form 5 (step S108). The feature amount is generated for each non-recorded character string in the document form 5 by the feature amount extraction unit 103. That is, the feature extraction unit 103 generates a feature for each document form and for each non-recorded character string. This feature quantity for each document form and for each non-recorded character string is referred to as an individual second feature quantity. The individual second feature amount may include either or both of character attributes and coordinates indicating a range of character strings.

The feature amount extraction unit 103 associates the individual second feature amounts of one or more non-recorded character strings in each document form 5 with the identifier of the document form 5 and the identifier of the non-recorded character string, and stores them in the database 4. Record (step S109).

As the identifier of the non-recorded character string, for example, coordinate values indicating the position of the non-recorded character string can be used. The feature amount extraction unit 103 extracts each individual character string that is included in the document form 5 shown in FIG. The second feature quantity is recorded in the database 4 in association with the identifier of the document form 5 and the identifier of the non-recorded character string.

The database 4 records image data of a plurality of different document forms 5 and information on recorded character strings corresponding to the image data. The acquisition unit 102 of the image processing device 1 repeats the processing from step S101 to step S109 until the image data and recorded character string information for all document forms 5 are read.

In step S101, if the acquisition unit 102 determines that all of the image data and recorded character string information regarding the document form 5 have been read (step S101: YES), the group classification unit 106 determines whether the information contained in the image data of the document form 5 includes The document form 5 is divided into groups based on the individual second feature amount and the individual first feature amount (step S112).

For example, the group classification unit 106 classifies each document form 5 based on the degree of matching of non-recorded character strings indicated by the individual second feature amount, the degree of matching of emblem images, the degree of matching of coordinate ranges of non-recorded character strings, etc. Divide into groups. Furthermore, the group classification unit 106 groups the recorded character strings based on the similarity of the characteristics (character attributes, number of characters, character height, font) of the recorded character strings indicated by the individual first feature amounts. In this way, the group classification unit 106 groups the document form 5 based not only on the individual second feature amount but also on the individual first feature amount. The group classification unit 106 determines the group identifier of the document form 5 in this grouping process.

The group classification unit 106 determines whether grouping has been completed for all document forms 5 (step S113). If grouping of all document forms 5 is not completed (step S113: NO), the group classification unit 106 repeats the process of step S112.

When the grouping of all the document forms 5 is completed (step S113: YES), the group classification unit 106 associates the identifier of the document form 5 with the group identifier assigned to the document form 5 and stores it in the database. is recorded in the recording table (step S114).

The reading target feature generating unit 104 reads each individual first feature and each individual second feature of one or more document forms 5 belonging to a certain group from the database 4 . The reading target feature amount generation unit 104 generates each group first feature amount and each group second feature amount corresponding to each individual first feature amount and each individual second feature amount of the document forms 5 belonging to the group (step S115).

The reading target feature amount generation unit 104 calculates a first feature amount for each group and a second feature amount for each group for each group, and records them in the database 4 in association with the group identifier (step S116).

Next, as shown in FIG. 7A, the operator performs an operation to cause the image reading device 2 to read a new document form. Thereby, the image reading device 2 generates image data of a new document form and outputs it to the image processing device 1. The acquisition unit 102 of the image processing device 1 acquires image data of a new document form from the image reading device 2 (step S201).

The acquisition unit 102 outputs the image data of the new document form to the feature amount extraction unit 103. The feature amount extraction unit 103 performs OCR processing on the image data of the new document form to detect a character string, the characteristics of the characters included in the character string, and the coordinates in the image data of the range of the character string. (Step S202).

The feature extraction unit 103 generates a third feature including the detected information for each character string in the image data of the new document form (step S203). Information indicating the characteristics of the character string included in the document form of the newly read image data is referred to as a third feature amount.

The group identifying unit 107 reads, from the database 4, a group second feature used to identify a new group of document forms from among a plurality of group second features (step S2031). The group second feature amount may be, for example, an individual second feature amount corresponding to the orderer's emblem image 502 (see FIG. 4) displayed in the image data of the document form.

The group identifying unit 107 determines whether the information indicated by the group second feature amount can be identified from the image data of the new document form acquired in step S201 (step S2032). The group specifying unit 107 performs similar processing using the group second feature amounts for all groups.

