JP7620585B2 - Program, display position determining method, and automated transaction device - Google Patents

Description

The present invention relates to a program, a display position determination method, and an automated transaction device.

Conventionally, automated transaction devices such as ATMs (Automatic Teller Machines) are equipped with means for reading media such as slips (see, for example, Patent Documents 1 to 4).

In an automated transaction device equipped with an OCR (Optical Character Recognition) unit as a means for reading media, for example, a camera captures an image of a medium placed in a predetermined position and reads the medium. In this case, the OCR unit detects the area of the medium shown in the image by image processing of the image captured by the camera, identifies the area of the characters to be read in that area, and recognizes the characters to be read in that area. Here, the detection of the area of the medium is performed by detecting the outline of the medium. The detection of the outline of the medium is performed based on the contrast between the medium and the background shown in the image, and on the predefined shape and size of the medium. The identification of the area of the characters to be read in the detected area of the medium is performed based on the predefined layout (format) of the medium.

In an automated transaction device where media is placed on the display surface of a display unit that displays various transaction screens and is read, a polarizing filter is provided on the display surface of the display unit to prevent the light of the screen displayed on the display unit from entering the camera. This makes it possible to realize an image with a black background when the media placed on the display surface of the display unit is imaged by the camera, ensuring contrast with the media on a white background.

JP 2020-27352 A JP 2019-105991 A JP 2012-9080 A JP 2007-164367 A

Depending on the installation environment of the automated transaction device, external light from ceiling lights, etc. may be reflected as white on the display surface of the display unit. In such a case, when a medium is placed on the display surface of the display unit and read, the reflection of that external light will also appear in the image captured by the camera. For example, if the reflection of the external light overlaps with the outline of the medium shown in the image, it is not possible to ensure contrast between the medium shown in the image and the background, and the outline of the medium cannot be detected. As a result, the medium cannot be read.

In view of the above-mentioned circumstances, the present invention aims to provide a program, a display position determination method, and an automated transaction device that, when a medium is placed on the display surface of a display unit and is being read, enables the medium to be read even when external light from a ceiling light or the like is reflected in white on the display surface.

One aspect of the program is a program that has a processor execute a process to image a display unit that displays a medium placement area that specifies an area for placing the medium to be read, using an imaging unit used to read the medium to be read, detect the contour of an external light reflection area shown in the image captured by the imaging unit, and determine the position of the medium placement area to be displayed on the display unit based on the position of the contour of the external light reflection area in the image , wherein the direction of the vertical component of the imaging direction of the imaging unit is the same as the direction of the vertical component of the incident direction of the reflected external light .

One aspect of the method is a display position determination method in which an automatic transaction device performs the following steps: captures an image of a display unit displaying a medium placement area that specifies the area in which the medium to be read is to be placed, using an imaging unit used to read the medium to be read, detects the contour of an external light reflection area shown in the image captured by the imaging unit, and determines the position of the medium placement area to be displayed on the display unit based on the position of the contour of the external light reflection area in the image; and the direction of the vertical component of the imaging direction of the imaging unit is the same as the direction of the vertical component of the incident direction of the reflected external light .

One aspect of the device is an automatic transaction device comprising a display unit that displays a medium placement area that specifies the area in which the medium to be read is placed, an imaging unit used to read the medium to be read, a detection unit that detects the contour of an external light reflection area shown in an image of the display unit captured by the imaging unit, and a determination unit that determines the position of the medium placement area to be displayed on the display unit based on the position of the contour of the external light reflection area in the image, wherein the direction of the vertical component of the imaging direction of the imaging unit is the same as the direction of the vertical component of the incident direction of the reflected external light .

According to the present invention, when a medium is placed on the display surface of the display unit and is read, the medium can be read even if external light from a ceiling light or the like is reflected white on the display surface.

