JP7282257B2 - Image processing device, control method and control program - Google Patents

Description

The present disclosure relates to an image processing device, control method and control program, and more particularly to an image processing device, control method and control program for processing an image including a seal or logo.

In a company where the person in charge manually converts forms such as invoices into data, the work burden on the person in charge becomes large when it is necessary to convert a huge number of forms into data. There is a growing demand for In general, the person in charge needs to perform a predetermined task for the business partner described in each form. Identifiable is desired.

A business document processing device is disclosed that estimates a character string that overlaps a seal impression by collating character strings that exist in the vicinity of the seal impression with a database (Patent Document 1). This business document processing apparatus extracts character information that exists in the vicinity of the seal imprint area and part of the characters are unclear due to the seal imprint area as the seal imprint related information. The business document processing device identifies the attribute of the imprint-related information, refers to a customer database that stores information on the customer who stores the character string candidate, and based on the imprint-related information classified by attribute, Estimate the string that overlaps with the region and is ambiguous.

JP 2010-157107 A

Image processing apparatuses are required to more accurately identify an area containing the name of a business partner from an input image.

An object of the image processing device, the control method, and the control program is to make it possible to more accurately identify an area containing the name of a supplier from an input image.

An image processing apparatus according to an aspect of an embodiment includes a storage unit that stores a keyword related to the name of a business partner, an acquisition unit that acquires an input image, and an image that detects an image area including a seal or a logo in the input image. an area detection unit, a character string area detection unit that detects a character string area from an input image, and a candidate area detection unit that detects a candidate area containing a client name from the character string area based on a keyword; a customer area identification unit that identifies one customer area from among the plurality of candidate areas based on the distance from the image area when a plurality of candidate areas are detected by the candidate area detection unit; and an output control unit that outputs information about the area.

A control method according to an aspect of the embodiment is a control method for an image processing apparatus having a storage unit, in which the image processing apparatus stores a keyword related to the name of a business partner in the storage unit and acquires an input image. , Detect image areas that contain a seal or logo in the input image, detect character string areas from the input image, and detect candidate areas that include the name of the client from the character string areas based on the keyword. Then, when a plurality of candidate areas are detected, one supplier area is specified from among the plurality of candidate areas based on the distance from the image area, and information regarding the supplier area is output.

Further, a control program according to one aspect of the embodiment is a control program for a computer having a storage unit, stores a keyword related to the name of a business partner in the storage unit, acquires an input image, and uses a seal or Detect image regions that include logos, detect character string regions from the input image, detect candidate regions that include customer names from among the character string regions based on keywords, and detect multiple candidate regions. If detected, cause the computer to identify one customer area from among the plurality of candidate areas based on the distance from the image area, and output information about the customer area.

According to the present embodiment, the image processing apparatus, control method, and control program can more accurately identify an area containing the name of a business partner from an input image.

The objects and advantages of the invention may be realized and obtained by means of the elements and combinations particularly pointed out in the claims. Both the foregoing general description and the following detailed description are exemplary and explanatory, and are not limiting of the invention as claimed.

1 is a diagram showing a schematic configuration of an image processing system 1 according to an embodiment; FIG. It is a figure which shows an example of the data structure of an item table. 2 is a diagram showing a schematic configuration of a second storage device 210 and a second processing circuit 220; FIG. 4 is a flowchart showing the operation of image reading processing; 4 is a flowchart showing the operation of recognition processing; 6 is a schematic diagram showing an example of an input image 600; FIG. 9 is a flowchart showing an example of operation of area specifying processing; FIG. 8 is a schematic diagram showing an example of an input image 800; 9 is a schematic diagram showing an example of an input image 900; FIG. 9 is a flow chart showing an example of the operation of a first specifying process; 9 is a flow chart showing an example of the operation of a second specifying process; 12 is a schematic diagram showing an example of an input image 1200; FIG. FIG. 13 is a schematic diagram showing another example of the input image 1300; FIG. 3 is a block diagram showing a schematic configuration of another second processing circuit 230; FIG.

An image processing apparatus, a control method, and a control program according to one aspect of the present disclosure will be described below with reference to the drawings. However, it should be noted that the technical scope of the present invention is not limited to those embodiments, but extends to the invention described in the claims and equivalents thereof.

FIG. 1 is a diagram showing a schematic configuration of an image processing system 1 according to an embodiment. As shown in FIG. 1, the image processing system 1 has an image reading device 100 and an information processing device 200 .

The image reading device 100 is, for example, a scanner device. The image reading device 100 is connected to the information processing device 200 . The information processing device 200 is an example of an image processing device, such as a personal computer.

The image reading device 100 has a first interface device 101 , an imaging device 102 , a first storage device 110 and a first processing circuit 120 .

The first interface device 101 has an interface circuit conforming to a serial bus such as USB (Universal Serial Bus), and is electrically connected to the information processing device 200 to transmit and receive image data and various information. Also, instead of the first interface device 101, a communication device having an antenna for transmitting and receiving wireless signals and a wireless communication interface circuit for transmitting and receiving signals through a wireless communication line according to a predetermined communication protocol may be used. good. The predetermined communication protocol is, for example, a wireless LAN (Local Area Network).

The imaging device 102 has a reduction optical system type imaging sensor that includes imaging elements that are CCDs (Charge Coupled Devices) linearly arranged in the main scanning direction. Further, the imaging device 102 includes a light source for irradiating light, a lens for forming an image on an imaging element, and an A/D converter for amplifying and analog/digital (A/D)-converting an electric signal output from the imaging element. have a vessel. In the imaging device 102, the imaging sensor images a conveyed medium to generate and output an analog image signal, and the A/D converter A/D converts the analog image signal to digital input. Generate and output an image. The input image is a color multivalued image in which each pixel data is composed of a total of 24 bits of R (red) value, G (green) value, and B (blue) value represented by 8 bits for each color of RGB, for example. Incidentally, instead of the CCD, a CIS (Contact Image Sensor) of an equal-magnification optical system type having an imaging element of CMOS (Complementary Metal Oxide Semiconductor) may be used.

The first storage device 110 includes memory devices such as RAM (Random Access Memory) and ROM (Read Only Memory), fixed disk devices such as hard disks, or portable storage devices such as flexible disks and optical disks. The first storage device 110 also stores computer programs, databases, tables, etc. used for various processes of the image reading apparatus 100 . The computer program may be installed in the first storage device 110 from a computer-readable portable recording medium using a known setup program or the like. Examples of portable recording media include CD-ROMs (compact disk read only memory) and DVD-ROMs (digital versatile disk read only memory). Also, the first storage device 110 stores an input image and the like generated by the imaging device 102 .

The first processing circuit 120 operates based on a program stored in advance in the first storage device 110 . The first processing circuit 120 is, for example, a CPU (Control Processing Unit). As the first processing circuit 120, a DSP (digital signal processor), LSI (large scale integration), ASIC (Application Specific Integrated Circuit), FPGA (Field-Programming Gate Array), or the like may be used.

The first processing circuit 120 is connected to the first interface device 101, the imaging device 102, the first storage device 110, etc., and controls these units. The first processing circuit 120 performs medium reading control of the imaging device 102, data transmission/reception control with the information processing device 200 via the first interface device 101, and the like.