If information matching the group second feature read from the database 4 can be identified from the image data of the newly read document form (step S2032: YES), the group identifying unit 107 selects the group second feature. This group is identified as the group of image data of the newly read document form (step S2033). The format indicated by the second feature amount of the identified group is set as the format of the new document form.

However, for the group second feature indicating a group of second features created from information on non-recorded character strings, if there are no or few non-recorded character strings such as in a tabular format form, the group cannot be correctly grouped. cannot be determined.

Therefore, in the image processing device 1 according to the embodiment, when matching information cannot be identified from the newly read document form for the group second feature amounts for all groups (step S2032: NO), the image processing device 1 Groups are identified using one feature.

Specifically, similarly to the second feature amount, it is determined whether the information indicated by the first feature amount regarding all recorded character strings in a certain group can be specified from the image data of the new document form acquired in step S201. (Step S2034).

If the group first feature amount and third feature amount match for all types of recorded character strings (step S2034: YES), it is determined that they are the same group. Then, the group having the first group feature amount is identified as the group of image data of the newly read document form (step S204).

As a matching condition, the value of the y-coordinate (vertical direction) of the group first feature quantity is not included. That is, as a matching condition, only the x-coordinate (horizontal direction) value is compared without comparing the y-coordinate (vertical direction) value. Thereby, even if the length of the table changes in the y direction, such as in a tabular document form, determination can be made without being affected by the length of the table.

The determination method will be explained using FIG. 4. Group first feature amounts that match all of the recorded character strings from date 51 to amount 55 are determined to be the same group. If there are multiple candidates like the product name 53, it is sufficient that there is at least one matching candidate.

Furthermore, if there are a plurality of group first feature amounts that match the image data of the document form, the group that has the most matching character strings is determined to be the same group.

That is, if the individual second feature of a new document form does not match any of the group second features recorded in advance, the image processing device 1 selects a new feature from among the group first features recorded in advance. The first feature amount that has the largest number of recorded character strings that match the individual first feature amount of the document form may be specified as the specific first feature amount.

After step S204, as shown in FIG. 7B, the group identifying unit 107 reads one or more group first feature amounts for the group from the database 4 (step S205). The group specifying unit 107 outputs the third feature amount and one or more group first feature amounts to the recording unit 105. The group first feature amount is a feature amount for specifying one or more recorded character strings in the document form belonging to the group.

The recording unit 105 acquires a third feature amount and one or more group first feature amounts for one or more character strings in the image data. The recording unit 105 uses the coordinates indicating the range of the character string included in each group first feature to determine whether all third features having coordinates corresponding to the coordinates indicated by each group first feature exist. Determination is made (step S206).

If all the third feature amounts having coordinates corresponding to the coordinates of the first feature amount of each group are present, there are descriptions of characters in all entries in the document form corresponding to the recorded character string. On the other hand, if all the third feature amounts having coordinates corresponding to the coordinates of the first feature amount of each group do not exist, there is no character entry in any entry in the document form.

If YES in step S206, the recording unit 105 determines whether the character attribute included in the group first feature amount and the character attribute included in the corresponding third feature amount specified based on the coordinates match each other. (Step S207).

If the determination result in step S207 is YES and the character attributes match, the recording unit 105 creates a rectangular frame in the range of the recorded character string based on the coordinates indicated by one or more third feature amounts in the image data currently being processed. Generates a confirmation screen that displays . The recording unit 105 outputs the confirmation screen to the monitor (step S208).

The operator can check the rectangular area displayed on this confirmation screen and check the recording character string that the image processing device 1 is going to record. This allows the operator to check whether there is any shortage of recorded character strings. An icon image of either an OK or NG button is displayed on the confirmation screen. By selecting the OK button from among the icon images of the buttons, the operator can indicate that there is no shortage of selections for the recorded character strings.

On the other hand, by selecting an NG button from among the button icon images, the operator can indicate that the selection as a recorded character string is insufficient. The recording unit 105 determines whether the selection of recording character strings is sufficient in response to the button icon image pressed by the operator (step S209).

If there is no shortage, the recording unit 105 records the character string included in the third feature amount in the recording table in association with the identification information of the document form (step S210).

In the case of NO in step S206, in the case of NO in step S207, in the case of NO in step S209, the recording unit 105 stores the coordinates of the first feature of the group for which the third feature of the corresponding coordinate in the image data does not exist. Input image data of a form image with input fields provided in the range is generated and output to the monitor (step S211).