FIG. 2 is a diagram illustrating a functional configuration of an automatic transaction device according to an embodiment. 1 is a perspective view illustrating an example of an external configuration of an automatic transaction apparatus according to an embodiment; 4 is a diagram illustrating an example of an imaging range of an OCR camera unit; FIG. 2 is a diagram illustrating a hardware configuration of an automatic transaction device according to an embodiment. 11 is a flowchart illustrating the flow of a display position determination process. FIG. 11 is a diagram illustrating parameters used in the determination process. 11 is a flowchart illustrating the flow of a determination process. 13A and 13B are diagrams illustrating an example of medium placement after a display position determination process in a case where external light is reflected at one location on the display surface of the display/input unit, in comparison with a conventional medium placement example. 13A and 13B are diagrams illustrating an example of medium placement after a display position determination process in a case where external light is reflected at multiple points on the display surface of the display/input unit, in comparison with a conventional medium placement example. 11 is a flowchart illustrating the flow of a medium reading process.

Hereinafter, an embodiment of the present invention will be described with reference to the drawings.
FIG. 1 is a diagram illustrating a functional configuration of an automated transaction device according to one embodiment. The functional configuration illustrated in FIG. 1 shows a configuration related to a display position determination function provided in the automated transaction device, and configurations related to other functions provided in the automated transaction device are not shown. The display position determination function is a function for determining the position of a medium placement area to be displayed on the display unit. The medium placement area is an area that specifies an area in which a medium to be read is placed. The medium to be read is, for example, a document such as a deposit slip or a transfer form. The automated transaction device is, for example, an ATM, and is installed in a financial institution such as a bank.

As illustrated in FIG. 1, the automated transaction device 1 includes a display unit 10, an imaging unit 20, a detection unit 30, and a determination unit 40.

The display unit 10 displays a medium placement area that specifies the area where the medium to be read is placed. The imaging unit 20 is used to read the medium to be read. The detection unit 30 detects the contour of the external light reflection area shown in the image of the display unit 10 captured by the imaging unit 20. The determination unit 40 determines the position of the medium placement area to be displayed on the display unit 10 based on the position of the contour of the external light reflection area detected by the detection unit 30 in the image of the display unit 10 captured by the imaging unit 20.

The detection unit 30 may detect the contour of an external light reflection area that is displayed within the readable range of an image captured by the imaging unit 20 on the display unit 10.

The determination unit 40 may calculate an evaluation value based on the shortest distance between the contour of the corresponding area and the contour of the external light reflection area at each position of the corresponding area, which is an area corresponding to the medium placement area and which the display unit 10 may place on the image captured by the imaging unit 20, and determine the display position of the display unit 10 corresponding to the position of the corresponding area with the maximum evaluation value as the position of the medium placement area to be displayed on the display unit 10.

In this case, the evaluation value may be set to 0 when the outline of the corresponding area is not included within the readable range of the image captured by the image capturing unit 20 on the display unit 10. Also, the shortest distance when the outline of the corresponding area and the outline of the external light reflection area overlap may be set to 0.

In addition, when the detection unit 30 detects a single contour of an external light reflection area, the evaluation value may be the shortest distance between the contour of the corresponding area and the contour of the external light reflection area.

In addition, the evaluation value may be set to 0 when there are multiple contours of the external light reflection area detected by the detection unit 30 and at least one of the shortest distances between the contour of the corresponding area and each of the contours of the multiple external light reflection areas is 0.

In addition, when there are multiple contours of the external light reflection area detected by the detection unit 30 and all of the shortest distances between the contour of the corresponding area and each of the contours of the multiple external light reflection areas are greater than 0, the evaluation value may be the minimum value of those shortest distances.

With this type of automated transaction device 1, when a medium to be read is placed on the display surface of the display unit 10 and read, even if external light from a ceiling light or the like is reflected white on the display surface, it is possible to prevent the outline of the medium from overlapping with the reflection of the external light, making it possible to read the medium. The automated transaction device 1 will be described in more detail below.

Figure 2 is a perspective view illustrating the external configuration of an automated transaction device according to one embodiment. Figure 3 is a diagram illustrating the imaging range of the OCR camera unit.