The information processing device 200 has a second interface device 201 , an input device 202 , a display device 203 , a second storage device 210 and a second processing circuit 220 . Each unit of the information processing apparatus 200 will be described in detail below.

The second interface device 201 has an interface circuit similar to that of the first interface device 101 of the image reading device 100 and connects the information processing device 200 and the image reading device 100 . Further, instead of the second interface device 201, a communication device having an antenna for transmitting and receiving wireless signals and a wireless communication interface circuit for transmitting and receiving signals through a wireless communication line according to a predetermined communication protocol such as a wireless LAN is provided. may be used.

The input device 202 has an input device such as a keyboard and a mouse, and an interface circuit for acquiring signals from the input device, and outputs a signal corresponding to a user's operation to the second processing circuit 220 .

The display device 203 is an example of an output unit. The display device 203 has a display made up of a liquid crystal, an organic EL (Electro-Luminescence), or the like, and an interface circuit that outputs image data to the display. The display device 203 displays various information on the display according to instructions from the second processing circuit 220 .

The second storage device 210 is an example of a storage unit, and includes a memory device similar to the first storage device 110 of the image reading device 100, a fixed disk device, a portable storage device, and the like. The second storage device 210 stores computer programs, databases, tables, etc. used for various processes of the information processing device 200 . The computer program may be installed in the second storage device 210 from a computer-readable portable recording medium such as CD-ROM, DVD-ROM, etc. using a known setup program or the like.

Further, in the second storage device 210, keywords related to names of business partners, item tables, etc. are stored in advance as data. The keywords related to the name of the business partner are characters that can be included in the name of the business partner, such as a corporation, a limited company, and the like. Details of the item table will be described later.

The second processing circuit 220 operates based on a program stored in advance in the second storage device 210 . The second processing circuit 220 is, for example, a CPU. A DSP, LSI, ASIC, FPGA, or the like may be used as the second processing circuit 220 .

The second processing circuit 220 is connected to the second interface device 201, the input device 202, the display device 203, the second storage device 210, etc., and controls these units. The second processing circuit 220 performs data transmission/reception control with the image reading device 100 via the second interface device 201, input control of the input device 202, display control of the display device 203, and the like.

FIG. 2 is a diagram showing an example of the data structure of an item table.

The item table stores a keyword and/or format related to each item in association with each item included in the form. The multiple items include items other than the name of the business partner, such as zip code, address, phone number, contact person, email address, bank account, date and document number. A keyword is a character that can be included in a character string indicating each item. A keyword for each item is an example of a second keyword. The format is a character string format that indicates each item. For example, if the item is a postal code, the format is set to "〒", "three digits", "-", and "four digits", and if the item is a phone number, the format is "2 A combination of "digit", "-", "four-digit number", "-", and "four-digit number" is set.

FIG. 3 is a diagram showing a schematic configuration of the second storage device 210 and the second processing circuit 220. As shown in FIG.

As shown in FIG. 3, the second storage device 210 stores an acquisition program 211, an image area detection program 212, a character string area detection program 213, a candidate area detection program 214, a first area detection program 215, and a second area identification program 216. , a customer area identification program 217 and an output control program 218 are stored. Each of these programs is a functional module implemented by software running on a processor. The second processing circuit 220 reads each program stored in the second storage device 210 and operates according to each read program. As a result, the second processing circuit 220 includes an acquisition unit 221, an image area detection unit 222, a character string area detection unit 223, a candidate area detection unit 224, a first area detection unit 225, a second area identification unit 226, a customer area It functions as a specifying unit 227 and an output control unit 228 .

FIG. 4 is a flow chart showing the operation of image reading processing by the image reading apparatus 100 . The operation of the image reading process will be described below with reference to the flowchart shown in FIG. The operation flow described below is executed mainly by the first processing circuit 120 in cooperation with each element of the image reading apparatus 100 based on a program stored in the first storage device 110 in advance.

First, the image capturing device 102 captures a form such as a bill, notification, or certificate containing a seal or logo as a document, generates an input image, and stores the input image in the first storage device 110 (step S101). .

Next, the first processing circuit 120 transmits the input image saved in the first storage device 110 to the information processing device 200 via the first interface device 101 (step S102), and ends the series of steps.

FIG. 5 is a flow chart showing the operation of recognition processing by the information processing apparatus 200 . The operation of the recognition process will be described below with reference to the flowchart shown in FIG. The operation flow described below is executed mainly by the second processing circuit 220 in cooperation with each element of the information processing device 200 based on a program stored in the second storage device 210 in advance.

First, the acquisition unit 221 acquires an input image from the image reading device 100 via the second interface device 201, and stores it in the second storage device 210 (step S201).

FIG. 6 is a schematic diagram showing an example of an input image 600. As shown in FIG.

As shown in FIG. 6, an input image 600 includes a seal 605, a postal code 606, an address 607, a company name 608, a telephone number 609, a business partner name 610, an address 611, a bank account 612, a bank account 613, and the like. include.

Next, the image area detection unit 222 divides the input image into a plurality of partial areas (step S202). The image region detection unit 222 divides the input image into a plurality of partial regions, for example, by dividing the input image into a predetermined number each in the vertical direction and/or the horizontal direction.

In the example shown in FIG. 6, an input image 600 is divided into two each in the vertical direction and the horizontal direction into four partial regions 601-604. The information processing apparatus 200 detects the name of the business partner based on the positional relationship with the image area containing the seal or logo in the input image in which the form is captured. In a general form, the name of the customer is written in the upper left, upper right, lower left, or lower right part, and a seal or logo is written in the vicinity thereof. Therefore, the information processing apparatus 200 divides the input image into two in the vertical direction and the horizontal direction so that the name of the business partner and the seal or logo are included in the same partial area. can do.

Next, the image area detection unit 222 detects a seal area containing a seal or a logo area containing a logo in the input image (step S203). Hereinafter, the seal area and the logo area may be collectively referred to as graphic areas. A graphic area is an example of an image area that contains a stamp or logo. The image area detection unit 222 detects a predetermined figure area for each of a plurality of partial areas within the input image. The seal stamp area includes a red seal stamp area including a red stamp and/or a black seal stamp area including a black stamp.

To detect the red stamp area, the image area detection unit 222 first converts the RGB (red, green, blue) values of each pixel in the input image into HSL (hue, saturation, luminance) values. Next, the image region detection unit 222 detects pixels whose hue value is less than a first threshold value (eg, 30) and pixels whose hue value is greater than a second threshold value (eg, 330) (that is, pixels having colors ranging from orange to reddish purple). ). Next, the image region detection unit 222 calculates the sum of the value obtained by subtracting the G value from the R value and the value obtained by subtracting the B value from the R value as the red intensity for each extracted pixel. Note that the image region detection unit 222 may use the S (saturation) value of each extracted pixel as the red intensity.

