JP7484229B2 - Card cutting device and program - Google Patents

Description

The present invention relates to a card cutting device and a program for cutting out the area of a card from an image of the card.

Nowadays, it is common to use images of the face of an ID card, such as a driver's license, to carry out various administrative procedures. For example, images of the face of an ID card are used to verify the identity of a person, obtain text information and facial images from the face of the ID card, etc.

The ID card face image used in these processes may be one taken with a mobile device such as a smartphone, as in Patent Document 1, and the first step in the above process may be to cut out the area of the card from the image of the card. When cutting out the area of the card, it is common to detect the boundary between the card and its background by differences in brightness (whiteness) or color.

JP 2017-215815 A

When detecting the boundary between a card and its background based on differences in brightness and color like this, there are cases where the boundary cannot be detected properly depending on the background when the card is photographed, and the area of the card cannot be accurately cut out. For example, if the background has the same brightness and color as the card, it is difficult to detect the boundary between the card and its background, and if the background itself has a pattern, the pattern may be mistakenly detected as the boundary.

The present invention was made in consideration of the above problems, and aims to provide a card cutting device etc. that can more accurately cut out the area of a card from an image that shows the card.

The first invention for solving the above-mentioned problems is a card cutting device that cuts out the area of a card from an image containing the card, comprising: an extraction means that extracts a first closed area of the face of the card from the image; and a cutting means that determines the area for cutting out the card based on the difference between the first closed area extracted by the extraction means and a template, and cuts out the area of the card from the image , wherein the extraction means extracts the first closed area by sequentially searching from the starting point outward for pixels that are determined to be the same color as a pixel of a starting point, and the pixel of the starting point is a pixel whose color is determined to be the same as the colors of the pixels above, below, left and right of the pixel .

In the present invention, the closed area of the card face is extracted from the image, and the range for cutting out the card from the image is determined according to the difference between the extracted closed area and a template (information indicating a positive closed area). This makes it possible to cut out only the area of the card without being influenced by the background of the card, and the area of the card can be cut out more accurately from an image in which the card appears. In addition, by being able to suitably extract the closed area from the image and determining the start point of the process as described above, it is possible to avoid extracting minute closed areas such as character parts and to extract the dominant closed area on the card face, which helps to improve the accuracy of cutting out.

It is desirable that the extraction means extracts a second closed area on the face of the card from the image, and that the card cutting device further has a determination means for determining whether or not the difference is correct based on the second closed area extracted by the extraction means and a template for the second closed area.
This makes it possible to check whether the above-mentioned differences are correct, enabling more accurate segmentation.

It is preferable that the difference be found by comparing the position of a portion of interest between the first closed region extracted by the extraction means and a template for the first closed region.
This makes it possible to obtain the above-mentioned difference through simple processing.

The card dispenser is, for example, a mobile terminal having a camera for photographing the card.
By using a mobile device with a camera, such as a smartphone or tablet, as a card cutting device, a user can cut out the area of the card from an image of the card taken with the mobile device and use this for various administrative purposes.

A second invention is a program for causing an information processing device to function as a card cutting device that cuts out the area of a card from an image containing the card, the program having an extraction means for extracting a first closed area of the face of the card from the image, and a cutting means for determining the area for cutting out the card based on the difference between the first closed area extracted by the extraction means and a template, and cutting out the area of the card from the image, wherein the extraction means extracts the first closed area by sequentially searching from the starting point outward for pixels that are determined to be the same color as a pixel of the starting point, and the pixel of the starting point is a pixel whose color is determined to be the same as the colors of the pixels above, below, left and right of the pixel .

The present invention provides a card cutting device that can more accurately cut out the area of a card from an image that shows the card.

FIG. FIG. 2 is a diagram showing the hardware configuration of the card cutting device 1. FIG. 2 is a diagram showing the functional configuration of the card cutting device 1. FIG. 1 is a diagram showing an outline of a card 10. 4 is a flowchart showing a process executed by the card dispensing device 1. 1 is an example of an image 100 captured of a card 10. FIG. 4 is a diagram for explaining determination of a start point. 4A and 4B are diagrams for explaining extraction of a closed region A; 5A and 5B are diagrams for explaining calculation of a difference between a closed region A and a template 200. A diagram showing the templates 200, 300 and the positive cut-out range 400 of the card 10. A diagram showing a starting point 150. FIG. 13 is a diagram for explaining a determination of whether a difference is correct or not. FIG. 13 shows a cut-out range 180 of the card 10. 5A to 5C are diagrams illustrating correction of a cut-out image.