As shown in FIG. 2, the automated transaction device 1 includes an OCR camera unit 101, a passbook insertion/ejection unit 102, a card insertion/ejection unit 103, a coin deposit/withdrawal unit 104, a banknote deposit/withdrawal unit 105, and a display/input unit 106.

The OCR camera unit 101 captures an image of a medium placed in a medium placement area displayed on the display input unit 106. The imaging range (angle of view) of the OCR camera unit 101 is the imaging range 101a illustrated in FIG. 3, which is a range that includes at least the medium placement area 201 that can be displayed on the display input unit 106. The shape and size of the medium placement area 201 are the same or approximately the same as the shape and size of the medium.

The passbook insertion and ejection unit 102 accepts and returns passbooks. The card insertion and ejection unit 103 accepts and returns cards (cash cards, etc.) used in transactions. The card insertion and ejection unit 103 also ejects receipts on which transaction details are printed. The coin deposit and ejection unit 104 accepts deposited coins and ejects withdrawn coins. The paper money deposit and ejection unit 105 accepts deposited banknotes and ejects withdrawn banknotes.

The display/input unit 106 displays various transaction screens (including the media reading screen) and accepts operational input from the customer.

Figure 4 is a diagram illustrating the hardware configuration of an automated transaction device according to one embodiment. As illustrated in Figure 4, the automated transaction device 1 includes a control unit 121, a display input unit 106, an OCR camera unit 101, a card processing unit 122, a passbook processing unit 123, a coin processing unit 124, a banknote processing unit 125, and a power supply unit (power supply panel) 126.

The control unit 121 includes a processor 131, a RAM (Random Access Memory) 132, a communication interface 133, a graphics processing unit 134, an external storage device 135, and an input/output interface 136, which are interconnected by a bus 137. The control unit 121 may further include an FPGA (Field Programmable Gate Array) 138 or an ASIC (Application Specific Integrated Circuit) (not shown) connected to the bus 137. In this case, some of the functions of the control unit 121 may be performed by the FPGA 138 or the ASIC.

The processor 131 is, for example, a CPU (Central Processing Unit). The processor 131 controls the overall operation of the automated transaction device 1 by executing various programs such as an OS (Operating System) program and application programs. For example, the processor 131 controls operations related to a display position determination function, such as causing the OCR camera unit 101 to capture an image of the display input unit 106, detecting the contour of an external light reflection area shown in the image captured by the OCR camera unit 101, and determining the position of the medium placement area 201 to be displayed on the display input unit 106 based on the position of the contour of the detected external light reflection area in the image. The RAM 132 temporarily stores part of the program executed by the processor 131 and is used as a working memory area for the processor 131.

The communication interface 133 is connected to a network (not shown) and communicates with, for example, a host computer (not shown) via the network.

The graphics processing unit 134 is connected to the display device 106a of the display input unit 106, and displays various trading screens on the display device 106a in accordance with control signals from the processor 131.

The external storage device 135 is, for example, a hard disk drive (HDD). The external storage device 135 stores various programs executed by the processor 131 and various data required for executing the programs.

The input/output interface 136 is connected to the touch panel 106b of the display/input unit 106, the OCR camera unit 101, the card processing unit 122, the passbook processing unit 123, the coin processing unit 124, and the banknote processing unit 125. Although not shown, the automated transaction device 1 may further include a portable recording medium interface capable of writing information to a portable recording medium and reading information from the portable recording medium, and the input/output interface 136 may further be connected to the portable recording medium interface.

The input/output interface 136 notifies the processor 131 of control signals to the touch panel 106b, the OCR camera unit 101, the card processing unit 122, the passbook processing unit 123, the coin processing unit 124, and the banknote processing unit 125. The input/output interface 136 also notifies the processor 131 of signals (data, information) from the touch panel 106b, the OCR camera unit 101, the card processing unit 122, the passbook processing unit 123, the coin processing unit 124, and the banknote processing unit 125.