Next, the image area detection unit 222 determines pixels whose calculated red intensity is equal to or greater than a third threshold value (for example, 32) to be stamp pixels. Next, in the input image, an area including an area where the extracted seal stamp pixels are connected is detected as a red seal stamp area in which the seal is shown. The image area detection unit 222 labels the areas surrounded by the seal stamp pixels that are adjacent to each other as one group and defines them as a red seal stamp area in which the seal is captured. Note that the image area detection section 222 may detect only groups whose size (area) is within a predetermined range as red seal stamp areas. The predetermined range is set in advance to a range that can be taken by a general seal. Further, the image area detection section 222 may detect only groups having a predetermined shape among the groups as red seal stamp areas. The predetermined shape is set in advance to a shape that can be taken by a general stamp, such as a circle or a square. For example, the image region detection unit 222 calculates a normalized cross-correlation value between each group and a predetermined shape, and only when the calculated normalized cross-correlation value is equal to or greater than a predetermined value, detects that group as a red seal region. do. Also, the image area detection section 222 may detect the circumscribed rectangle of each group as the red seal stamp area.

In addition, the image area detection unit 222 first converts the RGB values of each pixel in the input image into YUV (luminance, color difference) values in order to detect the logo area or the black seal area that does not overlap with the character string. do. Next, the image region detection unit 222 extracts pixels whose luminance values are less than a fourth threshold value (for example, 128) as black pixels. Next, the image area detection unit 222 puts together areas surrounded by black pixels that are adjacent to each other as one group by labeling, and detects a group having a predetermined shape from among each group as a logo area or a black seal area. Note that the image area detection section 222 may detect only groups whose size is within a predetermined range as the logo area or the black seal area. Also, the image area detection section 222 may detect the circumscribed rectangle of each group as a logo area or a black seal area.

In addition, in order to detect a black stamp area that overlaps the character string, the image area detection unit 222 uses the vertical pixel position in the input image as a class, and calculates the number of black pixels at each pixel position in the vertical direction. Generate a histogram of frequencies. The image area detection unit 222 determines that the number of consecutive classes whose frequencies are equal to or greater than a fifth threshold (e.g., 10) is within a predetermined range (e.g., equal to or greater than the number of pixels corresponding to 15 mm and equal to or less than the number of pixels corresponding to 30 mm). A certain vertical region is detected as a seal candidate line. Generally, in a horizontally written form, characters are arranged in the horizontal direction, so there is a blank space between lines in which the characters are written. However, in the area where the seal is pressed, the blank is removed because the seal exists between the lines. Therefore, there is a high possibility that a seal exists in an area where classes having a certain size of frequency are continuous to some extent in the histogram.

Next, the image area detection unit 222 puts together areas surrounded by mutually adjacent black pixels in the seal candidate line into one group by labeling. The image area detection unit 222 detects a group whose size is within a predetermined range and whose aspect ratio is within a predetermined aspect ratio range as a black seal stamp area. The predetermined range is set, for example, to a range equal to or greater than the number of pixels corresponding to 15 mm and equal to or less than the number of pixels corresponding to 30 mm. The aspect ratio range is set to, for example, 2/3 or more and 3/2 or less. In addition, the image area detection section 222 may detect the circumscribing rectangle of each group as the black seal stamp area.

In the example shown in FIG. 6, the area including the stamp 605 within the partial area 601 is detected as the figure area.

Next, the image area detection unit 222 determines whether or not a graphic area has been detected in the input image (step S204).

If no graphic area is detected in the input image, the supplier area identification unit 227 extracts from the input image based on the item name corresponding to the supplier name and the positional relationship with respect to the item name included in the input image. A supplier area is detected (step S205). The customer area is an area that contains a character string indicating the name of the customer in the input image. The customer area identification unit 227 uses a known OCR (Optical Character Recognition) technique to detect characters from the input image, and detects a plurality of characters having the first positional relationship as a character string. A plurality of characters having the first positional relationship are, for example, characters that overlap each other in the vertical direction and that are consecutive while being adjacent to each other within a certain range in the horizontal direction. The certain range is set to, for example, the number of pixels corresponding to the margin. The supplier area identification unit 227 extracts a character string including an item name corresponding to the supplier name from among the detected character strings. The item name corresponding to the name of the business partner is, for example, "company name".

The supplier area identification unit 227 detects a character string having the second positional relationship with respect to the extracted character string as a character string indicating the name of the supplier. The second positional relationship is preset, including direction (eg, right side, bottom side, bottom right side) and distance (eg, within pixels corresponding to 30 mm). The supplier area identification unit 227 detects an area including the detected character string as a supplier area. The supplier area identification unit 227 detects, for example, the circumscribed rectangle of the detected character string as the supplier area. Next, the supplier area identification unit 227 shifts the process to step S208.

On the other hand, if a figure area is detected in the input image, the character string area detection unit 223 removes the stamp component or the logo component from the input image (step S206).

When a red seal stamp area is detected in the input image, the character string area detection unit 223 first generates an expanded image by expanding the white component of the input image, and calculates the pixel value of each pixel in the expanded image ( R value, G value, B value) are determined as background information for each pixel in the corresponding input image. That is, the background information corresponding to the pixel of interest in the input image is set to the pixel value of the pixel with the highest luminance among the surrounding pixels of the pixel of interest. Next, the character string region detection unit 223 calculates the R value S _r , G value S _g , and B value S _b of each pixel in the red stamp region according to the following equations (1) to (3). Convert to D _r , D _g , D _b .
D _r = (B _r - S _r ) x (percentage of red) + S _r (1)
D _g = (B _g - S _g ) x (percentage of red) + S _g (2)
D _b = (B _b - S _b ) x (percentage of red) + S _b (3)

The red ratio is calculated, for example, by the following formula (4).
(red ratio)={(red intensity of pixel of interest)−β}/α (4)
where α is a factor (eg, 96) and is preset based on the red intensity bin width. Also, β is an offset value (eg, 32), which is preset based on the range of red intensities. The red ratio is set within a range of 0 or more and 1 or less. That is, when the value calculated by Equation (4) is 0 or less, the red color ratio is corrected to 0, and when the value calculated by Equation (4) is 1 or more, the red color ratio is corrected to 1. .

The character string area detection unit 223 synthesizes the pixel value of each pixel in the red stamp area with the background information of each pixel according to the red ratio of each pixel, thereby leaving the character component in the red stamp area. Only the components of the red stamp can be removed satisfactorily.

When a logo area or a black seal stamp area that does not overlap with a character string is detected in the input image, the character string area detection unit 223 first detects an area that exists around the logo area or the black seal area and is close to white. Identify a pixel. The character string area detection unit 223 identifies pixels that are within a certain range from the logo area or the black seal area and whose pixel values (R value, G value, B value) are equal to or greater than a threshold value (eg, 200). The character string area detection unit 223 replaces the pixel value of each pixel in the logo area or black seal area with the pixel value of the specified pixel, thereby removing the stamp component or the logo component.

Further, the character string area detection unit 223 does not remove the black seal component from the black seal stamp area that overlaps the character string in the input image.

Next, the second processing circuit 220 executes region identification processing (step S207). In the region identifying process, the second processing circuit 220 identifies the client region from the input image based on the positional relationship with the graphic region. The details of the area specifying process will be described later.