While looking at this input image data, the operator operates an input device such as a keyboard of the image processing device 1 to input a recorded character string into an input field within the input image data displayed on the monitor. A save button (record button) is displayed on the input image data, and when the save button is pressed, the recording unit 105 records a new input image in addition to the third feature amount already acquired for the document form. A third feature amount including the character string input in the data input field is generated (step S212). The recording unit 105 records the identifier of the form image data and the character string input in the input field in the database 4 in association with each other.

The image processing device 1 updates the group first feature amount and the group second feature amount and expands the range of recordable character strings by executing the processing flow shown in FIG. 6 again. As a result, the next time the same document form is processed, the character string can be automatically recorded, saving the operator the trouble of inputting the character string. The recording unit 105 records the character strings included in each of all the third feature amounts in the document form recording table (step S213).

As shown in steps S2031 to S204, the image processing device 1 adds determination using the group first feature amount to the group classification unit 106, thereby determining whether there is no second feature amount (non-recorded character string). Alternatively, even in the case of a document form with a small number of tabular forms, it is possible to correctly determine the format of the document form.

In addition, the image processing device 1 uses the image data and recorded character strings of a plurality of different document forms recorded by the operator in advance, and uses the image data of the document form, regardless of the type of the newly input document form. Record strings can be recorded. Thereby, the image processing apparatus 1 can reduce the labor of the operator in recording the recorded character string in the document form.

<Features>
Features of the embodiment will be described below.
When specifying (discriminating) the format of a document form, the image processing device 1 uses the attributes of recorded character strings (numbers, alphabets, hiragana, Group individual first feature quantity generated for each document form based on the coordinates of the recorded character string (coordinates of the first character included in the character string, coordinates of the end character, etc.) By using , the format of the document form is determined.

Although the present invention has been described as a hardware configuration in the above embodiment, the present invention is not limited to this. The present invention can also be realized by causing a CPU (Central Processing Unit) to execute a computer program to execute the processing of each component.

In the embodiments described above, the program can be stored and delivered to the computer using various types of non-transitory computer readable media. Non-transitory computer-readable media includes various types of tangible storage media. Examples of non-transitory computer-readable media include magnetic recording media (specifically, flexible disks, magnetic tapes, hard disk drives), magneto-optical recording media (specifically, magneto-optical disks), and CD-ROMs (Read Only Memory ), CD-R, CD-R/W, semiconductor memory (specifically, mask ROM, PROM (Programmable ROM), EPROM (Erasable PROM)), flash ROM, and RAM (Random Access Memory). The program may also be provided to the computer on various types of transitory computer readable media. Examples of transitory computer-readable media include electrical signals, optical signals, and electromagnetic waves. The temporary computer-readable medium can provide the program to the computer via wired communication channels, such as electrical wires and fiber optics, or wireless communication channels.

Note that the present invention is not limited to the above embodiments, and can be modified as appropriate without departing from the spirit.

1... Image processing device 2... Image reading device 3... Recording device 4... Database 5... Document form 10... System 100... Feature extraction generation section 101... Control section 102... Acquisition section 103... Feature amount extraction section 104... Feature to be read Quantity generation section 105...Recording section 106...Group classification section 107...Group identification section

Claims (9)

a feature quantity extraction generation unit that extracts an individual first feature quantity indicating a feature of a recorded character string included in a document form, and an individual second feature quantity indicating recognition information of a non-recorded character string included in the document form; ,
The document form is divided into groups based on the individual first feature amount and the individual second feature amount, and a group first feature amount includes a plurality of the individual first feature amounts, and a plurality of the individual second feature amounts. a group second feature quantity including; a group classification unit that generates ;
If the individual second feature of the new document form does not match any of the group second features recorded in advance, the new document form's individual second feature is selected from among the group first features recorded in advance. a control unit that identifies a specific first feature amount that matches the individual first feature amount, and sets a format indicated by the specific first feature amount as a format of the new document form;
Equipped with
When the individual second feature of the new document form does not match any of the pre-recorded group second feature, the control unit may match the individual second feature of the new document form to any of the pre-recorded group first feature. from, specifying the first feature amount having the largest number of recorded character strings that match the individual first feature amount of the new document form as the specific first feature amount;
Image processing device. The group classification unit classifies the document form according to at least one of the degree of correspondence of the non-recorded character string, the degree of correspondence of the emblem image, and the degree of correspondence of the coordinate range of the non-recorded character string indicated by the individual second feature amount. , and performing the grouping based on the similarity of the features of the recorded character strings indicated by the individual first feature amounts.
The image processing device according to claim 1. further comprising a group identifying unit that identifies the group in which the individual first feature of the new document form is included, and identifies the group in which the individual second feature of the new document form is included;
The image processing device according to claim 1 or 2. a storage unit that records the document form and the individual first feature amount of the document form in association with each other, and records the document form and the individual second feature amount of the document form in association with each other; Further prepare,
The image processing device according to any one of claims 1 to 3. further comprising an acquisition unit that acquires the document form and the recorded character string included in the document form from a database, and acquires the new document form from an image reading device;
The image processing device according to any one of claims 1 to 4. The individual first feature amount of the document form includes at least one of attribute information of the recorded character string and coordinate information of the recorded character string,
The attribute information of the recorded character string includes at least one of numbers, alphabets, hiragana, kanji, number of characters, character height, and font information,
The coordinate information of the recorded character string includes at least one of the coordinates of the first character and the coordinate of the end character of the recorded character string.
The image processing device according to any one of claims 1 to 5. Equipped with an image processing device, an image reading device, a recording device, and a database,
The image reading device includes:
a reading output unit that optically acquires image data of a document form and outputs the image data to the image processing device;
The image processing device includes:
The image data of the document form is character-recognized, and an individual first feature amount indicating a feature of a recorded character string included in the document form and an individual second feature amount indicating recognition information of a non-recorded character string included in the document form are provided. a feature amount extraction unit that extracts the feature amount;
a processing output unit that outputs the character recognition result of the character recognition to the recording device;
The document form is divided into groups based on the individual first feature amount and the individual second feature amount, and a group first feature amount includes a plurality of the individual first feature amounts, and a plurality of the individual second feature amounts. a group second feature quantity including; a group classification unit that generates ;
If the individual second feature of the new document form does not match any of the pre-recorded group second features, the new document form is selected from among the pre-recorded group first features. a control unit that specifies a specific first feature amount that matches the individual first feature amount, and sets a format indicated by the specific first feature amount as a format of the new document form;
If the individual second feature of the new document form does not match any of the pre-recorded group second feature, the control unit may match the individual second feature of the new document form to any of the pre-recorded group first feature. from, specifying the first feature amount having the largest number of recorded character strings that match the individual first feature amount of the new document form as the specific first feature amount;
The recording device includes:
a character recognition acquisition unit that acquires the character recognition result from the image processing device;
a character recognition recording unit that records the character recognition results in the database;
The database includes:
comprising a recording table for recording the character recognition results obtained from the recording device;
system. extracting an individual first feature amount indicating a feature of a recorded character string included in a document form, and an individual second feature amount indicating recognition information of a non-recorded character string included in the document form;
The document form is divided into groups based on the individual first feature amount and the individual second feature amount, and a group first feature amount includes a plurality of the individual first feature amounts, and a plurality of the individual second feature amounts. and generating a group second feature including;
If the individual second feature of the new document form does not match any of the group second features recorded in advance, the new document form's individual second feature is selected from among the group first features recorded in advance. identifying a specific first feature amount that matches the individual first feature amount, and setting a format indicated by the specific first feature amount as a format of the new document form;
If the individual second feature of the new document form does not match any of the group second features recorded in advance, the new specifying a first feature amount having the largest number of recorded character strings that match the individual first feature amount of the document form as the specific first feature amount;
How to prepare. extracting an individual first feature amount indicating a feature of a recorded character string included in a document form, and an individual second feature amount indicating recognition information of a non-recorded character string included in the document form;
The document form is divided into groups based on the individual first feature amount and the individual second feature amount, and a group first feature amount includes a plurality of the individual first feature amounts, and a plurality of the individual second feature amounts. and generating a group second feature including;
If the individual second feature of the new document form does not match any of the group second features recorded in advance, the new document form's individual second feature is selected from among the group first features recorded in advance. identifying a specific first feature amount that matches the individual first feature amount, and setting a format indicated by the specific first feature amount as a format of the new document form;
If the individual second feature of the new document form does not match any of the group second features recorded in advance, the new specifying a first feature amount having the largest number of recorded character strings that match the individual first feature amount of the document form as the specific first feature amount;
A program that causes a computer to execute.

Images