The display input unit 106 includes a display device 106a and a touch panel 106b arranged on the display device 106a. The display device 106a is, for example, an LCD (Liquid Crystal Display). A polarizing filter is provided on the display surface of the display device 106a so that light from the screen displayed on the display device 106a does not enter the OCR camera unit 101. This allows a black background to be realized in an image captured by the OCR camera unit 101 of a medium placed in the medium placement area 201 displayed on the display input unit 106 (display device 106a), ensuring contrast with the white medium. The touch panel 106b detects an operation input to the touch panel 106b and notifies the processor 131 of the detection result.

The OCR camera unit 101 includes an image sensor such as a CCD (Charge Coupled Device) or a CMOS (Complementary Metal Oxide Semiconductor).

The card processing unit 122 inserts and ejects cards from the card insertion/ejection section 103. It also reads and writes information (data) from the magnetic stripe and IC (Integrated Circuit) chip of the inserted card. The card processing unit 122 also prints transaction details on a receipt and ejects it from the card insertion/ejection section 103.

The passbook processing unit 123 inserts and ejects passbooks from the passbook insertion and ejection unit 102. It also reads information (data) recorded on the magnetic stripe of the inserted passbook and writes information into the passbook.

The coin processing unit 124 deposits and withdraws coins from the coin deposit/withdrawal section 104. It also counts and validates deposited coins and counts withdrawn coins.

The banknote processing unit 125 deposits and withdraws banknotes through the banknote deposit/withdrawal unit 105. It also counts and validates deposited banknotes and counts withdrawn banknotes.

The power supply unit 126 supplies power to each part of the automated transaction device 1.

In the automated transaction device 1 illustrated in FIG. 4, the display input unit 106 (display device 106a) is an example of the display unit 10. The OCR camera unit 101 is an example of the imaging unit 20. The processor 131 is an example of the detection unit 30 and the determination unit 40.

Next, the process related to the display position determination function performed by the automated transaction device 1 (hereinafter referred to as the "display position determination process") will be described with reference to Figures 5 to 7. Note that this display position determination process is a process that is performed by the processor 131 executing the above-mentioned program, for example, when the automated transaction device 1 is installed or started up (when the power is turned on).

Figure 5 is a flow chart illustrating the flow of the display position determination process. As illustrated in Figure 5, when the automated transaction device 1 starts the display position determination process, it first captures an image 301 by using the OCR camera unit 101 to capture the display input unit 106 with no medium placed on it (step (hereinafter simply referred to as "S") 101). Here, assume that external light such as a ceiling light is reflected on the display surface of the display input unit 106, and the captured image 301 shows the reflection of that external light (e.g., external light reflection 302).

Next, the automated transaction device 1 performs image processing to convert the image 301 to grayscale, and obtains a grayscale image 303 (S102).

Next, the automated transaction device 1 performs binarization image processing on the grayscale image 303 to obtain a binarized image 304 (S103). In this binarization image processing, the binarization threshold may be determined based on the pixel values of the surrounding pixels.

Next, the automated transaction device 1 detects the contour of the external light reflection area (e.g., external light reflection area 306) represented within the OCR possible range (readable range) 305 of the binary image 304 (S104). Here, the OCR possible range 305 is predefined based on the performance (resolution, angle of view, range in focus, etc.) of the OCR camera unit 101. In addition, the detection of the contour of the external light reflection area is performed by detecting the closed contour of the object by labeling within the OCR possible range 305 of the binary image 304. As a labeling method, for example, an algorithm described in "Satoshi Suzuki and others. Topological structural analysis of digitized binary images by border following. Computer Vision, Graphics, and Image Processing, 30(1):32-46, 1985." (hereinafter referred to as "Document A") can be used. Note that in S104, the contour of the external light reflection area shown within the OCR possible range 305 of the binarized image 304 is detected, but the contour of the external light reflection area shown in the binarized image 304 may be detected not only within the OCR possible range 305.