Next, the output control unit 228 outputs information about the supplier area detected in step S205 or the supplier area specified in the area specifying process in step S207 (step S208), and ends the series of steps. . The output control unit 228 outputs information on the customer area by displaying it on the display device 203 . Note that the output control unit 228 may output the information on the supplier area by transmitting it to another information processing device via the second interface device 201 . The output control unit 228 outputs a character string indicating the name of the supplier in the input image, which is included in the supplier area, for example, as information about the supplier area. Note that the output control unit 228 may output the input image and the position information indicating the coordinates of the supplier area in the input image as the information about the supplier area. Further, the output control unit 228 may output an image or the like obtained by cutting out the customer area as the information about the customer area.

It should be noted that the process of step S206 may be omitted, and the second processing circuit 220 may specify the supplier area without removing the stamp component or the logo component from the input image.

FIG. 7 is a flow chart showing an example of the operation of area specifying processing. The operation flow shown in FIG. 7 is executed in step S207 of the flowchart shown in FIG.

The processing of steps S301 to S310 shown in the flowchart of FIG. 7 is executed for each partial area in which a graphic area is detected among the plurality of partial areas.

In the example shown in FIG. 6, the processing of steps S301 to S310 is performed only for the partial area 601 in which the figure area including the stamp 605 is detected.

First, the character string area detection unit 223 detects a character string area from a partial area to be processed of the input image (step S301). The character string area detection unit 223 uses a known OCR technique to detect characters from the input image, and detects a plurality of characters having the first positional relationship as character strings. The character string area detection unit 223 detects an area including the detected character string as a character string area. The character string region detection unit 223 detects, for example, a circumscribing rectangle of the detected character string as a character string region.

In the example shown in FIG. 6, in the partial area 601, the area containing the postal code 606, address 607, company name 608, telephone number 609, business partner name 610, address 611, bank account 612, and bank account 613 are characters. Detected as a column region.

Next, the candidate area detection unit 224 selects the name of the customer from the character string area detected by the character string area detection unit 223 based on the keyword related to the name of the customer stored in the second storage device 210 . is detected (step S302). The candidate area detection unit 224 detects, as a candidate area, a character string area including a keyword related to the name of the business partner among the detected character string areas.

In the example shown in FIG. 6, the company's name 608 and the business partner's name 610 include the keyword "corporation", and character string areas containing the company's name 608 and the business partner's name 610 are detected as candidate areas. be done.

Next, the first area detection unit 225 detects the first area from among the character string areas detected by the character string area detection unit 223 based on the keyword or format for each item stored in the item table ( step S303). The first area is an area containing any one of the items stored in the item table. The first area detection unit 225 detects, as a first area, a character string area that includes a keyword related to any item or a character string area that satisfies a format related to any item from among the detected character string areas. .

In the example shown in FIG. 6, character string areas including postal code 606, address 607, telephone number 609, address 611, bank account 612 and bank account 613 are detected as the first area.

Next, the second area identification unit 226 identifies second areas that have not been detected as candidate areas or first areas from among the character string areas detected by the character string area detection unit 223 (step S304). The second area specifying unit 226 specifies, as a second area, a character string area that is not detected as a candidate area and not detected as a first area, among the detected character string areas.

In the example shown in FIG. 6, all character string areas are detected as candidate areas or first areas, and second areas are not specified.

FIG. 8 is a schematic diagram showing an example of an input image 800 including the second area.

As shown in FIG. 8, an input image 800 includes a seal 805, a date 806, a document number 807, a supplier name 808, a person in charge 809, an e-mail address 810, a postal code 811, an address 812, a telephone number 813, and the like. is An input image 800 is divided into four partial areas 801 to 804, and an area including a stamp 805 in the partial area 802 is detected as a graphic area. Also, areas containing the date 806, document number 807, supplier name 808, person in charge 809, mail address 810, postal code 811, address 812, and telephone number 813 are detected as character string areas in the partial area 802. . The supplier name 808 does not include the keyword “corporation”, and no candidate area is detected from the partial area 801 . Also, a character string area including date 806, document number 807, person in charge 809, mail address 810, zip code 811, address 812, and telephone number 813 is detected as the first area in partial area 801. FIG. Therefore, the character string area containing the name 808 of the business partner that was not detected as the candidate area and was not detected as the first area is specified as the second area.

Next, the supplier area identification unit 227 determines whether or not the candidate area is detected by the candidate area detection unit 224 (step S305).

If no candidate area is detected by the candidate area detection unit 224, the supplier area identification unit 227 executes first identification processing (step S306). In the first identification process, the supplier area identification unit 227 identifies the most promising candidate area, which is the most promising candidate area within the partial area, based on the second area. Details of the first identification process will be described later.

On the other hand, if a candidate area is detected, the client area identification unit 227 determines whether a plurality of candidate areas have been detected by the candidate area detection unit 224 (step S307).

If a plurality of candidate areas are not detected, that is, if the number of detected candidate areas is one, the supplier area identification unit 227 executes second identification processing (step S308). In the second identification process, the supplier area identification unit 227 identifies the most likely candidate area based on the positional relationship between the candidate area and the second area. Details of the second specifying process will be described later.

On the other hand, when a plurality of candidate areas are detected, that is, when a plurality of candidate areas are detected, the supplier area identification unit 227 selects a candidate area from among the plurality of candidate areas based on the distance from the graphic area. One most promising candidate area is identified (step S309).

The supplier area identification unit 227 calculates the distance between each of the plurality of candidate areas and all the graphic areas in the partial area. The customer area identification unit 227 identifies a combination of edge positions (upper left, lower left, upper right, or lower right) located closest to each other in the input image in the candidate area and the graphic area, and determines the identified edge position. The Euclidean distance (the number of pixels corresponding to the linear distance) between the regions is calculated as the distance between the regions. Note that the client region identification unit 227 calculates the Manhattan distance between the identified end positions (the sum of the absolute value of the horizontal coordinate difference and the absolute value of the vertical coordinate difference) as the distance between the regions. You may Further, the supplier area identification unit 227 may calculate the Euclidean distance or the Manhattan distance between the centroid positions of the areas as the distance between the areas. Further, when the candidate area and the graphic area overlap (when the number of overlapping pixels is equal to or greater than a predetermined number), the supplier area identification unit 227 may set the distance between the areas to zero.

The customer area identification unit 227 identifies the graphic area with the shortest distance as the graphic area located closest to each candidate area for each of the plurality of candidate areas, and identifies the distance between the areas. The supplier area identification unit 227 identifies the combination of the candidate area and the graphic area having the smallest distance between the areas from among the combinations of the identified candidate area and the graphic area, and identifies the candidate area associated with the identified combination as the most promising candidate area. Identify as

In the example shown in FIG. 6, a candidate area including the company's name 608 and a candidate area including the supplier's name 610 are detected, and the supplier's name 610 is included at a short distance from the graphic area including the stamp 605. A candidate region is identified as the most likely candidate region.