The following describes in detail a preferred embodiment of the present invention with reference to the drawings.

(1. Card cutting device 1)
1 is a diagram showing a card cutting device 1 according to an embodiment of the present invention. The card cutting device 1 is a mobile terminal with a camera, such as a smartphone or a tablet terminal, and cuts out the area of the card 10 from an image of the card 10.

The cut-out image of the card 10 (image of the face of the card) is used for various administrative processes such as identity verification, obtaining text information from the face of the card, and acquiring a facial image, as described above, and the card cutting device 1 transmits the cut-out image of the card 10 to an external information processing device (not shown) as necessary for carrying out the above-mentioned processes.

Figure 2 is a diagram showing the hardware configuration of the card dispensing device 1. As shown in Figure 2, the card dispensing device 1 can be realized by an information processing device configured by connecting a control unit 101, a memory unit 102, a display unit 103, an input unit 104, a communication unit 105, a camera 106, an audio input/output unit 107, etc., via a bus or the like. However, this is not limited to this, and various configurations can be used as appropriate.

The control unit 101 is composed of a CPU, ROM, RAM, etc. The CPU loads programs related to the processing of the card dispensing device 1 stored in storage media such as the memory unit 102 and ROM into a work area on the RAM and executes them. The ROM is a non-volatile memory, and permanently stores programs such as the boot program and BIOS, as well as data. The RAM is a volatile memory, and temporarily stores programs and data loaded from the memory unit 102, ROM, etc., and also has a work area that the control unit 101 uses to perform various processes.

The memory unit 102 is a flash memory or the like, and stores the programs executed by the card dispensing device 1, data required for executing the programs, the OS, etc. These programs and data are read and executed by the control unit 101 as necessary. The memory unit 102 stores dedicated application programs for executing the processes described in FIG. 5 below.

The display unit 103 has a display device such as a liquid crystal panel, and a touch panel is provided as the input unit 104 for inputting data to the card dispensing device 1.

The communication unit 105 is a communication interface that mediates communication via a network, and communicates with external information processing devices, etc.

Camera 106 is a high-resolution area camera that is composed of an optical lens, an imaging element such as a CCD (Charge Coupled Device) or a CMOS (Complementary Metal-Oxide Semiconductor), an A/D (Analog/Digital) conversion unit, etc. Camera 106 photoelectrically converts the subject image input through the optical lens using the imaging element to generate an analog image signal. The A/D conversion unit then converts the analog image signal into digital image data.

The audio input/output unit 107 includes a microphone for audio input and a speaker for audio output.

(2. Functions of Card Dispenser 1)
Fig. 3 is a diagram showing the functional configuration of the card dispensing device 1. As shown in Fig. 3, the card dispensing device 1 has an extraction means 111, a determination means 112, a dispensing means 113, and the like.

The extraction means 111 is a means by which the control unit 101 of the card cutting device 1 extracts a closed area on the face of the card 10 from an image showing the card 10. In this embodiment, two closed areas are extracted, one of which is used to calculate the difference from a template (information indicating a positive closed area). This difference is used to determine the range in which the card 10 is cut out from the image, and details of this will be described later.

The determination means 112 is a means for the control unit 101 of the card dispensing device 1 to determine whether the above difference is correct or not. This determination is made using the other of the two closed areas extracted by the extraction means 111.

The cutting means 113 is a means for causing the control unit 101 of the card cutting device 1 to determine the range for cutting out the card 10 in accordance with the above difference, and cutting out the range of the card 10 from the image in which the card 10 appears.

(3. Card 10)
In this embodiment, the card 10 is, for example, an ID card (identification card) such as a driver's license, a My Number card, a residence card, or a special permanent resident card, but is not limited to these and may be any other type of card.

Figure 4 is a diagram showing an overview of card 10, and is a schematic representation of the face of card 10. Card 10 has various lines 12, facial images 13, characters 14, etc. printed on a roughly rectangular card substrate 11, and the configuration of the face of the card varies depending on the type of card 10.

(4. Processing of Card Dispenser 1)
Fig. 5 is a flow chart showing the processing executed by the card dispensing device 1. Steps S1 to S10 in Fig. 5 are executed by the control unit 101 of the card dispensing device 1 by controlling each unit of the card dispensing device 1.

In this embodiment, the user first operates the card dispenser 1 to launch a dedicated application program stored in the memory unit 102. The card dispenser 1 then displays a menu screen (not shown) on the display unit 103, starts the camera 106 in response to the user's selection on the menu screen, and photographs the card 10 with the camera 106 in response to the user's operation (S1). This obtains an image showing the card 10 together with the background.

Figure 6 is an example of an image 100 captured of a card 10. In this embodiment, a user captures a photo of the card 10 against a variety of backgrounds 20, and as shown in Figure 6, the background 20 of the card 10 may have a pattern. The card cutting device 1 of this embodiment executes the cutting method from S2 onwards in Figure 5, making it possible to cut out the area of the card 10 without being affected by the background 20.

Furthermore, the orientation of the camera 106 when photographing the card 10 is not constant, and there are cases where the card 10 is photographed tilted within a plane as shown in the figure. The position of the center of the card 10 in the image 100 also shifts due to the position of the camera 106, and the shape of the card 10 in the image 100 may also be distorted due to the position and orientation of the camera 106. With the cut-out method from S2 onwards in Figure 5, even if the card 10 is tilted, misaligned, distorted, etc. as described above due to the position and orientation of the camera 106, the area of the card 10 can be correctly cut out accordingly.

In this embodiment, when cutting out the area of the card 10 from the image 100 shown in FIG. 6, two area extraction processes (a first area extraction process and a second area extraction process) are performed to extract two closed areas on the face of the card 10.

In this embodiment, the two closed regions to be extracted are the regions indicated by the symbols A and B in FIG. 6. Closed regions A and B are closed regions with a contiguous color that is the same (or can be considered to be the same), and in this embodiment, closed regions A and B are the portions of the background color of card 10 (white in the example shown). Closed region A is located in the center of image 100, and closed region B is located above closed region A. In the first region extraction process, closed region A (first closed region) is extracted, and in performing region extraction, the starting point of the process is first determined (S2).

In this embodiment, the starting point of the process is determined to avoid character portions and the like, so as not to extract tiny closed areas such as the character portions on the face of the document. Therefore, in S2, the center of the image 100 is first set as a tentative starting point, and the color of that pixel 110 is compared with the colors of one or more pixels 120 located above, below, to the left, and to the right of the pixel 110, as shown in FIG. 7(a).

For all of these pixels 120, if it is determined that the color of pixel 120 is the same as the color of pixel 110, pixel 110 is determined to be the starting point for processing. Differences in color also include differences in brightness, and in the above determination, if the difference in pixel value between pixels 110 and 120 is equal to or less than a predetermined value, the color of pixel 120 can be determined to be the same as the color of pixel 110.

On the other hand, as shown in FIG. 7B, if at least one of the pixels 120 located above, below, left, or right of pixel 110 is covered by text or the like, the colors of all pixels 120 will not be the same as the color of pixel 110, and pixel 110 will not be the starting point of the process. In this case, the card cutting device 1 moves pixel 110, which is a candidate for the starting point, from the center of image 100 while repeatedly comparing the colors of pixels 110 and 120 as described above. As a result, if a pixel 110' is detected such that the colors of one or more pixels 120' above, below, left, and right are all determined to be the same as the color of pixel 110', as shown in pixel 110', then pixel 110' is determined to be the starting point of the process.

The amount and direction of movement when performing the above movement are not particularly limited, and if the starting point cannot be determined even after a predetermined number of comparisons, an error message indicating that the starting point could not be determined is displayed on the display unit 103, and processing can be terminated.

The card cutting device 1 starts the first area extraction process from the starting point determined in S2 (S3). This area extraction process is a process for extracting a closed area A of the card 10 including the starting point. There is no particular restriction on the specific method of the process, but as an example, a filling algorithm such as that shown in Figure 8 can be applied, and closed area A can be extracted by sequentially searching from the starting point outward for pixels that are determined to be the same color as the pixel of the starting point. The above determination can be made by comparing pixel values, as described above.

In the example of Figure 8, when pixel a is determined to be the starting point as shown in Figure 8(a), pixel a is treated as a pixel of closed area A, and is filled with a predetermined color different from the original color (background color) of pixel a as shown in Figure 8(b), and is then placed in the processing queue. Then, pixels b, c, d, and e, which are located above, below, to the left, and right of pixel a at the top of the queue and have the same color as pixel a's original color, are treated as pixels of closed area A and filled with the above-mentioned predetermined color as shown in Figure 8(c), and instead of removing pixel a from the processing queue, pixels b, c, d, and e are placed in the processing queue. Note that in the example shown, the pixels are placed in the processing queue in the following order: pixel b above pixel a, pixel c to the left, pixel d to the right, and pixel e below.

Next, the pixels above, below, left and right of pixel b at the top of the queue are checked, and similarly to the above, pixel f determined to be the same color as pixel a's original color is filled in with the above-mentioned predetermined color as shown in Figure 8(d) as a pixel of closed area A, and pixel f is added to the end of the processing queue instead of removing pixel b from the processing queue. This makes pixel c the top of the queue, so the pixels above, below, left and right of pixel c are checked, and pixel g determined to be the same color as pixel a's original color is filled in with the above-mentioned predetermined color as shown in Figure 8(e) as a pixel of closed area A, and pixel g is added to the end of the processing queue instead of removing pixel c from the processing queue.

After this, pixels d-m shown in Figure 8(f) are sequentially placed at the top of the queue, and the above process is repeated until the queue is finally empty. At that time, closed area A can be extracted using pixels a-m that have been filled with the specified color. The gray area in Figure 9(a) is an example of closed area A extracted from image 100 in Figure 6.

Next, the card cutting device 1 compares the extracted closed area A with a template indicating a positive closed area A, and determines the difference between them (S4).

Here, reference numeral 200 in FIG. 10 denotes a template showing a positive closed area A, i.e., a closed area A in an image P obtained by correctly photographing the card 10. In the example of FIG. 10, it is assumed that the card 10 is not tilted in the image P, the center of the card 10 is located at the center of the image P, and there is no distortion of the card 10.

The memory unit 102 of the card cutting device 1 stores the position of the attention area 210 of the template 200, as illustrated by the ● in Figure 10. The attention area 210 is, for example, a feature point such as a corner of the outline of the template 200, but there is no particular limitation on what the attention area 210 is, and it is also possible to use an edge of the outline as the attention area 210.

In S4, as shown by the circle in FIG. 9(a), the position of the attention portion 140 corresponding to the attention portion 210 is detected for the closed region A extracted in S3, and the positions of the corresponding attention portions 140 and 210 are compared to compare the closed region A extracted in S3 with the template 200.

For example, as shown in FIG. 9(b), when the closed area A extracted in S3 is rotated by α° in a plane with the center C of the image 100 as the reference, when the error in the position of the corresponding attention areas 140, 210 is minimized, the inclination of the closed area A extracted in S3 with respect to the template 200 (i.e., the inclination of the card 10 in the image 100) can be obtained as -α°.

In the example of FIG. 9(b), since there are multiple pairs (12 pairs) of corresponding attention areas 140, 210, the above-mentioned error can be calculated as the average error or the sum of the squares of the errors found for each pair of corresponding attention areas 140, 210, and this value can be calculated each time closed area A is rotated a specified angle. If the minimum error value is equal to or greater than a specified threshold, it is possible to display an error on the display unit 103 indicating that extraction of closed area A was not successful, and to end the process. Note that it is preferable to have multiple pairs of corresponding attention areas 140, 210 from the standpoint of accuracy.

In the example of FIG. 9(b), it is assumed that there are no differences between the closed region A extracted in S3 and the template 200 other than the inclination, but the differences between the closed region A and the template 200 are not limited to this. For example, when the error in the position of the corresponding attention areas 140, 210 is smallest when the position of the closed region A extracted in S3 is shifted by (dx, dy), the positional deviation of the closed region A extracted in S3 with respect to the template 200 (i.e., the positional deviation of the center of the card 10 in the image 100) can be obtained as (-dx, -dy). Note that dx indicates the horizontal displacement of the image 100, and dy indicates the vertical displacement of the image 100.

Similarly, the shape of the card 10 in the image 100 is distorted depending on the position and orientation of the camera 106 when the card 10 is photographed, so when a predetermined projective transformation is performed on the closed area A extracted in S3, the error in the position of the corresponding focus areas 140, 210 is minimized, and the distortion of the closed area A extracted in S3 relative to the template 200 (i.e., the distortion of the card 10 in the image 100) can be obtained as the inverse transformation of the above projective transformation.

In S4, the above-mentioned processing is used to determine the difference between the closed area A extracted in S3 and the template 200 as the inclination, positional deviation, distortion, or a combination of these of the closed area A relative to the template 200. In the following explanation, it is assumed that the closed area A extracted in S3 is inclined by -α° relative to the template 200 as shown in Figure 9(b), i.e., the orientation of the card 10 in the image 100 is inclined by -α° (there are no other differences).

The card cutting device 1 then performs a second area extraction process to extract another closed area B (second closed area) on the face of the card 10, and determines whether the difference found in S4 is correct based on the extracted closed area B and a template indicating a positive closed area B. Here, the starting point of the second area extraction process is first determined (S5).

As described above, closed area B is located above closed area A extracted earlier, and in this embodiment, as in the above, the position of the attention area 310 of the template 300 indicating the positive closed area B is stored in the memory unit 102 of the card cutting device 1 as shown in FIG. 10. Also, in this embodiment, the positive starting point 320 of the second area extraction process is set in advance as a point located within closed area B and in the background color part of the card 10, and is stored in the memory unit 102 of the card cutting device 1.

In S5, the difference calculated in S4 and the position of the starting point 320 can be referenced to calculate the position of the starting point 150 in the image 100 as shown in FIG. 11, and the area extraction process is started from this starting point 150.

By performing a process similar to S3, the card cutting device 1 extracts a closed area B including the starting point 150 from the image 100 (S6), as shown by the gray area in Figure 12 (a), and detects the position of the attention area 170 of the closed area B that corresponds to the attention area 310 (see Figure 10).

Then, the card dispenser 1 determines whether the difference found in S4 is correct (S7).

In S7, it is possible to determine whether the difference is correct or not by checking whether closed region B extracted in S6 matches template 300 when the closed region B is transformed using the difference found in S4.

In this embodiment, the difference is that the orientation of the card 10 is tilted by -α°, so as shown in FIG. 12(b), the positions of the corresponding focus areas 170, 310 are compared when the closed area B is rotated by α° around the center C of the image 100, and a judgment is made using the error. If the error is equal to or less than a predetermined value, the difference found in S4 is deemed correct, and if it exceeds the predetermined value, the difference found in S4 is deemed incorrect.

In the example of FIG. 12(b), since there are multiple (four) pairs of corresponding attention areas 170, 310, the above-mentioned error can be the average of the errors or the sum of the squares of the errors calculated for each pair of corresponding attention areas 170, 310. As mentioned above, it is preferable from the standpoint of accuracy to have multiple pairs of attention areas 170, 310.

If the card cutting device 1 determines that the difference found in S4 is correct (S7; YES), it determines the cutting range of the card 10 based on this difference and cuts out the range of the card 10 from the image 100 (S8).

In this embodiment, the positive cut-out range of the card 10 indicated by reference numeral 400 in FIG. 10 is stored in the memory unit 102 of the card cutting device 1. In S8, the difference obtained in S4 and the position of the positive cut-out range 400 are referenced, and the position of the cut-out range 180 of the card 10 in the image 100 can be calculated as shown in FIG. 13. In the example of FIG. 13, the cut-out range 180 is tilted from the positive cut-out range 400 in accordance with the difference obtained in S4.

In this way, the card cutting device 1 defines a cutting range 180 for cutting out the card 10 from the image 100, and cuts out the cutting range 180 from the image 100 (S8). The image (face image) of the card 10 cut out in S8 is shown in Figure 14 (a).

Then, the card cutting device 1 uses the difference found in S4 to correct the cut-out image as shown in FIG. 14(b) (S9). In the example of FIG. 14(b), the cut-out image is rotated according to the difference found in S4 and correction is performed.

If it is determined in S7 that the difference found in S4 is incorrect (S7; NO), the card dispensing device 1 displays an error message on the display unit 103 (S10) and ends the process.

As described above, the card cutting device 1 of this embodiment extracts the closed area A of the face of the card 10 from the image 100, and determines the cutting-out range 180 for cutting out the card 10 from the image 100 based on the difference between the extracted closed area A and its template 200. This makes it possible to cut out only the area of the card 10 without being influenced by the background 20 of the card 10, and allows the area of the card 10 to be cut out more accurately from an image in which the card 10 appears.

In addition, in this embodiment, it is determined whether the above differences are correct based on the closed region B extracted from the image 100 and its template 300, and the correctness of the differences is confirmed, enabling more accurate extraction.

Furthermore, in this embodiment, the above difference is obtained by comparing the positions of the closed area A extracted from the image 100 and the points of interest 140, 210 of the template 200, so the above difference can be found by simple processing.

In addition, in this embodiment, the closed area A can be suitably extracted from the image 100 by performing the area extraction process as illustrated in FIG. 8. Furthermore, when performing the area extraction process, the starting point of the process is determined as explained in FIG. 7, so that it is possible to avoid extracting minute closed areas such as character portions and to extract the dominant closed area A on the face of the card, which helps to improve the extraction accuracy.

In addition, in this embodiment, by using a mobile terminal with a camera, such as a smartphone or tablet terminal, as the card cutting device 1, the user can cut out the area of the card 10 from the image 100 taken of the card 10 with the mobile terminal and use this for various administrative tasks.

However, the present invention is not limited to the above embodiment. For example, in this embodiment, a mobile terminal with a camera owned by the user is used as the card cutting device 1, but the card cutting device 1 is not limited to this and may be a stationary information processing device without a photographing function, such as a server. The information processing device can receive the image 100 from the user's mobile terminal, and perform the processing from S2 onwards on the image 100, thereby cutting out the area of the card 10 from the image 100.

In this embodiment, two closed regions A and B are extracted, the difference is determined using closed region A, and the correctness of this difference is determined using closed region B, but the latter may be omitted, in which case it is sufficient to extract only one closed region A.

In this embodiment, the closed areas A and B are extracted as the background color of the face of the card 10, but the closed areas A and B are not limited to this and may be continuous areas printed in the same color on the card substrate 11, such as the areas surrounding the characters "1234567890" in Figure 4.

In addition, the method of the present invention can be used in conjunction with a conventional method that uses differences in brightness and color to extract the boundary between the card 10 and its background 20, and sets this as the cut-out range for the card 10. For example, it is possible to apply the method of the present invention in cases where the cut-out range cannot be determined using the conventional method.

The above describes preferred embodiments of the present invention with reference to the attached drawings, but the present invention is not limited to these examples. It is clear that a person skilled in the art can come up with various modified or revised examples within the scope of the technical ideas disclosed in this application, and it is understood that these also naturally fall within the technical scope of the present invention.

1: Card cutting device 10: Card 100: Image 111: Extraction means 112: Determination means 113: Cutting means 140, 170, 210, 310: Part of interest 180, 400: Cutting range 200, 300: Templates A, B; Closed region A

Claims (5)

A card cutting device that cuts out a range of a card from an image of the card,
an extraction means for extracting a first closed area of the face of the card from the image;
a cutout means for determining an area for cutting out the card in accordance with a difference between the template of the first closed area extracted by the extraction means and the template, and for cutting out the area of the card from the image;
having
the extraction means extracts a first closed region by sequentially searching from the starting point toward the outside for pixels determined to have the same color as the pixel at the starting point;
The starting pixel is a pixel whose color is determined to be the same as the colors of the pixels above, below, left, and right of the starting pixel.
A card dispensing device characterized by the above. The extraction means extracts a second closed area of the face of the card from the image,
The card dispensing device is
2. The card dispensing device according to claim 1, further comprising a determination means for determining whether or not the difference is correct based on the second closed area extracted by the extraction means and a template for the second closed area. The card cutting device according to claim 1 or 2, characterized in that the difference is found by comparing the position of the attention part of the first closed area extracted by the extraction means with a template for the first closed area. 4. The card dispensing device according to claim 1, wherein the card dispensing device is a mobile terminal having a camera for photographing the card. An information processing device,
A card cutting device that cuts out a range of a card from an image of the card,
an extraction means for extracting a first closed area of the face of the card from the image;
a cutout means for determining an area for cutting out the card in accordance with a difference between the template of the first closed area extracted by the extraction means and the template, and for cutting out the area of the card from the image;
A program for causing a card dispensing device to function as a card dispensing device having the following:
the extraction means extracts a first closed region by sequentially searching from the starting point toward the outside for pixels determined to have the same color as the pixel at the starting point;
The starting pixel is a pixel whose color is determined to be the same as the colors of the pixels above, below, left, and right of the starting pixel.
A program characterized by :

Images