Next, the automated transaction device 1 calculates an evaluation value based on the shortest distance between the contour of the corresponding area 307 and the contour of the external light reflection area (e.g., the external light reflection area 306) at each position of the corresponding area 307, which is an area that can be placed on the binary image 304 and corresponds to the medium placement area 201, and determines the display position of the display input unit 106 corresponding to the position of the corresponding area 307 where the evaluation value is maximum as the position of the medium placement area 201 to be displayed on the display input unit 106 (S105). Since the imaging range of the OCR camera unit 101 is fixed, the display position of the corresponding display input unit 106 can be obtained from the position on the binary image 304 based on the imaging range. When S105 is completed, the automated transaction device 1 ends the display position determination process illustrated in FIG. 5.

Here, the process of S105 (hereinafter referred to as the "decision process") will be described in more detail with reference to Figures 6 and 7. Figure 6 is a diagram explaining the parameters used in the decision process. Figure 7 is a flowchart illustrating the flow of the decision process.

First, the parameters used in the determination process will be described with reference to FIG. 6. In the determination process, as shown in FIG. 6, the upper left vertex of the binary image 304 is the origin (0,0), the width direction of the binary image 304 is the X direction, and the height direction is the Y direction. The width of the binary image 304 is w dots, and the height is h dots. The OCR possible range 305 is represented as O(0). In addition, it is assumed that N (N is an integer equal to or greater than 1) contours of the external light reflection area have been detected, and the contours of the N external light reflection areas are represented as O(n) (n=1, 2, ..., N). FIG. 6 shows a case where two contours of the external light reflection area (O(1), O(2)) have been detected. In addition, the corresponding area 307, which is an area that can be placed on the binary image 304 and corresponds to the medium placement area 201, is represented as A(x,y). The (x, y) in A(x, y) represents the position of the top left vertex of the corresponding area 307, and also represents the position of the corresponding area 307. Furthermore, the threshold value used to determine whether the shortest distance between the contour of the corresponding area 307 and the contour of the external light reflection area is an allowable distance (corresponding to whether overlap between the contour of the medium and the external light reflection can be avoided) is represented as D. Furthermore, the evaluation value at each position of the corresponding area 307 that can be placed on the binarized image 304 is represented as R(x, y). Based on these assumptions, the determination process will be specifically described using FIG. 7.

As shown in FIG. 7, when the automated transaction device 1 starts the determination process, it first performs loop process A from x=1 to w (x=1, 2, ..., w) (S201 to S212).

In loop process A for each x, the automated transaction device 1 performs loop process B for y = 1 to h (y = 1, 2, ..., h) (S202 to S211).

In loop processing B for each y, the automated transaction device 1 sets R(x, y) = 0 (S203) and determines whether A(x, y) is included in O(0) (S204).

If the result of the determination in S204 is NO, the automated transaction device 1 sets R(x, y) = 0 (S205) and ends loop processing B for y at this time.

On the other hand, if the result of the determination in S204 is YES, the automated transaction device 1 performs loop processing C for n=1 to N (n=1, 2, ..., N) (S206 to S210).

In loop processing C for each n, the automated transaction device 1 calculates the shortest distance between the contour of A(x, y) and O(n) as d (S207). In this calculation, the shortest distance when the contour of A(x, y) and O(n) overlap is set to 0 (d=0). In addition, when the contour of A(x, y) and O(n) are separated, or when O(n) is included within the contour of A(x, y), d>0.

Next, the automated transaction device 1 judges whether d>0 (S208). If the judgment result of S208 is NO, the automated transaction device 1 ends loop process C regardless of the value of n at this time, sets R(x, y)=0 (S205), and ends loop process B for y at this time.

On the other hand, if the judgment result of S208 is YES, the automated transaction device 1 assigns d to R(x, y) if R(x, y)=0, and assigns the smaller of d and R(x, y) to R(x, y) if R(x, y)=0 is not true (S209).