When the image processing system 1 is used by a company that digitizes forms such as invoices received from business partners, the image processing system 1 needs to identify and manage the name of the business partner. Such invoices are likely to include the name of the company and the name of the business partner as the company name. However, it is possible that the seal or logo of the supplier is included near the name of the business partner who issued the form, and the seal or logo is not included near the name of the company to which the form is sent. highly sexual. Therefore, when a plurality of candidate areas are detected, the image processing system 1 accurately detects the area containing the customer name by specifying the candidate area closer to the graphic area as the most likely candidate area. be able to. Thereby, the image processing system 1 can suppress erroneous detection of a name other than the name of the business partner as the name of the business partner.

Next, the supplier area identification unit 227 determines whether or not the processing of steps S301 to S309 has been completed for all partial areas in which graphic areas have been detected (step S310). If the graphic area has been detected and partial areas that have not yet been processed remain, the customer area identification unit 227 returns the process to step S301 and repeats the processes of steps S301 to S309.

On the other hand, when the processing has been completed for all partial areas, the supplier area identification unit 227 determines whether or not the most promising candidate area has been identified from any partial area (step S311).

If the most promising candidate area is not identified from any of the partial areas, the supplier area identification unit 227 identifies the transaction from the input image based on the item name corresponding to the supplier name and the positional relationship from the item name. The destination area is detected (step S312), and the series of steps ends. The supplier area identification unit 227 detects the supplier area from the input image in the same manner as in step S205 of FIG.

On the other hand, if the most likely candidate area is identified from any of the partial areas, the customer area identification unit 227 identifies the customer area from the identified most likely candidate areas (step S313), and then performs a series of steps. exit. The customer area identification unit 227 identifies the customer area based on the distance between each most likely candidate area identified for each partial area in which the figure area is detected and the figure area corresponding to each most likely candidate area. Identify. A graphical area corresponding to each most likely candidate area is a graphical area that is included in the partial area that includes each most likely candidate area and that is positioned closest to each most likely candidate area within that partial area. The supplier area identification unit 227 calculates the distance between each most likely candidate area and each of all graphic areas in the partial area in the same manner as in step S309. The customer area identification unit 227 identifies the graphic area with the shortest distance for each most likely candidate area as the most likely graphical area positioned closest to each most likely candidate area, and identifies the distance between the areas. The customer area identification unit 227 identifies the combination of the most promising candidate area and the graphic area having the smallest distance between the areas among the identified combinations of the most likely candidate area and the graphic area, and identifies the most likely candidate for the identified combination. Identifies a region as a trading partner region.

In this way, the customer area specifying unit 227 selects the most likely candidate area that is the shortest from the corresponding figure area among the most likely candidate areas specified for each partial area in which the graphic area is detected as the customer area. Identify.

In the example shown in FIG. 6, the most likely candidate area including the supplier name 610 is detected in the partial area 601, and the most likely candidate area is not detected in the other partial areas 602-604. Therefore, the most likely candidate area containing the customer name 610 is identified as the customer area.

FIG. 9 is a schematic diagram showing an example of an input image 900 including a plurality of most likely candidate regions.

As shown in FIG. 9, an input image 900 includes a supplier's logo 905, company's name 906, supplier's seal stamp 907, supplier's name 908, and the like. An input image 900 is divided into four sub-regions 901-904. Then, in the partial area 901, an area containing the client's logo 905 is detected as a graphic area, and an area containing the company's name 906 is detected as the most probable candidate area. Also, in the partial area 902, an area including the customer's seal 907 is detected as a graphic area. Regions containing the business partner's name 908 are detected as the most likely candidate regions. The distance between the most likely candidate area containing the supplier's name 908 and the graphical area containing the supplier's seal 907 is the distance between the most likely candidate area including the company's name 906 and the graphical area including the supplier's logo 905. less than the distance between Therefore, the most likely candidate area containing the supplier name 908 is identified as the supplier area.

Depending on the form, the logo of the customer, the seal of the person in charge of the customer, or the like may be included in the same partial area as the name of the company. Even in such cases, there is a high possibility that the seal of the business partner is stamped near the name of the business partner, and the distance between the name of the business partner and the seal of the business partner is the same as the name of the company and the logo of the business partner. Or there is a high possibility that it is smaller than the distance between the seal of the person in charge of the business partner. When the most likely candidate area is detected in two or more partial areas, the image processing system 1 identifies the most likely candidate area that is close to the corresponding graphic area as the customer area, thereby identifying the area containing the name of the customer. can be detected with high accuracy. Thereby, the image processing system 1 can suppress erroneous detection of a name other than the name of the business partner as the name of the business partner.

FIG. 10 is a flow chart showing an example of the operation of the first specifying process. The flow of operations shown in FIG. 10 is executed in step S306 of the flowchart shown in FIG.

First, the customer area identification unit 227 determines whether the second area has been identified by the second area identification unit 226 (step S401).

If the second area is not specified, the customer area specifying unit 227 does not specify the most promising candidate area from the partial area (step S402), and ends the series of steps.

On the other hand, when the second area has been identified, the supplier area identification unit 227 determines whether or not a plurality of second areas have been identified by the second area identification unit 226 (step S403).

If a plurality of second areas are not specified, that is, if the number of specified second areas is one, the supplier area specifying unit 227 specifies the second area as the most likely candidate area (step S404), the series of steps ends.

In the example shown in FIG. 8, no candidate area is detected, and a second area containing the supplier name 808 is identified, and this second area is identified as the most likely candidate area.

In general, there are company names that do not contain general keywords such as "stock company" and "limited company", so it is difficult to completely detect the name of the customer based on only the keywords. When the information processing apparatus 200 fails to detect the name of the customer based on the keyword, the information processing apparatus 200 presumes that the name of the customer is a character string that does not correspond to items such as an address and a telephone number included in a general form. . As a result, the information processing apparatus 200 can accurately identify the name of the business partner that does not include the keyword.

On the other hand, when a plurality of second areas are identified, that is, when the number of identified second areas is plural, the customer area identification unit 227 determines the plurality of second areas based on the distance from the graphic area. (step S405), and the series of steps ends. The supplier area identification unit 227 calculates the distance between each of the plurality of second areas and all the graphic areas in the partial area in the same manner as in the process of step S309 in FIG. The customer area identification unit 227 identifies the graphic area with the shortest distance as the graphic area positioned closest to each second area for each of the plurality of second areas, and identifies the distance between the areas. The supplier area identification unit 227 identifies the combination of the second area and the graphic area having the smallest distance between the areas among the identified combinations of the second area and the graphic area, and determines the second area associated with the identified combination as the largest. Identify as a strong candidate region.

As mentioned above, when the invoice received from a business partner contains the name of the company and the name of the business partner, the business partner's seal or logo is included near the name of the business partner, and the name of the company It is highly probable that no seal or logo is included in the vicinity of . Therefore, when a plurality of second areas are identified, the image processing system 1 identifies the second area closer to the graphic area as the most likely candidate area, thereby accurately identifying the area containing the name of the customer. can be detected. Thereby, the image processing system 1 can suppress erroneous detection of a name other than the name of the business partner as the name of the business partner.

Note that if the candidate area detection unit 224 does not detect a candidate area in any of the partial areas, only the second area identified by the second area identification unit 226 is identified as the most probable candidate area. Therefore, in step S313 of FIG. 7, the customer area identifying section 227 identifies the customer area from the second areas identified by the second area identifying section 226. FIG. In particular, when the second area specifying unit 226 specifies a plurality of second areas for each partial area in which the image area is detected, the supplier area specifying unit 227 selects the corresponding A second area having the smallest distance from the graphic area is identified as the supplier area.

As described above, depending on the form, the logo of the customer, the seal of the person in charge of the customer, or the like may be included in the same partial area as the name of the company. When a plurality of second areas are identified as the most promising candidate areas in two or more partial areas, the image processing system 1 identifies a second area close to the corresponding graphic area as a customer area, thereby identifying the customer's A region containing a name can be detected with high accuracy. Thereby, the image processing system 1 can suppress erroneous detection of a name other than the name of the business partner as the name of the business partner.

FIG. 11 is a flow chart showing an example of the operation of the second specifying process. The flow of operations shown in FIG. 11 is executed in step S308 of the flowchart shown in FIG.

First, the supplier area identification unit 227 determines whether the distance between the candidate area detected by the candidate area detection unit 224 and the graphic area located closest to the candidate area is equal to or less than the distance threshold. Determine (step S501). The supplier area identification unit 227 calculates the distance between the candidate area and each of the graphic areas in the partial area in the same manner as in step S309 of FIG. The customer area identification unit 227 identifies the graphic area with the shortest distance as the graphic area positioned closest to the candidate area, and identifies the distance between the areas. The distance threshold is set, for example, to the number of pixels corresponding to a predetermined number of lines (eg, 5 lines).

If the distance between the candidate area and the corresponding graphic area is equal to or less than the distance threshold, the customer area identification unit 227 identifies the candidate area as the most likely candidate area (step S502), and terminates the series of steps. do. That is, if there is a seal or logo in the vicinity of the name of the supplier that includes the keyword in the form, the area containing the name of the supplier is specified as the most likely candidate area.

On the other hand, if the distance between the candidate area and the corresponding graphic area is greater than the distance threshold, the supplier area identification unit 227 determines whether there is a second area whose distance from the graphic area is smaller than that of the candidate area. is determined (step S503). 10, for each second area identified by the second area identifying unit 226, the supplier area identifying unit 227 identifies the graphic area closest to each second area, and , the distance between each second region and each graphic region. If the distance calculated for any of the second regions is smaller than the distance calculated for the candidate regions in step S501, the supplier region identification unit 227 determines that there is a second region whose distance from the graphic region is smaller than that of the candidate region. judge. On the other hand, if the distance calculated for the candidate area in step S501 is equal to or less than the distance calculated for all the second areas, the supplier area identification unit 227 determines that the second area whose distance from the graphic area is smaller than that of the candidate area is Determine that it does not exist.

If there is no second area whose distance from the graphic area is smaller than that of the candidate area, the supplier area identifying unit 227 identifies that candidate area as the most likely candidate area (step S502), and terminates the series of steps.

FIG. 12 is a schematic diagram showing an example of an input image 1200 in which candidate areas are separated from graphic areas.

As shown in FIG. 12, an input image 1200 includes a customer seal 1205, a customer name 1206, and the like. An input image 1200 is divided into four partial areas 1201 to 1204. In the partial area 1202, an area containing the customer's seal stamp 1205 is detected as a graphic area, and an area containing the customer's name 1206 is detected as a candidate area. . However, the partial area 1202 does not contain items other than the items stored in the item table, and the second area is not specified within the partial area 1202 . The distance between the candidate area containing the business partner's name 1206 and the graphic area containing the business partner's stamp 1205 is greater than the distance threshold. However, since the second area is not identified within the partial area 1202, the candidate area containing the customer name 1206 is identified as the most likely candidate area.

In this way, even if the seal or logo is far from the company name in the form, if the company name located closest to the seal or logo contains the keyword, the region containing the company name is the most suitable. It is identified as a strong candidate region.

On the other hand, if there is a second area whose distance from the graphic area is smaller than that of the candidate area, the supplier area identification unit 227 determines whether the distance between the second area and the graphic area is equal to or less than a predetermined distance. (Step S504). Among the second areas in the partial area, the supplier area identification unit 227 determines the second area that is the closest to the closest graphic area, and the second area and the second area that are closest to the second area. Determine the distance between the graphic regions located at . If the identified distance is less than or equal to the predetermined distance, supplier area identification unit 227 determines that the distance between the second area and the graphic area is less than or equal to the predetermined distance. It is determined that the distance between the two areas and the graphic area is greater than a predetermined distance.

If the distance between the second area and the graphic area is greater than the predetermined distance, the supplier area identification unit 227 identifies the candidate area as the most likely candidate area (step S502), and terminates the series of steps. That is, even if a company name that does not contain a keyword exists closer to the seal or logo than the company name that contains the keyword in the form, if the company name that does not contain the keyword is farther from the seal or logo than the company name that contains the keyword, the keyword are identified as the most likely candidate regions.

On the other hand, if the distance between the second area and the graphic area is equal to or less than the predetermined distance, the supplier area identification unit 227 identifies the second area as the most promising candidate area (step S505), and terminates the series of steps. . The supplier area identification unit 227 identifies, among the second areas in the partial area, the second area having the shortest distance from the closest graphic area as the most likely candidate area.

FIG. 13 is a schematic diagram showing an example of an input image 1300 in which candidate areas are separated from graphic areas.

As shown in FIG. 13, the input image 1300 includes the seal 1305 of the client, the name of the company 1306, the name 1307 of the client, and the like. An input image 1300 is divided into four sub-regions 1301-1304. Then, in the partial area 1301, an area containing the customer's seal 1305 is detected as a graphic area, an area including the company's name 1306 is specified as a candidate area, and an area including the customer's name 1307 is detected as a second area. be done. The distance between the candidate area containing the company's name 1306 and the graphic area containing the client's seal 1305 is greater than the distance threshold. On the other hand, the distance between the second area containing the supplier's name 1307 and the graphic area containing the seal 1305 is smaller than the distance between the candidate area containing the company's name 1306 and the graphic area containing the seal 1305, and , is less than or equal to a predetermined distance. Therefore, the second area including the customer name 1307 is identified as the most likely candidate area.

In this way, if the seal or logo is far from the company name that includes the keyword in the form, and if there is a company name that is closer to the seal or logo than the company name and does not include the keyword, the keyword is Regions containing non-containing company names are identified as the most likely candidate regions. Therefore, the image processing system 1 can accurately detect an area including the name of the supplier even if the name of the supplier does not include the keyword and the name of the company includes the keyword. Thereby, the image processing system 1 can suppress erroneous detection of a name other than the name of the business partner as the name of the business partner.

Note that the process of step S501 may be omitted. In that case, regardless of the distance between the candidate area and the graphic area, the customer area identification unit 227 determines the second area whose distance from the graphic area is smaller than the distance between the candidate area and the graphic area and is equal to or less than a predetermined distance. as a trading partner area.

Also in step S309 of FIG. 7, as in the second identification process, the supplier area identification unit 227 determines that the distance from the graphic area is smaller than the distance between the candidate area and the graphic area and is equal to or less than the predetermined distance. A second region may be identified as the most likely candidate region. If a plurality of candidate areas are detected, the client area identification unit 227 identifies a candidate area having the shortest distance from the nearest graphic area among the plurality of candidate areas. If there is a second area whose distance from the graphic area is smaller than the identified candidate area, and the distance between the second area and the graphic area is equal to or less than a predetermined distance, the supplier area identification unit 227 identifies the second area. Two regions are identified as the most likely candidates. Note that if there are a plurality of second areas whose distance from the graphic area is smaller than the distance between the identified candidate area and the graphic area and is equal to or less than a predetermined distance, the customer area identification unit 227 determines that the distance is the shortest. A second smaller region is identified as the most likely candidate region.

Note that the information processing apparatus 200 may perform the recognition processing and the region specifying processing collectively without dividing the input image into a plurality of partial regions. In that case, the process of step S202 is omitted in the recognition process of FIG. 5, and in step S203, the image area detection unit 222 detects a graphic area from the entire input image. Also, in the area specifying process of FIG. 7, the processes of steps S301 to S309 are performed on the entire input image, and the process of step S310 is omitted. In step S313 of FIG. 7, the customer area identification unit 227 selects the most promising candidate area identified in step S309 of FIG. 7, steps S404 and S405 of FIG. 10, or steps S502 and S505 of FIG. Identify.

Further, the information processing apparatus 200 may specify the supplier area without detecting the first area and specifying the second area. In that case, the processing of steps S303 and S304 in FIG. 7 is omitted. If no candidate area is detected in step S305, the supplier area identification unit 227 shifts the process to step S312, and based on the item name corresponding to the supplier name and the positional relationship from the item name, Detect supplier regions from the input image. In step S307, if the number of candidate areas detected is one, the supplier area identification unit 227 identifies the detected candidate area as the most likely candidate area. In this case, in step S313, the supplier area identification unit 227 identifies, as the supplier area, the candidate area with the shortest distance from the corresponding graphic area among the candidate areas identified for each partial area in which the graphic area is detected. do.

Further, in step S313, the supplier area identification unit 227 may preferentially identify the most promising candidate area detected as the candidate area as the supplier area. In that case, the supplier area identification unit 227 determines whether or not there is a most likely candidate area detected as a candidate area. If there is a most likely candidate area detected as a candidate area, the supplier area identifying unit 227 selects the most likely candidate area that is the shortest from the corresponding graphic area, among the most likely candidate areas detected as candidate areas. as a trading partner area. On the other hand, if there is no most likely candidate area detected as a candidate area, the supplier area identifying unit 227 selects the most likely candidate area identified as the second area, which has the smallest distance from the corresponding graphic area. Identify a strong candidate area as a trading partner area.

As described above, when a plurality of graphic regions are detected and a plurality of candidate regions corresponding to each graphic region are detected, the customer region identification unit 227 determines the distance from the corresponding graphic region among the plurality of candidate regions. The candidate region with the smallest is specified as the trading partner region. Further, when a plurality of graphic regions are detected, the candidate region corresponding to each graphic region is not detected, and the second region corresponding to any of the graphic regions is identified, the supplier region identification unit 227 The specified second area is specified as a supplier area. In particular, when a plurality of second regions corresponding to each graphic region are identified, the supplier region identification unit 227 selects the second region having the smallest distance from the corresponding graphic region among the plurality of second regions. identified as the destination region.

As described in detail above, the information processing apparatus 200, when a plurality of candidate areas for the name of the customer is detected based on the keyword, selects one customer based on the distance from the graphic area containing the seal or logo. Identify areas. As a result, the information processing apparatus 200 can more accurately identify the region containing the name of the business partner from the input image.

As a result, the information processing apparatus 200 no longer accepts correction of the supplier's name from the user using the input device 202, making it possible to reduce the processing load of the processor in the process of handling the input image. Further, when the information processing device 200 transmits information about the name of the client to another information processing device, the information processing device 200 does not receive a request to correct the name of the client from the other information processing device. It has become possible to reduce the amount of communication between other information processing apparatuses.

The image processing system 1 has made it possible to reduce the work load of the person in charge in a company that automates the form input work using OCR and RPA (Robotic Process Automation) technology. In particular, the image processing system 1 makes it possible to specify the names of business partners with high accuracy when digitizing forms such as invoices and automating operations such as payment requests, thereby improving user convenience. became possible.

FIG. 14 is a block diagram showing a schematic configuration of a second processing circuit 230 in an information processing apparatus according to another embodiment.

The second processing circuit 230 executes recognition processing instead of the second processing circuit 220 . The second processing circuit 230 includes an acquisition circuit 231, an image area detection circuit 232, a character string area detection circuit 233, a candidate area detection circuit 234, a first area detection circuit 235, a second area identification circuit 236, and a supplier area identification circuit 237. and an output control circuit 238 and the like.

The acquisition circuit 231 is an example of an acquisition section and has the same function as the acquisition section 221 . The acquisition circuit 231 acquires an input image from the image reading device 100 via the second interface device 201 and stores it in the second storage device 210 .

The image area detection circuit 232 is an example of an image area detection section and has the same function as the image area detection section 222 . The image area detection circuit 232 reads the input image from the second storage device 210 , detects graphic areas in the input image, and stores the detection result in the second storage device 210 .

The character string area detection circuit 233 is an example of a character string area detection section and has the same function as the character string area detection section 223 . The character string area detection circuit 233 reads the input image from the second storage device 210 , detects character string areas from the input image, and saves the detection result in the second storage device 210 .

The candidate area detection circuit 234 is an example of a candidate area detection section and has the same function as the candidate area detection section 224 . The candidate area detection circuit 234 reads out the detection result of the character string area and the keyword related to the name of the customer from the second storage device 210, detects the candidate area from the character string area, and stores the detection result in the second storage device 210. save.

The first area detection circuit 235 is an example of a first area detection section and has the same function as the first area detection section 225 . The first area detection circuit 235 reads the detection result of the character string area and the item table from the second storage device 210, detects the first area from the character string area, and saves the detection result in the second storage device 210. .

The second area identifying circuit 236 is an example of a second area identifying section and has the same function as the second area identifying section 226 . The second area identification circuit 236 reads the detection result of the character string area and the detection result of the first area from the second storage device 210, identifies the second area from the character string area, and stores the identification result in the second storage device. 210.

The supplier area identification circuit 237 is an example of a supplier area identification unit and has the same function as the supplier area identification unit 227 . The supplier area identification circuit 237 reads the detection result of the candidate area, the detection result of the first area, and the identification result of the second area from the second storage device 210, identifies the supplier area, and stores the identification result as the second area. Store in storage device 210 .

The output control circuit 238 is an example of an output control section and has the same function as the output control section 228 . The output control circuit 238 reads out the specified result of the customer area from the second storage device 210 and outputs information about the customer area to the display device 203 .

As described in detail above, the information processing apparatus, even when using the second processing circuit 230, can more accurately identify an area containing the name of the customer from the input image.

Although the preferred embodiments have been described above, the embodiments are not limited to these. For example, the division of functions between the image reading device 100 and the information processing device 200 is not limited to the example of the image processing system 1 shown in FIG. 200 can be changed as appropriate. Alternatively, the image reading device 100 and the information processing device 200 may be configured as one device.

For example, the first storage device 110 of the image reading device 100 may store each program and each data stored in the second storage device 210 of the information processing device 200 . Also, the first processing circuit 120 of the image reading device 100 may operate as each unit realized by the second processing circuit 220 of the information processing device 200 . Also, the image reading device 100 may have a processing circuit similar to the second processing circuit 230 of the information processing device 200 .

In that case, the image reading device 100 has a display device similar to the display device 203 . Since the recognition process is executed by the image reading apparatus 100, the transmission/reception process of the input image in steps S102 and S201 is omitted. Each process of steps S202 to S208 is executed by the first processing circuit 120 of the image reading apparatus 100. FIG. These processing operations are the same as those executed by the second processing circuit 220 or the second processing circuit 230 of the information processing device 200 . In this case, the image reading device 100 operates as an image processing device.

In the image processing system 1, the first interface device 101 and the second interface device 201 are connected via a network such as the Internet, a telephone line network (including a mobile terminal line network and a general telephone line network), an intranet, or the like. good too. In that case, the first interface device 101 and the second interface device 201 are provided with a communication interface circuit of the network to be connected. In that case, multiple information processing devices are distributed over a network so that image processing services can be provided in the form of cloud computing, and each information processing device cooperates to perform recognition processing, etc. You may make it share. As a result, the image processing system 1 can efficiently perform recognition processing on input images read by a plurality of image reading devices.

200 information processing device 210 second storage device 221 acquisition unit 222 image area detection unit 223 character string area detection unit 224 candidate area detection unit 225 first area detection unit 226 second area identification unit 227 customer area identification unit 228 output control unit

Claims (9)

a storage unit that stores a keyword related to the name of a business partner and stores a second keyword or format related to each item for each of a plurality of items included in a form;
an acquisition unit that acquires an input image;
an image area detection unit that detects an image area containing a seal or a logo in the input image;
a character string area detection unit that detects a character string area from the input image;
a candidate area detection unit that detects a candidate area including a client name from the character string area based on the keyword;
a first area detection unit that detects a first area including any one of the plurality of items from the character string area based on the second keyword or the format;
a second region identifying unit that identifies a second region that has not been detected as the candidate region or the first region from the character string region;
When a plurality of candidate areas are detected by the candidate area detection unit, one client area is specified from among the plurality of candidate areas based on the distance from the image area , and the candidate area detection unit a customer area identifying unit that identifies the second area identified by the second area identifying unit as the customer area when no candidate area is detected;
an output control unit that outputs information about the trading partner area;
An image processing device comprising: The customer area identification unit includes a second area having a smaller distance from the image area than the candidate area detected by the candidate area detection unit, and a distance between the second area and the image area. 2. The image processing apparatus according to claim 1 , wherein said second area is specified as said customer area when the distance is equal to or less than a predetermined distance. When the second area identifying unit identifies a plurality of second areas, the customer area identifying unit selects one customer from among the plurality of second areas based on the distance from the image area. 3. The image processing apparatus according to claim 1 , wherein the area is specified. a storage unit that stores keywords related to names of business partners;
an acquisition unit that acquires an input image;
an image area detection unit that detects an image area including a seal or a logo for each of a plurality of partial areas in the input image;
a character string area detection unit that detects a character string area for each partial area in which the image area is detected;
a candidate area detection unit for detecting a candidate area including a client name from the character string area based on the keyword for each partial area where the image area is detected;
When a plurality of candidate areas are detected by the candidate area detection unit, among the candidate areas detected for each of the partial areas in which the image areas are detected, a candidate area having the shortest distance from the corresponding image area is traded. a customer area identification part for identifying a customer area;
an output control unit that outputs information about the trading partner area;
An image processing device comprising: The storage unit further stores a second keyword or format for each item for each of a plurality of items included in the form,
A first area including any one of the plurality of items is detected from the character string area based on the second keyword or the format for each partial area where the image area is detected. a first area detection unit for
a second area identifying unit that identifies a second area that was not detected as the candidate area or the first area from the character string area for each partial area where the image area was detected;
When the candidate area detection unit does not detect a candidate area, the supplier area identification unit is configured to: 5. The image processing apparatus according to claim 4 , wherein a second area having the smallest distance from the corresponding image area is specified as the client area. A control method for an image processing device having a storage unit, the image processing device comprising:
storing a keyword related to the name of a business partner in the storage unit, and storing a second keyword or format related to each item in the storage unit for each of a plurality of items included in a form;
get the input image,
detecting an image area containing a seal or logo in the input image;
detecting a character string region from the input image;
Detecting a candidate area containing a customer name from the character string area based on the keyword,
based on the second keyword or the format, detecting a first area containing any one of the plurality of items from the character string area;
identifying a second region that was not detected as the candidate region or the first region from the character string region;
If a plurality of candidate areas are detected, one customer area is specified from among the plurality of candidate areas based on the distance from the image area, and if no candidate area is detected, the specified specifying the second area as the trading partner area;
outputting information about the trading partner area;
A control method characterized by: A control method for an image processing device having a storage unit, the image processing device comprising:
storing a keyword related to the name of the business partner in the storage unit;
get the input image,
detecting an image region containing a seal or a logo for each of a plurality of partial regions in the input image;
detecting a character string region for each partial region where the image region is detected;
detecting a candidate area containing a client name from the character string area based on the keyword for each partial area where the image area is detected;
when a plurality of candidate areas are detected, among the candidate areas detected for each partial area in which the image area is detected, a candidate area having the smallest distance from the corresponding image area is specified as a customer area;
outputting information about the trading partner area;
A control method characterized by: A control program for a computer having a storage unit,
storing a keyword related to the name of a business partner in the storage unit, and storing a second keyword or format related to each item in the storage unit for each of a plurality of items included in a form;
get the input image,
detecting an image area containing a seal or logo in the input image;
detecting a character string region from the input image;
Detecting a candidate area containing a customer name from the character string area based on the keyword,
based on the second keyword or the format, detecting a first area containing any one of the plurality of items from the character string area;
identifying a second region that was not detected as the candidate region or the first region from the character string region;
If a plurality of candidate areas are detected, one customer area is specified from among the plurality of candidate areas based on the distance from the image area, and if no candidate area is detected, the specified specifying the second area as the trading partner area;
outputting information about the trading partner area;
A control program characterized by causing the computer to execute: A control program for a computer having a storage unit,
storing a keyword related to the name of the business partner in the storage unit;
get the input image,
detecting an image region containing a seal or a logo for each of a plurality of partial regions in the input image;
detecting a character string region for each partial region where the image region is detected;
detecting a candidate area containing a client name from the character string area based on the keyword for each partial area where the image area is detected;
when a plurality of candidate areas are detected, among the candidate areas detected for each partial area in which the image area is detected, a candidate area having the smallest distance from the corresponding image area is specified as a customer area;
outputting information about the trading partner area;
A control program characterized by causing the computer to execute:

Images