In this way, when the processing of S201 to S212 is completed, the automated transaction device 1 acquires the maximum value of R(x, y) obtained in the processing of S201 to S212 as Rmax, and acquires the position of A(x, y) when the maximum value of R(x, y) was obtained as Amax (S213).

Next, the automated transaction device 1 judges whether Rmax>D (S214). If the judgment result of S214 is YES, the automated transaction device 1 determines the display position of the display input unit 106 corresponding to Amax as the position of the medium placement area 201 to be displayed on the display input unit 106 (S215).

On the other hand, if the determination result in S214 is NO, the automated transaction device 1 determines that it is impossible to avoid overlap between the outline of the medium and the reflection of external light (S216). In this case, the automated transaction device 1 may notify the bank clerk of the determination result. The notification method may be, for example, displaying the determination result on the display input unit 106, or notifying the determination result to a bank clerk terminal. Such notification allows the bank clerk to review the installation location of the automated transaction device 1, etc.

When S215 or S216 is completed, the automated transaction device 1 ends the decision process illustrated in FIG. 7.

By displaying the medium placement area 201 in the position determined in S215 in this manner, it is possible to avoid overlap between the outline of the medium placed in the medium placement area 201 and reflections of external light. Here, an example of medium placement after the display position determination process will be described using Figures 8 and 9, in comparison with a conventional example of medium placement. Note that for ease of explanation, the display surface of the display input unit 106 is shown in black in Figures 8 and 9. Also, illustrations of characters, lines, etc. printed on the medium are omitted.

Figure 8 is a diagram showing an example of medium placement after a display position determination process when there is external light reflection at one point on the display surface of the display input unit, in comparison with a conventional medium placement example. As shown in Figure 8, when there is external light reflection 311 at one point on the display surface 106c of the display input unit 106, in the past, the display position of the medium placement area 201 was fixed, so when a medium reading screen 312 including the medium placement area 201 was displayed on the display input unit 106 and a medium 313 was placed on the medium placement area 201, the outline of the medium 313 and the external light reflection 311 may overlap. In this case, the outline of the medium 313 cannot be detected, so the medium 313 cannot be read. In contrast, in this embodiment, the display position determination process determines the display position of the medium placement area 201 to be within the OCR possible range 305 and to a position where the outline of the placed medium 313 does not overlap with the external light reflection 311, so that when the medium reading screen 312 including the medium placement area 201 is displayed on the display input unit 106 and the medium 313 is placed on the medium placement area 201, the outline of the medium 313 does not overlap with the external light reflection 311. Therefore, the outline of the medium 313 can be detected, and the medium 313 can be read.

9 is a diagram showing an example of medium placement after the display position determination process when there is external light reflection at multiple locations on the display surface of the display input unit, in comparison with a conventional medium placement example. As shown in FIG. 9, when there is external light reflection 321 (321a, 321b) at multiple locations (two locations in this example) on the display surface 106c of the display input unit 106, in the past, the display position of the medium placement area 201 was fixed, so when the medium reading screen 312 including the medium placement area 201 was displayed on the display input unit 106 and the medium 313 was placed on the medium placement area 201, the outline of the medium 313 and the external light reflection 321 may overlap. In this case, the outline of the medium 313 cannot be detected, so the medium 313 cannot be read. In contrast, in this embodiment, the display position determination process determines the display position of the medium placement area 201 to be within the OCR possible range 305 and to be a position where the contour of the placed medium 313 does not overlap with the external light reflection 321. Therefore, when the medium reading screen 312 including the medium placement area 201 is displayed on the display input unit 106 and the medium 313 is placed on the medium placement area 201, the contour of the medium 313 does not overlap with the external light reflection 321. In the example of FIG. 9, the medium placement area 201 is displayed so that one of the two external light reflections, external light reflection 321a, is contained within the medium placement area 201 and the other external light reflection area 321b is located at a distance outside the medium placement area 201. Therefore, the contour of the medium 313 placed on the medium placement area 201 can be detected, and the medium 313 can be read.

Now, with reference to FIG. 10, we will explain the medium reading process performed by the automated transaction device 1 when a medium is placed in the medium placement area 201 displayed on the display input unit 106 after the above-mentioned display position determination process. This medium reading process is performed by the processor 131 executing the above-mentioned program. For ease of explanation, FIG. 10 omits illustrations of characters, lines, etc. printed on the medium shown in the image. Also, the detected contour and polygonal approximation contour are shown diagrammatically.

Figure 10 is a flow chart illustrating the flow of the medium reading process. As illustrated in Figure 10, when the automated transaction device 1 starts the medium reading process, it first captures an image 331 by the OCR camera unit 101 (S301). Here, although external light such as a ceiling light is reflected on the display surface of the display input unit 106, the outline of the medium placed in the medium placement area 201 displayed at the position determined by the above-mentioned display position determination process does not overlap with the external light reflection, so the outline of the medium 332 shown in image 331 does not overlap with the external light reflection (e.g., external light reflection 333).

Next, the automated transaction device 1 performs image processing to convert the image 331 to grayscale, and obtains a grayscale image 334 (S302).

Next, the automated transaction device 1 performs binarization image processing on the grayscale image 334 to obtain a binarized image 335 (S303). In this binarization image processing, the binarization threshold may be determined based on the pixel values of the surrounding pixels.

Next, the automated transaction device 1 detects contours in the binarized image 335 (S304). This contour detection is performed by detecting closed contours of objects through labeling. As a labeling method, for example, the algorithm described in the above-mentioned document A can be used. Through the process of S304, the contour 336a of the medium 336 shown in the binarized image 335 and the contour of the external light reflection area (for example, the contour 337a of the external light reflection area 337) are detected.

Next, the automated transaction device 1 performs polygon approximation to approximate the contour detected in S304 to a polygon (S305). The Douglas-Peucker algorithm or the like can be used as a polygon approximation method. The process of S305 results in contour 336b, which is a polygon approximation of contour 336a, or contour 337b, which is a polygon approximation of contour 337a, for example.

Next, the automated transaction device 1 identifies the contour of the medium 336 from the contours approximated as polygons in S305 (S306). This identification can be performed based on the predefined shape and size of the medium to be read. Here, contour 336b is identified as the contour of the medium 336.

Next, the automated transaction device 1 performs trapezoidal conversion image processing on the binarized image 335 so that the shape (trapezoidal shape) of the contour 336b of the medium 336 identified in S306 becomes similar to the shape (rectangular shape) of the actual medium (S307).

Next, the automated transaction device 1 cuts out a partial image 339 of the character area to be read in the area of the medium 338 represented in the binary image that was trapezoidally converted in S307 (S308). The cutting out of the partial image 339 at this time can be performed based on the predefined layout (format) of the medium to be read.

Next, the automated transaction device 1 performs character recognition on the partial image 339 of the character area to be read that was extracted in S308 (S309), and ends the medium reading process illustrated in FIG. 10.

According to the automated transaction device 1 described above, when the medium to be read is placed on the display surface of the display input unit 106 and reading is performed, even if external light from a ceiling light or the like is reflected white on the display surface, it is possible to prevent the outline of the medium from overlapping with the reflection of the external light, making it possible to read the medium.

Note that messages, illustrations, etc. may be displayed in the medium placement area 201 displayed on the display input unit 106. In this case, the customer may be prompted to specify the orientation of the medium to be placed in the medium placement area 201 using the message, illustration, etc.

Although the embodiment of the present invention has been described above, the present invention is not limited to the above-mentioned embodiment, and various improvements and modifications are possible without departing from the gist of the present invention.

REFERENCE SIGNS LIST 1 Automatic transaction device 10 Display unit 20 Imaging unit 30 Detection unit 40 Determination unit 101 OCR camera unit 101a Imaging range 102 Passbook insertion and ejection unit 103 Card insertion and ejection unit 104 Coin deposit and withdrawal unit 105 Paper money deposit and withdrawal unit 106 Display input unit 106a Display device 106b Touch panel 106c Display surface 121 Control unit 122 Card processing unit 123 Passbook processing unit 124 Coin processing unit 125 Paper money processing unit 126 Power supply unit 131 Processor 132 RAM
133 Communication interface 134 Graphic processing unit 135 External storage device 136 Input/output interface 137 Bus 138 FPGA
201 Medium placement area 301 Image 302 External light reflection 303 Grayscale image 304 Binarized image 305 OCR possible range 306 External light reflection area 307 Corresponding area 311 External light reflection 312 Medium reading screen 313 Medium 321, 321a, 321b External light reflection 331 Image 332 Medium 333 External light reflection 334 Grayscale image 335 Binarized image 336 Medium 336a, 336b Contour 337 External light reflection area 337a, 337b Contour 338 Medium 339 Partial image

Claims (10)

a display unit that displays a medium placement area that specifies an area in which a medium to be read is placed, the display unit being imaged by an image capture unit that is used to read the medium to be read;
Detecting an outline of an external light reflection area shown in an image captured by the imaging unit;
determining a position of the medium placement area to be displayed on the display unit based on a position of a contour of the external light reflection area in the image;
causing a processor to execute the process ;
The direction of the vertical component of the imaging direction of the imaging unit is the same as the direction of the vertical component of the incident direction of the reflected external light.
A program characterized by: In the detection, an outline of an area where external light is reflected and displayed within a readable range of the image is detected.
2. The program according to claim 1 . In the determination, an evaluation value is calculated based on the shortest distance between the contour of the corresponding area and the contour of the external light reflection area at each position of a corresponding area that can be arranged on the image and corresponds to the medium placement area, and the display position of the display unit corresponding to the position of the corresponding area where the evaluation value is maximum is determined as the position of the medium placement area to be displayed on the display unit.
3. The program according to claim 1 or 2. When the outline of the corresponding area is not included within the readable range of the image, the evaluation value is set to 0.
4. The program according to claim 3. The shortest distance when the contour of the corresponding area and the contour of the external light reflection area overlap is set to 0.
5. The program according to claim 4. The evaluation value when the external light reflection region in the image has one contour is set to the shortest distance between the contour of the corresponding region and the contour of the external light reflection region.
6. The program according to claim 5. When the image has a plurality of contours of the external light reflection region and at least one of the shortest distances between the contour of the corresponding region and each of the contours of the plurality of external light reflection regions is 0, the evaluation value is set to 0.
6. The program according to claim 5. When the image has a plurality of contours of the external light reflection region and all of the shortest distances between the contour of the corresponding region and each of the contours of the plurality of external light reflection regions are greater than 0, the evaluation value is set to a minimum value of the shortest distances.
6. The program according to claim 5. a display unit that displays a medium placement area that specifies an area in which a medium to be read is placed, the display unit being imaged by an image capture unit that is used to read the medium to be read;
Detecting an outline of an external light reflection area shown in an image captured by the imaging unit;
determining a position of the medium placement area to be displayed on the display unit based on a position of a contour of the external light reflection area in the image;
This is done by the automated transaction device,
The direction of the vertical component of the imaging direction of the imaging unit is the same as the direction of the vertical component of the incident direction of the reflected external light.
A display position determining method comprising: a display unit that displays a medium placement area for designating an area in which a medium to be read is placed;
An imaging unit used to read the medium to be read;
a detection unit that detects an outline of an external light reflection area displayed in an image captured by the imaging unit of the display unit;
a determination unit that determines a position of the medium placement area to be displayed on the display unit based on a position of a contour of the external light reflection area in the image;
Equipped with
The direction of the vertical component of the imaging direction of the imaging unit is the same as the direction of the vertical component of the incident direction of the reflected external light.
1. An automatic transaction device comprising: