JP2023120861A - Voucher information processing device, voucher information processing method, and program - Google Patents

Abstract

To provide a voucher information processing device, a voucher information processing method, and a program, that can reduce the burden of a user when a voucher is detected and shorten the time required for completion of reading voucher contents.SOLUTION: In a voucher information processing system comprising a communication section, an image processing section, a voucher information management section, and a storage section, a voucher information processing device includes: an image acquisition section 111 that acquires a first image M1 including a voucher from the communication section; a voucher position detection section 112 that detects a plurality of vouchers including a first voucher V1 and a second voucher V2 in the first image; an image clipping section 115 that clips a second image M2 including the first voucher from the first image; an image processing section 116 that processes an area of the second voucher from the second image so as not to disable character recognition when the first voucher and the second voucher overlap in the second image; and an image output section 118 that outputs the second image processed by the image processing section 116.SELECTED DRAWING: Figure 2

Description

The present disclosure relates to a voucher information processing device, a voucher information processing method, and a program.

In order to reduce paper handling costs and facilitate information sharing, paper documents such as receipts and application forms are digitized with a scanner and managed as image data (also known as digitized documents). The system has been put into practical use. In this system, paper documents are read by a scanner and created as electronic documents. After that, the character information on the imaged document is extracted as code data by OCR (Optical character recognition) technology, and is used as data for computer processing to support tabulation and analysis work.

The processing of receipts, invoices, and other evidence is an example of the electronic filing system as described above. For example, a certified public accountant office or a tax accountant office (hereinafter simply referred to as an "accounting firm") receives original documents such as receipts and receipts that are the basis of accounting processing from clients and process. At this time, there is a method of converting the original evidence (hereinafter also simply referred to as "evidence" or "receipt") into electronic data with a scanner or a smartphone camera and delivering it to the accounting office. The imaged data of this original document is later used for inputting accounting data by referring to it by staff members of the accounting office, or for automatic reading by OCR processing.

Patent Literature 1 discloses a voucher image acquisition method and an accounting processing system. This accounting processing system reads image data of one or more receipts or other original documents, and includes a document image cutting means for extracting the document part from the image data, and correcting the inclination and distortion of the clipped document image. A shaping means for shaping the image into a rectangular shape and a voucher image registering means for storing the shaped voucher image. The voucher image clipping means attempts to clip the voucher image from the image data using the first set of receipt extraction thresholds in the normal mode, and extracts the receipt when the image data voucher image cannot be clipped using the first set. Using a second set of receipt extraction thresholds in which the fluctuation range of the thresholds is changed, extraction of a voucher image from the image data is attempted.

Japanese Patent No. 6539844

The system of Patent Literature 1 detects each voucher by edge detection. For example, if edge detection is difficult because two or more vouchers are overlapped, the user needs to specify the presence of the edge through manual operation such as pointing operation, which may require time and effort to detect the voucher.

Further, in the system of Patent Document 1, when reading by OCR processing is difficult due to reasons such as inclination of the voucher, the voucher image is rotated by 90 degrees and read is attempted. For this reason, it may take time to rotate for reading, which may lengthen the time required for reading.

The present disclosure provides a voucher information processing device, a voucher information processing method, and a program that can reduce the user's effort when detecting vouchers and shorten the time required to complete reading of the contents of vouchers.

According to one aspect of the present disclosure, an acquisition unit that acquires a first image including a voucher, a detection unit that detects a plurality of vouchers including a first voucher and a second voucher in the first image, and When the cutting unit for cutting out a second image containing the first evidence from the first image and the first evidence and the second evidence overlap in the second image, the A voucher information processing apparatus comprising: a processing unit that processes a region of the second voucher from a second image so that characters cannot be recognized; and an output unit that outputs the second image processed by the processing unit. .

According to one aspect of the present disclosure, an acquisition unit that acquires a fourth image including a voucher, a detection unit that detects the voucher in the fourth image, and the voucher detected by the detection unit has a character orientation are different from each other, a fifth image including the first character area and a sixth image including the second character area. and an output unit for outputting the fifth image and the sixth image.

One aspect of the present disclosure includes the steps of acquiring a first image containing evidence, detecting a plurality of evidence in the first image, including a first evidence and a second evidence, and cutting out a second image containing the first evidence from the image of the second image when the first evidence and the second evidence overlap in the second image A voucher information processing method comprising the steps of: processing an area of the second voucher from the second voucher so that characters cannot be recognized; and outputting the processed second image.

One aspect of the present disclosure is a program that causes a computer to execute the voucher information processing method described above.

According to the present disclosure, it is possible to reduce the burden on the user when detecting a voucher, and shorten the time required to complete reading of the contents of the voucher.

1 is a block diagram of a voucher information processing system including a voucher information processing device according to an embodiment of the present disclosure; FIG. 3 is a block diagram of an overlapped voucher processing unit in the voucher information processing apparatus of Embodiment 1; FIG. 3 is a flowchart of a voucher information processing method executed by the voucher information processing apparatus of the first embodiment; FIG. 2A and 2B are conceptual diagrams of voucher information processing executed by the voucher information processing apparatus of Embodiment 1, where (A) is an image before processing, and (B) is a diagram showing an image after processing; 4A and 4B are conceptual diagrams of another example of voucher information processing executed by the voucher information processing apparatus of Embodiment 1, where (A) shows an image before processing and (B) shows an image after processing; FIG. 11 is a block diagram of an overlapped voucher processing unit in the voucher information processing apparatus of Embodiment 2; FIG. 11 is a flowchart of a voucher information processing method executed by the voucher information processing apparatus of the second embodiment; FIG. 13 depicts an example of two separate voucher images; FIG. 8 is a conceptual diagram of evidence information processing executed by the evidence information processing apparatus of Embodiment 2, where (A) shows an image before processing, (B) shows an image in an intermediate process, and (C) shows an image after processing. FIG. 11 is a flowchart of a voucher information processing method executed by the voucher information processing apparatus of the third embodiment; FIG. 11 shows screens of processing results by the voucher information processing apparatus of the third embodiment, which are screens displayed by the operation terminal, where (A) is an initial screen and (B) is a screen in which two voucher images are displayed side by side. This is a screen displayed by checking the check box for "superposition" in FIG. 11B. It is a screen of the processing result by the voucher information processing apparatus of Embodiment 4, and is a screen displayed by the operation terminal. 14 is a screen displayed by pressing the detail button of warning information in FIG. 13. FIG. It is a screen of the processing result by the documented evidence information processing apparatus of Embodiment 5, and is a screen which an operation terminal displays.

Hereinafter, embodiments will be described in detail with reference to the drawings as appropriate. However, more detailed description than necessary may be omitted. For example, detailed descriptions of well-known matters and redundant descriptions of substantially the same configurations may be omitted. This is to avoid unnecessary verbosity in the following description and to facilitate understanding by those skilled in the art. It should be noted that the accompanying drawings and the following description are provided to enable those skilled in the art to fully understand the present disclosure and are not intended to limit the claimed subject matter thereby.

(Configuration of voucher information processing system)
FIG. 1 is a block diagram of a voucher information processing system including a voucher information processing device according to an embodiment of the present disclosure. The voucher information processing system 200 includes a voucher information processing device 1, a plurality of operation terminals 50 (50A, 50B), a plurality of scanners 60 (60A, 60B), and a communication network 70 that communicably connects these devices. Configured.

The operation terminal 50 is a terminal used by an applicant (an example of a user) who inputs evidence or a person in charge of processing the evidence (for example, an accounting person) (an example of a user). The operation terminal 50 is, for example, a PC (Personal Computer) or a tablet terminal. The operation terminal 50 includes, for example, a control section, a communication section, a storage section, an operation section, and a display section, like a PC. A control part is comprised by a processor, for example. The storage unit includes, for example, ROM (Read Only Memory) or RAM (Random Access Memory). The operation unit may be a keyboard, mouse, touch panel, or the like. The display unit is, for example, a liquid crystal display device. The display unit may display images, data, information, or the like processed by the documented evidence information processing apparatus 1, for example. The user can check the displayed image, data, information, or the like. The operation terminal 50A is a terminal used by a person in charge of processing. The operation terminal 50B is a terminal used by the applicant.

The scanner 60 is a device that reads paper documents including proofs such as receipts, receipts, application forms, statements of delivery, contracts, etc., and creates electronic documents. The data format of the electronic document may be general image data. For example, bitmap format or JPEG format can be used. The scanner 60 is, for example, a scanner equipped with an auto-feeder mechanism, and may read standard size paper documents. The scanner 60 is, for example, a stand-type scanner that reads an image of a document placed on a desk, and may read irregular-sized paper documents.

(Basic Configuration of Voucher Information Processing Device)
The voucher information processing apparatus 1 takes in the image data of the voucher read by the scanner 60, and performs various processes. As a result, the voucher information processing apparatus 1 can reduce the user's trouble in detecting the voucher, and shorten the time required to complete the reading of the contents of the voucher.

The voucher information processing device 1 includes a communication unit 30 , an image processing unit 10 , a voucher information management unit 20 and a storage unit 40 . The storage unit 40 includes a registration voucher database (DB) 40D.

The communication unit 30 communicates various data or information. The communication method by the communication unit 30 may be a wired communication method or a wireless communication method, such as LAN (Local Area Network), WAN (Wide Area Network), power line communication, or cellular communication for mobile phones (e.g., 5G, LTE (Long Term Evolution)) and other communication methods may be included. The communication unit 30 functions as a receiving unit that receives various data (eg, image data) transmitted from the operation terminal 50, the scanner 60, or the like via the communication network 70, for example. The communication unit 30 functions, for example, as a transmission unit that transmits various data (such as image data) to the operation terminal 50 or the scanner 60 via the communication network 70 . The data received by the communication unit 30 is processed by the image processing unit 10, for example. The communication unit 30 transmits data processed by the image processing unit 10, for example.

The image processing unit 10 is configured by, for example, a processor. The processor may include an MPU (Micro processing Unit), a CPU (Central Processing Unit), a DSP (Digital Signal Processor), a GPU (Graphical Processing Unit), or the like. The processor may include various integrated circuits (eg, LSI (Large Scale Integration) or FPGA (Field Programmable Gate Array)) and the like. The image processing unit 10 implements various functions by executing programs held in the storage unit 40 . The image processing unit 10 comprehensively controls each unit of the voucher information processing apparatus 1 and performs various processes.

The image processing unit 10 includes, as a functional configuration, an overlapping voucher processing unit 11, an OCR processing unit 12, a duplicate registration confirmation processing unit 13, an input support processing unit 14, and an input schedule processing unit 15. The image processing unit 10 is, for example, a part responsible for the main functions of the voucher information processing apparatus 1, and when the image data received by the communication unit 30 via the communication network 70 is the image data of a voucher, it performs the processing described later.

The overlapped voucher processing unit 11 performs processing described later when the acquired image data of the voucher is image data in which two or more vouchers overlap (first embodiment). Further, the overlapping voucher processing unit 11 performs processing described later when the acquired image data of the voucher is image data including two or more character areas with different character orientations (embodiment 2). The details of the overlapping voucher processing unit 11 will be described later.

The OCR processing unit 12 recognizes characters or text portions of the image data processed by the overlapping document processing unit 11 and converts them into character data. The OCR processing unit 12 performs general OCR (Optical Character Reader) processing, that is, optical character recognition.

The overlapping registration confirmation processing unit 13 compares the image data newly processed by the overlapping voucher processing unit 11 with other image data, for example, the image data of the voucher registered in the registered voucher database 40D, and confirms that the same image (same voucher ). Thereby, the duplicate registration confirmation processing unit 13 confirms whether or not the information regarding the same voucher is duplicately registered. The input support processing unit 14 performs processing related to support for input by a person in charge of processing vouchers. The input schedule processing unit 15 performs processing related to prompting the applicant to input the voucher that is scheduled. Details of the duplicate registration confirmation processing unit 13, the input support processing unit 14, and the input schedule processing unit 15 will be described in the third to fifth embodiments.

The voucher information management unit 20 has a function for the user (applicant or person in charge of processing) to manage the voucher information processing device 1 .

The storage unit 40 includes a primary storage device (for example, RAM (Random Access Memory) or ROM (Read Only Memory)). The storage unit 40 may include a secondary storage device (for example, HDD (Hard Disk Drive) or SSD (Solid State Drive)) or a tertiary storage device (for example, optical disk or SD card). The storage unit 40 may include other storage devices. The storage unit 40 may be detachable from the voucher information processing apparatus 1 . The storage unit 40 holds various data or information.

The registered voucher database 40D contains the image of the voucher processed by the overlapping voucher processing unit 11, the voucher information (first voucher information) related to the contents of the voucher obtained by character recognition from the image of the voucher by the OCR processing unit 12, and the like. accumulate. Therefore, the image of the voucher and the voucher information are accumulated as the results of the past vouchers. Also, the registered voucher database 40D may include an original paper image obtained by scanning the original paper of the voucher. The processed voucher image, original paper image, and voucher information may be associated and accumulated. A user (applicant and person in charge of processing) of the operation terminal 50 can utilize the processed image data, evidence information, or the like registered in the registered voucher database 40D.

The evidence information includes information for individually identifying the evidence. The voucher information includes voucher identification information for identifying vouchers. If the voucher is an invoice, the voucher identification information may include invoice number, amount, or payee account number. If the voucher is a receipt, the voucher identification information may include date, amount, counterparty or name, and the like.

The registered voucher database 40D may be arranged outside the voucher information processing apparatus 1, and may be an independent database server, for example. Via the communication unit 30, the documented evidence information processing apparatus 1 may be able to access the database server and acquire accumulated images, data, information, or the like.

(Embodiment 1)
Next, the documented evidence information processing apparatus 1 according to the first embodiment will be described. FIG. 2 is a block diagram of the overlapped voucher processing unit 11 in the voucher information processing apparatus 1 of the first embodiment. This embodiment is useful when the first image M1, which is the image data received by the communication unit 30, includes two or more images of evidence.

In the illustrated example, the first image M1 includes a second image M2 containing the first voucher V1 and a third image M3 containing the second voucher V2 (before processing). The first voucher V1 and the second voucher V2 are different receipts (outline of the receipt and its details), and the entire surface of the second voucher V2 overlaps the first voucher V1, which has a larger area. It is listed as In addition, the orientations of the first voucher V1 and the second voucher V2 are slanted with respect to each other, and the orientations of the respective characters are also slanted. The first image M1 includes not only the second image M2 and the third image M3, but also the image behind the two images, such as the image of the mount on which the first voucher V1 and the second voucher V2 are pasted. may include

When a conventional OCR device reads such a voucher, reading errors can occur due to the different orientation of the image. However, in the present embodiment, the overlapped evidence processing unit 11 performs the following processing on the first image M1, resulting in a second image M2 including the first evidence V1 and a second image M2 including the second evidence V2. The third image M3 is separately output in a separated state (after processing). In addition, the overlapping voucher processing unit 11 processes the processing area MX of the third image M3 in the second image M2 so that the characters cannot be recognized. In this example, black coating processing is applied. As a result, the OCR processing unit 12 can properly recognize characters or text portions in the second image M2 and the third image M3 processed by the overlapping document processing unit 11 and convert them into character data. Details of the overlapped voucher processing unit 11 will be described below.

In this embodiment, the overlapped voucher processing unit 11 includes an image acquisition unit 111, a voucher position detection unit 112, a size relationship calculation unit 113, an overlap ratio calculation unit 114, an image clipping unit 115, and an image processing unit 116. , an image rotation correction unit 117 , and an image output unit 118 .

The image acquisition unit 111 functions as an acquisition unit that acquires the first image M<b>1 including the voucher from the communication unit 30 . The voucher position detection unit 112 functions as a detection unit that detects a plurality of vouchers including the first voucher V1 and the second voucher V2 in the first image M1.

The voucher position detection unit 112 may use a detector that has been trained as artificial intelligence to detect the plurality of vouchers. The voucher position detection unit 112 generates and updates learned parameters and a learned model (detector) by performing learning by executing a learning program using a set of learning data when learning the detector. A learned parameter is a parameter (coefficient) adjusted by learning. A trained model is a program that incorporates trained parameters. The voucher position detection unit 112 creates learning data by, for example, annotation processing in the AI (Artificial Intelligence) learning field. The voucher position detection unit 112 inputs, for example, a large number of images containing arbitrary vouchers, inputs the xy coordinate values of the four corners of the region where the vouchers exist in the input image, and generates an image with coordinates. Generate learning data based on data conversion of The input images may include overlapping and non-overlapping images. The voucher position detection unit 112 learns the presence/absence of a voucher in an arbitrary image, the position and size of the voucher with respect to the image, the tilt of the voucher, etc. based on the learning data. As a result, the voucher position detection unit 112 can detect (the position of) an individual voucher image even if a plurality of voucher images overlap. The voucher position detection unit 112 can detect the position, size, inclination of the voucher, etc. with respect to the image. Note that the detector learning itself may be performed outside the voucher position detection unit 112 .

The magnitude relation calculation unit 113 calculates the magnitude relation between the second image M2 including the first voucher V1 detected by the voucher position detection unit 112 and the third image M3 including the second voucher V2. The size relationship is represented by dimensions, areas, etc. In this example, the size relationship is M2>M3.

The duplication rate calculation unit 114 calculates the second image M2 and the third image M3 from the second image M2 and the third image M3 detected by the voucher position detection unit 112 and the magnitude relation calculated by the magnitude relation calculation unit 113. and the image M3, that is, the degree of overlap between the two images. Overlap rate calculation section 114 can use, for example, IoU (Intersection over Union) as the overlap rate. IoU is the ratio of the overlapping area to the total area.

The image clipping unit 115 functions as a clipping unit that clips a second image M2 including the first voucher V1 from the first image M1. The image clipping unit 115 may clip the third image M3 including the second voucher V2 from the first image M1.

When the first voucher V1 and the second voucher V2 overlap in the second image M2, the image processing unit 116 converts the second image M2 to a machining region MX corresponding to the region of the second voucher V2. It functions as a processing unit that processes the characters so that they cannot be recognized. In this example, the image processing unit 116 blacks out the processing area MX corresponding to the area of the second voucher V2, thereby disabling character recognition of the processing area MX by OCR processing. Processing that disables character recognition may be not only black painting processing, but also white painting processing, blurring, and other processing methods.

The image rotation correction unit 117 processes the second image M2 processed by the image processing unit 116 by rotation correction. Rotation correction is a process of changing the orientation of the image of the voucher so that the voucher and its characters face a predetermined direction (for example, the horizontal direction of the paper surface of the drawing). The image rotation correction unit 117 may also perform rotation correction of the third image M3. The image rotation correction unit 117 may, for example, detect the top and bottom of the voucher included in each image based on the characteristics of each image, and adjust the angle of each image based on the top and bottom of the voucher to correct the rotation. The image rotation correction unit 117 may perform rotation correction based on the direction of the characters or the frame indicating the outer circumference of the voucher. The image rotation correction unit 117 can align the orientations of the second image M2 and the third image M3 by rotating the second image M2 and the third image M3.

The image output unit 118 functions as an output unit that outputs at least the second image M2 processed by the image processing unit 116 and the image rotation correction unit 117 . The image output unit 118 may also output the third image M3.

Upon receiving the processing result of the image output unit 118, the OCR processing unit 12 functions as a recognition unit that recognizes characters in the second image M2 and the third image M3 based on general OCR processing. The OCR processing unit 12 recognizes characters in each image and converts them into text data. This text data may be stored, for example, in the registration voucher database 40</b>D, or may be transmitted to the operation terminal 50 via the communication network 70 by the communication section 30 . Also, the duplicate registration confirmation processing unit 13 may check the text data. Further, the OCR processing unit 12 may classify the vouchers by type of vouchers (for example, receipts, invoices) based on the text data obtained by the OCR processing. The registered voucher database 40D may hold image data, text data, or voucher information related to vouchers for each type of voucher.

FIG. 3 is a flowchart of the voucher information processing method executed by the voucher information processing apparatus 1 of the first embodiment. As a premise for this, for example, it is assumed that the communication unit 30 has received the first image M1 from the scanner 60 via the communication network 70 . It is also assumed that the first image M1 includes one or more vouchers.

The image acquisition unit 111 acquires the first image M1 from the communication unit 30 (step S1). The voucher position detection unit 112 detects the voucher image (evidence image, voucher area) in the first image M1 (step S2), and determines whether or not a plurality of voucher images exist in the first image M1. (step S3). When only one image exists in the first image M1 (step S3; No), the image clipping unit 115 clips this one voucher image (step S4). The image rotation correction unit 117 rotates and corrects the clipped image of the voucher (step S5), and the image output unit 118 outputs the corrected image to the OCR processing unit 12 (step S6).

In step S3, when the voucher position detection unit 112 determines that a plurality of images exist in the first image M1 (step S3; Yes), the size relationship calculation unit 113 calculates the size relationship between the two voucher images. calculate. In addition, the overlap ratio calculation unit 114 calculates the overlap ratio (overlap ratio) of the two images of the voucher, that is, the second image M2 corresponding to the first voucher V1 and the third image M3 corresponding to the second voucher V2. degree) is calculated. The duplication rate calculator 114 calculates, for example, the duplication rate between the first voucher V1 and the second voucher V2 in the second image M2. The duplication rate calculation unit 114 uses, for example, IoU (Intersection over Union) as the duplication rate.

If the overlap rate is smaller than a predetermined threshold value (for example, 0.075) (step S7; No), the degree of overlap is small, and ignoring the overlap portion is considered to cause little actual harm. Therefore, the image clipping unit 115 clips a plurality of voucher images, that is, the second image M2 and the third image M3 (step S8). The documented evidence information processing apparatus 1 can reduce the processing load by omitting a process such as calculation of the duplication rate. The image rotation correction unit 117 rotates and corrects the clipped second image M2 and third image M3 (step S5), and the image output unit 118 outputs the corrected second image M2 and third image M3. Output to the OCR processor 12 (step S6).

In step S7, if the overlap rate is smaller than the predetermined threshold value (step S7; No), the degree of overlap is high, and ignoring the overlap is highly likely to cause actual damage such as OCR reading errors. it is conceivable that. Therefore, the image processing unit 116 blacks out the processing area MX corresponding to the second voucher V2 from the large second image M2, thereby disabling character recognition in the processing area MX (step S9).

The image clipping unit 115 clips a small third image M3 corresponding to the second voucher V2 existing in the processing area MX (step S10). The image rotation correction unit 117 rotates and corrects the processed second image M2 and the third image M3 (step S5), and the image output unit 118 outputs the corrected second image M2 and the third image M3. is output to the OCR processor 12 (step S6).

4A and 4B are conceptual diagrams of the evidence information processing executed by the evidence information processing apparatus 1 of the first embodiment, in which (A) shows an image before processing and (B) shows an image after processing. The second image M2 is an image of the first voucher V1, which is a receipt, and the third image M3 is an image of the second voucher V2, which is a receipt. As explained above, the second image M2 has a larger image range than the third image M3. Particularly in this example, as shown in FIG. 4A, the entire area of the third image M3 is superimposed on the second image M2, and the third image M3 is separated from the edge of the second image M2. It does not protrude. With a technique such as conventional edge detection, it is difficult to detect the second image M2 and the third image M3 from the first image M1.

The documented evidence information processing apparatus 1 of the present embodiment processes and outputs the unprocessed image shown in FIG. 4(A) as shown in FIG. 4(B). That is, the documented evidence information processing apparatus 1 can detect the second image M2 and the third image M3 with high accuracy even when at least a part of the second image M2 and the third image M3 overlap. Then, the documented evidence information processing apparatus 1 processes the area where the third image M3 exists in the second image M2 into a processed area MX in which characters cannot be recognized. As a result, the documented evidence information processing apparatus 1 can remove an area in the second image M2 that may have a character orientation different from the character orientation of the second image M2. OCR reading errors of M2 can be suppressed.

Furthermore, the voucher information processing apparatus 1 can also cut out the third image M3. Accordingly, both the second image M2 and the third image M3, which originally exist in the first image M1, can be read separately. In addition, the documented evidence information processing apparatus 1 corrects the rotation of the second image M2 and the third image M3, thereby aligning the orientation of the characters in both images and adjusting the orientation of the characters to be suitable for OCR processing. The reading accuracy of OCR processing can be improved.

5A and 5B are conceptual diagrams of another example of voucher information processing executed by the voucher information processing apparatus of the first embodiment, where (A) shows an image before processing and (B) shows an image after processing. The image before processing in FIG. 5A is the same as the image in FIG. 4A. In this example, when the image cropping unit 115 cuts out the small third image M3 corresponding to the second voucher V2 existing in the processing area MX (step S10 in FIG. 3), the image processing unit 116 By blacking out the processing area MX1 corresponding to the second image M2 existing around the third image M3, character recognition in the processing area MX1 is disabled.

In the example described above, the entire surface of the second voucher V2 is placed on the first voucher V1 having a larger area. However, the present embodiment can be applied not only to such cases but also to cases in which part of the first document V1 and part of the second document V2 overlap each other. In addition, this embodiment can also be applied to a case where the first voucher V1 and the second voucher V2 are the same size, or a case where three or more vouchers are overlapped.

(Embodiment 2)
Next, the documented evidence information processing apparatus 1 according to the second embodiment will be described. FIG. 6 is a block diagram of the overlapped voucher processing unit 11 in the voucher information processing apparatus 1 of the second embodiment. This embodiment is useful when the first image M1, which is the image data received by the communication unit 30, includes a plurality of character areas with different character orientations.

In the illustrated example, the fourth image M4 including the voucher includes a fifth image M5 including the first character area C1 and a sixth image M6 including the second character area C2 ( before treatment). The first character area C1 and the second character area C2 are character areas in which the orientations of characters are different from each other. In this example, the first character area C1 and the second character area C2 are present adjacent to each other on the fourth image M4 corresponding to one receipt, and the orientations of the characters are different by 90 degrees. there is Note that the difference in character orientation may be other than 90 degrees. For example, the first character area C1 and the second character area C2 originally correspond to the same third voucher V3 (receipt), but the orientation of the characters may differ. The fourth image M4 may include not only the third voucher V3 but also an image behind the voucher, for example, an image of a mount on which the third voucher V3 is pasted.

If an OCR device reads the voucher in such a state, reading errors may occur. However, in the present embodiment, the overlapping document processing unit 11 performs the following processing on the fourth image M4, resulting in a fifth image M5 including the first character region C1 and a second character region C2. and a sixth image M6 containing , are output separately in a separated state (after processing). As a result, the OCR processing unit 12 can properly recognize the characters or text portions of the fifth image M5 and the sixth image M6 processed by the overlapping document processing unit 11 and convert them into character data. Details of the overlapped voucher processing unit 11 will be described below.

In this embodiment, the overlapping voucher processing unit 11 includes an image acquisition unit 111, a voucher position detection unit 112, a character direction detection unit 119, an image clipping unit 115, an image rotation correction unit 117, and an image output unit 118. And prepare.

The image acquisition unit 111 functions as an acquisition unit that acquires the fourth image M4 including the third voucher V3 from the communication unit 30. FIG. The voucher position detection unit 112 functions as a detection unit that detects the third voucher V3 in the fourth image M4.

The voucher position detection unit 112 may use a learned detector to detect the voucher. The detector learning method may be the same as that described in the first embodiment.

The character orientation detection unit 119 detects the orientation of the characters of the third voucher V3 detected by the voucher position detection unit 112 . As with the voucher position detector 112, the character orientation detector 119 may use a detector trained as artificial intelligence to detect the orientation of the characters of the third voucher V3. When the detector is trained, a learning data set is used to execute a learning program for learning, thereby generating and updating a trained parameter and a trained model (detector). The character direction detection unit 119 creates learning data by processing annotations in the AI learning field, for example. The character orientation detection unit 119 inputs, for example, a large number of images containing arbitrary evidence by character orientation, inputs the xy coordinate values of the four corners of the region in which the evidence exists in the input image, and calculates the coordinate Learning data may be generated based on digitizing the attached image. At this time, the training data includes an image that includes a plurality of character areas with different character orientations in one voucher image. The character orientation detection unit 119 learns orientations (orientations of characters) of one or more character regions included in any image described above, based on the learning data. Furthermore, the voucher position detection unit 112 also learns whether or not vouchers with different character orientations are included in the same voucher. As a result, the documented document position detection unit 112 can detect the orientation (angle) of each of the plurality of character areas present in the image. Note that the detector learning itself may be performed outside the voucher position detection unit 112 .

Therefore, the voucher position detection unit 112 and the character direction detection unit 119 detect the image on the image by a detector that has been trained as artificial intelligence, even if one voucher in the image has a plurality of regions with different directions of character regions. It is possible to detect regions in which the direction of the character region is different.

The third voucher V3 in this example includes two character areas with different character orientations. Therefore, the character orientation detection unit 119 may detect the first character region C1 and the second character region C2 having different character orientations.

The image clipping unit 115 serves as a clipping unit that clips a fifth image M5 including the first character region C1 and a sixth image M6 corresponding to the second character region C2 from the fourth image M4. Function.

The image rotation correction unit 117 processes the fifth image M5 and the sixth image M6 cut out by the image cutout unit 115 by rotation correction. Rotation correction is a process of changing the orientation of the image of the voucher so that the voucher and its characters face a predetermined direction (for example, the horizontal direction of the paper surface of the drawing). The image rotation correction unit 117 may, for example, detect the top and bottom of the voucher included in each image based on the characteristics of each image, and adjust the angle of each image based on the top and bottom of the voucher to correct the rotation. The image rotation correction unit 117 may perform rotation correction based on the direction of the characters or the frame indicating the outer circumference of the voucher. Therefore, the image rotation correction unit 117 can align the orientations of the fifth image M5 and the sixth image M6 by rotating the fifth image M5 and the sixth image M6.

The image output unit 118 functions as an output unit that outputs the fifth image M5 and the sixth image M6 processed by the image rotation correction unit 117. FIG.

Upon receiving the processing result of the image output unit 118, the OCR processing unit 12 functions as a recognition unit that recognizes the characters of the fifth image M5 and the sixth image M6 based on general OCR processing. The OCR processing unit 12 recognizes characters in each image and converts them into text data. This text data may be stored, for example, in the registration voucher database 40</b>D, or may be transmitted to the operation terminal 50 via the communication network 70 by the communication section 30 . Also, the duplicate registration confirmation processing unit 13 may check the text data. Further, the OCR processing unit 12 may classify the vouchers by type of vouchers (for example, receipts, invoices) based on the text data obtained by the OCR processing. The registered voucher database 40D may hold image data, text data, or voucher information related to vouchers for each type of voucher.

FIG. 7 is a flowchart of the evidence information processing method executed by the evidence information processing apparatus 1 of the second embodiment. In FIG. 7, for example, it is assumed that the communication unit 30 has received the fourth image M4 from the scanner 60 via the communication network . It is also assumed that the fourth image M4 includes one or more character areas of the voucher.

The image acquisition unit 111 acquires the fourth image M4 from the communication unit 30 (step S11). The voucher position detection unit 112 detects the voucher image (evidence image, voucher area) in the fourth image M4 (step S12). The character direction detection unit 119 detects the character orientation of the third voucher V3 detected by the voucher position detection unit 112, and detects a plurality of character areas having different character orientations, that is, the first character area C1 and the second character area C1. It is determined whether or not the character area C2 exists (step S13).

If only one character area exists in the fourth image M4 (step S13; No), the image clipping unit 115 clips the evidence image corresponding to this one character area (step S14). The image rotation correction unit 117 rotates and corrects the clipped image of the voucher (step S15), and the image output unit 118 outputs the corrected image to the OCR processing unit 12 (step S16).

In step S13, when the character direction detection unit 119 determines that a plurality of character regions with different character orientations exist in the fourth image M4 (step S13; Yes), the image clipping unit 115 A fifth image M5 including one character area C1 and a sixth image M6 including a second character area C2 are separately cut out (step S17). The image rotation correction unit 117 rotates and corrects the image of the evidence corresponding to each of the cut out fifth image M5 and sixth image M6 (step S15), and the image output unit 118 corrects the image to the OCR processing unit 12. (step S16).

As shown in FIG. 8, there may exist not only the fourth image M4 including the third voucher V3, but also the image MY including the voucher VY separated from the third voucher V3. Since the fifth image M5 and the sixth image M6 are cut out from the fourth image M4, the fifth image M5, the sixth image M6 and the image MY are output and subjected to OCR processing.

9A and 9B are conceptual diagrams of the evidence information processing executed by the evidence information processing apparatus 1 of the second embodiment, in which (A) is an image before processing, (B) is an image in an intermediate process, and (C) is an image after processing. Show the image. A fourth image M4 is an image of the third voucher V3, which is a receipt. As described above, as shown in FIG. 9A, the fifth image M5 including the first character area C1 and the sixth image M6 including the second character area C2 are combined into the fourth image. They are arranged adjacently in the image M4. The orientation of the characters in the first character area C1 differs from the orientation of the characters in the second character area C2 by 90 degrees. It is difficult to detect the fifth image M5 and the sixth image M6 from the fourth image M4 by a technique such as conventional edge detection.

The documented evidence information processing apparatus 1 of the present embodiment separately cuts out a fifth image M5 and a sixth image M6 from the unprocessed image shown in FIG. 9A as shown in FIG. 9B. Output. That is, the documented evidence information processing apparatus 1 detects the directions of the first character region C1 and the second character region C2 whose character orientations are different from each other, and detects both the fifth image M5 and the sixth image M6. Both the fifth image M5 and the sixth image M6 can be read separately.

Furthermore, as shown in FIG. 9C, the documented evidence information processing apparatus 1 corrects the rotation of the fifth image M5 and the sixth image M6 to make the orientation of the characters of both images suitable for OCR processing. It can be adjusted to align with the direction of characters, and reading accuracy can be improved in OCR processing.

Note that in the above example, two character areas exist in one voucher. However, the present embodiment can be applied not only to such cases, but also to cases in which three or more character areas exist in one voucher.

(Embodiment 3)
Examination of the evidences applied for by the applicants is one of the major issues in the processing of the evidences. For example, it is important to examine whether or not the requested evidence is inconsistent with past evidence that has already been processed (whether or not it overlaps with past evidence). It is also important to examine whether or not the content of the requested voucher matches the original document of the voucher. As a typical example, examination of invoices, which is a type of evidence, places a heavy burden on the processing personnel in the accounting departments of companies.

Hitherto, the above examination has been carried out, for example, by a checksum comparison method for examining the identity of the data of two vouchers. However, in this method, the data are not considered identical unless they match perfectly. For example, even if the original paper of the data of the two vouchers is the same, the data of the two vouchers may be different even if they originally have the same content due to factors such as the difference in the timing of scanning by the scanner and the difference in the model. It may be recognized as evidence, and the accuracy of the examination may decrease.

The person in charge of processing the voucher judges from memory whether or not the voucher newly applied by the applicant overlaps with the past voucher. Also, when using an accounting system that manages original documents together with electronic data, the person in charge of processing visually confirms that the documented evidence data applied and the original documents of the voucher match. Such processing increases the burden on the person in charge of processing, and may induce oversight in examination.

Therefore, the voucher information processing apparatus 1 of the present embodiment supports the work related to the voucher by the person in charge of processing. The voucher information processing device 1 generates the voucher data (current voucher information) based on the requested voucher image (current voucher image), and the previously processed voucher image or scanned original voucher image ( Voucher data (registered voucher information) based on the registered voucher image) is compared to determine whether or not the voucher information matches. If they match, the evidence information processing apparatus 1 issues a warning, and the person in charge of processing checks the evidence based on the warning information, thereby improving the accuracy of examination and reducing the burden on the person in charge of processing.

The above processing is mainly performed by the duplicate registration confirmation processing unit 13 in FIG. 1, for example. The duplicate registration confirmation processing unit 13 receives the current evidence information regarding the applied evidence. The duplicate registration confirmation processing unit 13 compares the current voucher information with registered voucher information registered in the registered voucher database 40D. When the duplicate registration confirmation processing unit 13 determines that the two vouchers are duplicated by comparison, the communication unit 30 transmits warning instruction information including an instruction to present warning information indicating duplication via the communication network 70. to the operation terminal 50A.

In the operation terminal 50A, the communication section receives the warning instruction information, and the display section or the like presents the warning information based on the warning instruction information. The person in charge of processing can confirm the warning information on the display section of the operation terminal 50A. The communication unit 30 of the documented evidence information processing apparatus 1 may transmit the warning instruction information not only to the operation terminal 50A of the person in charge of processing but also to the operation terminal 50B of the applicant.

Also, when using an accounting system that manages original paper together with electronic data, the operation terminal 50B acquires various information input by the applicant via the operation unit. The information input by the applicant is information related to evidence that can be included in the current evidence information, and is also referred to as input evidence information. The operation terminal 50A may acquire and display the input voucher information input by the applicant. The person in charge of processing using the operation terminal 50 may also visually check whether the input voucher information and the original document of the voucher match.

In this way, the person in charge of processing confirms whether or not the current voucher information matches the registered voucher information based on the judgment performed by the voucher information processing apparatus 1, and visually checks the current voucher image and the original document of the voucher. Also check. If the two confirmations do not match, that is, if the current documented evidenced information is not yet registered, the approval is completed. Further, the voucher information processing apparatus 1 may similarly issue warning information when the input voucher information is different from the current voucher information.

FIG. 10 is a flowchart of the voucher information processing method executed by the voucher information processing apparatus 1 of the third embodiment. The flow of FIG. 10 may be performed, for example, after the flow of Embodiment 1 of FIG. 3 or the flow of Embodiment 2 of FIG. Note that the flow of FIG. 10 is not necessarily performed after the flow of FIG. 3 or 7, and can be performed independently of these flows.

The OCR processing unit 12 performs OCR processing on the voucher image (step S21). The images of the vouchers to be subjected to the OCR processing here are, for example, the second image M2, the third image M3, the fifth image M5, and the sixth image, which are the output images in the first and second embodiments. At least one of M6. Note that the image of the voucher other than the image of the voucher obtained in the first embodiment or the second embodiment may be used. Voucher information (current voucher information) is obtained from the image of the voucher by OCR processing.

Duplicate registration confirmation processing unit 13, based on the current evidence information obtained by OCR processing unit 12, determines whether or not the same evidence information (registered evidence information) has been registered in registered evidence database 40D in the past. Determine (step S22). In other words, the duplicate registration confirmation processing unit 13 functions as a determination unit that determines whether or not the voucher information registered in the registered voucher database 40D and the current voucher information are the same voucher information.

The duplicate registration confirmation processing unit 13 compares, for example, the registered voucher identification information included in the registered voucher information with the current voucher identification information included in the current voucher information. If the registered voucher identification information and the current voucher identification information are the same, the duplicate registration confirmation processing unit 13 may determine that they are the same voucher information.

If the duplicate registration confirmation processing unit 13 determines that the registered voucher information and the current voucher information are the same voucher information, the image of the voucher (current voucher image) in step S21 and the registered voucher database 40D There is a high possibility that the image of the voucher linked to the registered voucher information (registered voucher image) will be duplicated. Therefore, the communication unit 30 transmits warning instruction information to the operation terminal 50A via the communication network 70 (step S23). The warning instruction information may include, for example, warning information, current evidenced image, and registered evidenced image.

The communication unit of the operation terminal 50A receives the warning instruction information, and the display unit displays the warning information. In addition, the display unit may display the two images of the vouchers (the current voucher image and the registered voucher image) compared in step S22 together with the warning information (step S24). By checking this display, the person in charge of processing can easily visually determine whether or not the current evidenced image and the registered evidenced image match. The operation terminal 50A makes one of the images of the two vouchers translucent and displays the images of the two vouchers so that the person in charge of processing can easily recognize the difference between the images of the two vouchers. , it is possible to more easily visually confirm whether the document is really duplicated.

FIG. 11 is a screen showing the result of processing by the documented evidence information processing apparatus 1 of the third embodiment, and is a screen displayed by the operation terminal 50A of the person in charge of processing. The display unit of the operation terminal 50A displays the initial screen of FIG. 11(A). This initial screen contains warning information for "past bill duplication". When the person in charge of processing presses the confirmation button for the voucher (invoice) on which this warning information is displayed, the display unit displays two voucher images (the current voucher image and the registered voucher image) as shown in FIG. evidence image) are displayed side by side.

FIG. 12 shows a screen displayed by the operation terminal 50A of the person in charge of processing, and shows an image in which one of the above two images of the voucher is translucent and the two images are superimposed and displayed. When the person in charge of processing checks, for example, the "Overlay" box in FIG. 11B via the operation unit of the operation terminal 50A, the display unit of the operation terminal 50A displays the screen shown in FIG. In FIG. 12, one description is "52,000 yen", while the other description is "53,000 yen", and other descriptions are the same. The person in charge of processing can easily recognize that there is an error in the description of the amount portion by checking the superimposed display as shown in FIG. In addition, the documented evidence information processing apparatus 1 is operated by the person in charge of processing via the operation unit to operate the "send back" button in FIG. (e.g. applicant).

(Embodiment 4)
In general, the operation terminal 50B used by the applicant who applies for (submits) the evidence, inputs the information of each item of the application form for the evidence via the operation unit of the operation terminal 50B before submitting the evidence, This information (input voucher information) is transmitted to the operation terminal 50A of the person in charge of processing. The person in charge of processing may check the input voucher information while visually comparing it with the image data of the voucher processed by the voucher information processing apparatus 1 . This work also imposes a heavy burden on the person in charge of processing.

In the documented evidence information processing apparatus 1 of the present embodiment, the communication unit 30 receives the application form data (input documented evidence information) from the applicant's operation terminal 50B via the communication network 70 . The input support processing unit 14 determines whether or not predetermined items of the voucher application form match the items of the voucher data read by the voucher information processing apparatus 1 in the past. That is, the voucher information processing apparatus 1 determines whether or not the predetermined information included in the input voucher information matches the corresponding predetermined information included in the registered voucher information.

When the above predetermined items match, the communication unit 30 transmits matching registered voucher information to the operation terminal 50A. In the operation terminal 50A, the communication unit receives the registered evidence information, and the display unit displays it as the initial value of the data of the application form for the evidence. Further, when the above predetermined information matches, the input support processing unit 14 instructs the operation terminal 50A to perform display (for example, display of similar information) in consideration of misrecognition of reading by the OCR processing unit 12. You can

When the above predetermined items do not match, the communication unit 30 may transmit warning instruction information including warning information to that effect to the operation terminal 50A. In the operation terminal 50A, the communication section may receive the warning instruction information, and the display section may display the warning information. The person in charge of processing may confirm the display of the warning information and perform processing such as new registration and correction of information including the relevant item via the operation unit.

The above processing is mainly performed by the input support processing unit 14 in FIG. 1, for example. The input support processing unit 14 receives the OCR-processed voucher image sent from the OCR processing unit 12 . The input support processing unit 14 identifies, for example, characters corresponding to a representative item, such as a bank account number of a bill, from the image of the document that has been applied for. The input support processing unit 14 refers to the registered voucher database 40D, and searches for the item linked to the registered voucher that has been processed in the past, that is, the bank account number in the registered voucher information.

The input support processing unit 14 determines whether or not registered voucher information having information matching representative items of the requested voucher (for example, bank account number) exists in the registered voucher database 40D. If it exists, the communication unit 30 transmits the registered documented evidence information including the item to the operation terminal 50A of the person in charge of processing via the communication network 70 .

FIG. 13 shows a screen of a processing result by the documented evidence information processing apparatus 1 of the fourth embodiment, and is a screen displayed by the operation terminal 50A of the person in charge of processing. In this example, the voucher is an example of a bill.

In this example, the input support processing unit 14 compares the payee account number of the invoice obtained by OCR processing with the payee account number included in the registered voucher information registered in the registered voucher database 40D. do. As a result of the comparison, if the two transfer account numbers match, the communication unit 30 transmits the registered voucher information including the transfer account number to the operation terminal 50A of the person in charge of processing via the communication network 70. . The registered voucher information here includes basic information and detailed information about the voucher (see FIG. 13). In the operation terminal 50A, the communication unit receives the registered evidenced information from the operation terminal 50A, and the display unit displays at least part of the evidenced information.

Note that the input support processing unit 14 may generate a similar character string similar to the payee's account number obtained by the OCR processing in consideration of erroneous recognition in the OCR processing. In this case, the input support processing unit 14 generates a similar character string whose reliability is equal to or higher than a predetermined value when replacing the character string of the bank account number obtained by OCR processing. The input support processing unit 14 determines whether or not the deposit account number that matches the generated similar character string instead of the character string of the deposit account number obtained by the OCR process is stored in the registered voucher database 40D. . If the matching bank account number is accumulated, the communication unit 30 may transmit the registered documented evidence information including the matching bank account number to the operation terminal 50A together with the reliability information. good.

FIG. 13 shows a case where the two transfer account numbers do not match as a result of comparing the two transfer account numbers. In this case, the input support processing unit 14 generates a similar character string for the payee's account number obtained by OCR processing. The input support processing unit 14 refers to the registered voucher database 40D and acquires registered voucher information including the bank account number that matches the generated similar character string. The communication unit 30 transmits warning instruction information to the operation terminal 50A. This warning instruction information may include warning information indicating that the two payee account numbers do not match, the acquired registered voucher information, and the payee account number obtained by OCR processing. In the operation terminal 50A, the communication unit receives the warning instruction information, and the display unit displays, for example, the warning information, the acquired registered documented evidence information, and the payee account number obtained by OCR processing. do.

The person in charge of processing can see the warning information W1 regarding the transfer account number on the screen of FIG. Furthermore, the screen of FIG. 14 is displayed on the operation terminal 50A when the person in charge of processing presses the details button adjacent to the warning information W1 via the operation unit. FIG. 14 is an example of a screen displayed on the operation terminal 50A. FIG. 14 makes it possible to confirm the cause of the warning. In the display of FIG. 14, the transfer account number entered by the applicant (input content, for example, "2975936"), the transfer account number by OCR processing (OCR result, for example, "2915936"), and the registered voucher database 40D and similar bank account numbers (similar candidates, for example, "2915935") registered in .

The person in charge of processing confirms these three transfer account numbers, and if, for example, the transfer account number entered by the applicant is correct, the transfer account number entered by the applicant is confirmed via the operation unit of the operation terminal 50A. You can point me in the right direction. This instruction information may be transmitted to the voucher information processing device 1 via the communication unit. In the voucher information processing device 1, the communication unit 30 receives this instruction information, and since the transfer account number input by the applicant is correct, the input voucher information including this transfer destination account number is registered in the registered voucher database 40D. You can

The person in charge of processing confirms these three transfer account numbers, and if, for example, the transfer account number entered by the applicant is incorrect, the correct transfer account number (for example, OCR bank account numbers of results or similar candidates) may be indicated. This instruction information may be transmitted to the voucher information processing device 1 via the communication unit. The communication unit 30 of the voucher information processing apparatus 1 may receive this instruction information, replace the payee account number in the input voucher information with the correct payee account number, and register the input voucher information in the registered voucher database 40D.

According to the fourth embodiment, the evidence information processing apparatus 1, for example, when the person in charge of processing determines the validity of the data of the application form for the evidence, the input evidence information, the OCR result, or the registered evidence information, etc. are compared, and based on whether or not this information matches, the processing staff can assist the input work for confirming the application form. Therefore, the documented evidence information processing apparatus 1 can reduce the burden on the person in charge of processing in examining the contents of the application.

Although it is mainly exemplified that it is determined whether or not the payee's account number matches in order to compare the input voucher information and the registered voucher information, the present invention is not limited to this. For example, the input support processing unit 14 may search registered documented evidence information using the invoice number, the recipient of the transfer, the transfer amount, the transfer date, or the like as a key. These items are information that is typically included on a receipt.

Although it has been shown that the payee's account number is registered in the registered voucher database 40D, the present invention is not limited to this. A payee master database that accumulates information on payees may be provided separately from the registered voucher database 40D. The payee master database may include, for example, information on the payee account owner (name, address, telephone number, etc.), payee account number, and the like. When the payee master database manages information on payees, the correct payee account number information described above may be reflected in the payee master database.

(Embodiment 5)
In Embodiment 5, the operation terminal 50B used by the applicant who applies for (submits) the evidence may make a schedule application for notifying the schedule of the application before applying for the evidence. In the scheduled application, the operation terminal 50B inputs information (scheduled voucher information) indicating the scheduled content of each item of the application form of the voucher via the operation unit, and transmits this scheduled input voucher information to the voucher information via the communication unit. Send to the processing device 1 . The scheduled input voucher information is information scheduled to be input as input voucher information, and includes information similar to the voucher information (part of which is scheduled information). Including transfer destination. In addition, the schedule input voucher information also includes the contact information (e.g., e-mail address, etc.) of the operation terminal 50B of the schedule applicant. The control unit of the operation terminal 50B cooperates with the registered voucher database 40D via the communication unit, and inputs (generates) at least part of the scheduled input voucher information based on at least part of the registered voucher information. good too.

The input schedule processing unit 15 of the voucher information processing apparatus 1 receives schedule input voucher information from the operation terminal 50B via the communication unit 30, and registers it in the registered voucher database 40D. The input schedule processing unit 15 performs a process of prompting submission of a voucher (for example, a bill) based on the scheduled input voucher information. As a result, the voucher information processing apparatus 1 can prevent failure to register or create vouchers.

Here, the voucher image and the voucher information based on the OCR processing of the voucher image are not registered at the schedule application stage. For example, when the scheduled application is approved by the superior of the scheduled applicant, the input schedule processing unit 15 generates an input prompt notification prompting uploading of an image of a document related to the scheduled application. The input prompt notification includes, for example, the URL of the data upload destination of the document image. The communication unit 30 transmits the generated input prompt notification to the operation terminal 50B.

FIG. 15 is a screen showing the result of processing by the documented evidence information processing apparatus 1 of the fifth embodiment, and is a screen displayed by the operation terminal 50B of the prospective applicant. In this example, the voucher is an example of a bill.

The prospective applicant confirms the screen of FIG. 15 and refers to, for example, "Waiting for bill registration" in the status on the screen. Then, in the operation terminal 50B, for example, the communication unit may upload the image of the invoice to the specified location based on the operation of the planned applicant to the operation unit. Therefore, an applicant who has made a scheduled application can easily upload an image of the bill using the URL for uploading the data of the bill (certificate). The image of the uploaded voucher (bill) is stored in the registered voucher database 40D in association with the schedule input voucher information related to this schedule application. Scheduled input evidence information and evidence images such as invoices are confirmed by the person in charge of processing at the scheduled processing timing (for example, the scheduled month of processing), and processed by the evidence information processing device 1 or the operation terminal 50A of the person in charge of processing. It is possible.

If there is an error in the scheduled application by the applicant (such as an error in the transfer account number), as described in the fourth embodiment, the input support processing unit 14 creates warning information, and the operation terminals 50A and 50B may be sent to

According to the fifth embodiment, the evidence information processing apparatus 1 has the billing schedule input function as described above, so that the input prompting notification to the scheduled applicant allows the voucher information to be input at a timing different from the input timing of the schedule input voucher information. can assist with the registration of

As described above, the evidence information processing apparatus 1 of the above-described embodiment includes an acquisition unit (for example, the image acquisition unit 111) that acquires the first image M1 including the evidence, and the first evidence and the A detection unit (e.g., the evidence position detection unit 112) that detects a plurality of evidences including the second evidence, and a clipping unit (e.g., image clipping unit) that cuts out a second image including the first evidence from the first image M1. 115), and when the first voucher and the second voucher overlap in the second image, the region of the second voucher (for example, the processing region MX) cannot be recognized from the second image. and an output unit (eg, image output unit 118) that outputs the second image processed by the processing unit (for example, image processing unit 116).

As a result, the evidence information processing apparatus 1 can detect each voucher individually by, for example, artificial intelligence even if a plurality of (for example, two) vouchers overlap. Therefore, the evidence information processing apparatus 1, for example, when a plurality of vouchers overlap, particularly when one of the plurality of vouchers completely overlaps such that the entirety of one of the plurality of vouchers is included in the other, each voucher can be detected individually. As a result, the voucher information processing apparatus 1 can reduce the burden on the user when detecting the voucher, and can shorten the time required to complete the reading of the content of the voucher.

Also, the clipping unit may clip a third image including the second documentary evidence from the first image M1. The output unit may output the third image. As a result, the documented evidence information processing apparatus 1 cuts out the image of each documented document included in the first image M1 separately. can suppress the decrease in

Also, the documented evidence information processing apparatus 1 may include a recognition unit that recognizes characters of the second image and the third image. As a result, even when a plurality of vouchers overlap, the voucher information processing apparatus 1 can derive separate images containing each of the plurality of vouchers. A decrease in recognition accuracy can be suppressed.

Also, the second image may have a larger image range than the third image. As a result, even when a plurality of vouchers overlap and the sizes of the plurality of images (image ranges), that is, the sizes of the plurality of vouchers are different, the voucher information processing apparatus 1 can display the characters when the plurality of vouchers overlap. A decrease in recognition accuracy can be suppressed.

Further, when the overlapping rate of the first evidence and the second evidence in the second image is equal to or higher than the threshold, the processing unit processes the area of the second evidence from the second image so that the characters cannot be recognized. You can As a result, the voucher information processing apparatus 1 performs processing to render characters unrecognizable only when the overlap rate of a plurality of vouchers is high and the probability of occurrence of character recognition errors is high. Therefore, the voucher information processing apparatus 1 can improve the character recognition accuracy of a plurality of vouchers while reducing the processing load related to processing.

Further, the evidence information processing apparatus 1 of the above embodiment includes an acquisition unit that acquires a fourth image including the evidence, a detection unit that detects the evidence in the fourth image, and the evidence detected by the detection unit, When there are a plurality of character areas including a first character area and a second character area with different character orientations, a fifth image including the first character area and a second image including the second character area. 6 images, and an output unit for outputting the fifth image and the sixth image.

As a result, the voucher information processing apparatus 1 can individually detect each character region by artificial intelligence or the like even if a single voucher includes a plurality of character regions with different character orientations. Therefore, character recognition accuracy can be improved for characters included in a plurality of character areas by including each character area in a separate image for character recognition. As a result, the voucher information processing apparatus 1 can reduce the burden on the user when detecting the voucher, and can shorten the time required to complete the reading of the content of the voucher.

The voucher information processing apparatus 1 may also include a correction unit (for example, the image rotation correction unit 117) that rotates and corrects the fifth image and the sixth image separately. The output unit may output the rotation-corrected fifth and sixth images.

As a result, the documented evidence information processing apparatus 1 corrects the rotation of the fifth image and the sixth image so that the orientations of characters included in a plurality of character regions are aligned, thereby facilitating character recognition by character recognition processing. Become. Therefore, the voucher information processing apparatus 1 can improve the accuracy of character recognition. In addition, since the occurrence of character recognition errors is suppressed, it becomes unnecessary to manually adjust the angle of the voucher image in order to accurately perform character recognition. Therefore, the documented evidence information processing apparatus 1 can reduce the user's trouble and shorten the time until character recognition is completed.

Further, the documented evidence information processing apparatus 1 may include a recognition unit that recognizes characters of the fifth image and the sixth image. As a result, the voucher information processing apparatus 1 can correct the angle of each image by deriving separate images containing each of the plurality of character regions even when the direction of the characters in the plurality of character regions in the voucher is different. Since the character string can be output as a frontal image with no angle, character recognition is performed based on each derived image, thereby suppressing a decrease in character recognition accuracy.

The voucher information processing apparatus 1 may also include an accumulation unit (for example, the registered voucher database 40D) for accumulating first voucher information related to vouchers. The evidence information processing apparatus 1 obtains the first evidence information accumulated by the accumulation unit and character recognition of at least one of the second image, the third image, the fifth image, and the sixth image. A determination unit (for example, duplicate registration confirmation processing unit 13) that determines whether or not the received second evidence information is the same evidence information may be provided. The evidence information processing apparatus 1 includes a transmission unit (for example, the communication unit 30) that transmits an instruction to present warning information when it is determined that the first evidence information and the second evidence information are the same evidence information. and may be provided.

As a result, the evidence information processing apparatus 1 transmits an instruction to present warning information to, for example, the operation terminal 50A, thereby combining the second evidence information obtained this time with the previously obtained and accumulated first evidence information. You can check the consistency of the evidence information of As a result, the voucher information processing apparatus 1 can prevent the same voucher information from being registered multiple times by the person in charge of processing, and can improve processing accuracy related to accounting.

Also, the first voucher information may include first voucher identification information for identifying the voucher. The second voucher information may include second voucher identification information that identifies the voucher. When the first evidenced document identification information and the second evidenced evidenced identification information are the same, the determining unit may determine that the first evidenced evidenced information and the second evidenced evidenced information are the same evidenced evidenced information.

As a result, the evidence information processing apparatus 1 can determine the consistency between the second evidence information obtained this time and the first evidence information obtained in the past, using the evidence identification information as a key.

Various embodiments have been described above with reference to the drawings, but it goes without saying that the present disclosure is not limited to such examples. It is obvious that a person skilled in the art can conceive of various modifications or modifications within the scope described in the claims, and these also belong to the technical scope of the present disclosure. Understood. Also, the components in the above embodiments may be combined arbitrarily without departing from the gist of the disclosure.

Also, in the above embodiments, the processor may be physically configured in any way. Moreover, if a programmable processor is used, the content of processing can be changed by changing the program, so that the degree of freedom in designing the processor can be increased. The processor may be composed of one semiconductor chip, or physically composed of a plurality of semiconductor chips. When configured with a plurality of semiconductor chips, each control of the above embodiments may be realized by separate semiconductor chips. In this case, it can be considered that the plurality of semiconductor chips constitutes one processor. Also, the processor may be composed of a member (capacitor, etc.) having a function different from that of the semiconductor chip. Also, one semiconductor chip may be configured to implement the functions of the processor and other functions. Also, a plurality of processors may be composed of one processor.

INDUSTRIAL APPLICABILITY The present disclosure is useful for a voucher information processing device, a voucher information processing method, a program, and the like, which can reduce the user's effort when detecting vouchers and shorten the time required to complete reading of the contents of vouchers.

1 Voucher information processing device 10 Image processing unit 11 Overlapping voucher processing unit 12 OCR processing unit 13 Duplicate registration confirmation processing unit 14 Input support processing unit 15 Input schedule processing unit 20 Voucher information management unit 30 Communication unit 40 Storage unit 40D Registered voucher database 50 Operation terminal 60 Scanner 70 Communication network 111 Image acquisition unit 112 Voucher position detection unit 113 Size relation calculation unit 114 Duplication ratio calculation unit 115 Image clipping unit 116 Image processing unit 117 Image rotation correction unit 118 Image output unit 200 Voucher information processing system C1 First character area C2 Second character area M1 First image M2 Second image M3 Third image M4 Fourth image M5 Fifth image M6 Sixth image MX Processing area V1 First voucher V2 Second Evidence

Claims (13)

an acquisition unit that acquires a first image containing a voucher;
a detection unit that detects a plurality of vouchers including a first voucher and a second voucher in the first image;
a clipping unit for clipping a second image containing the first voucher from the first image;
a processing unit that processes an area of the second voucher from the second image so that characters cannot be recognized when the first voucher and the second voucher overlap in the second image;
an output unit that outputs the second image processed by the processing unit;
A voucher information processing device. The clipping unit clips a third image containing the second document from the first image,
The output unit outputs the third image,
The voucher information processing apparatus according to claim 1. further comprising a recognition unit that recognizes characters in the second image and the third image;
The voucher information processing apparatus according to claim 2. wherein the second image has a larger image range than the third image;
The voucher information processing apparatus according to claim 2 or 3. When the overlapping rate of the first evidence and the second evidence in the second image is equal to or higher than a threshold, the processing unit removes the area of the second evidence from the second image as text. process unrecognizable,
A voucher information processing apparatus according to any one of claims 2 to 4. an acquisition unit that acquires a fourth image containing the voucher;
a detection unit that detects evidence in the fourth image;
When the evidence detected by the detection unit includes a plurality of character areas including a first character area and a second character area having different character orientations, a fifth character area including the first character area is detected. and a sixth image including the second character region;
an output unit that outputs the fifth image and the sixth image;
A voucher information processing device. further comprising a correction unit that rotates and corrects the fifth image and the sixth image separately;
The output unit outputs the rotation-corrected fifth image and the sixth image.
The voucher information processing apparatus according to claim 6. further comprising a recognition unit that recognizes characters in the fifth image and the sixth image;
The voucher information processing apparatus according to claim 6 or 7. an accumulation unit for accumulating first voucher information related to vouchers;
Determining whether or not the first evidenced information accumulated by the accumulation unit and the second evidenced information obtained by character recognition of the second image or the third image are the same evidenced information. a determination unit for
a transmission unit that transmits an instruction to present warning information when it is determined that the first evidenced information and the second evidenced information are the same evidenced information,
A voucher information processing apparatus according to any one of claims 2 to 5. an accumulation unit for accumulating first voucher information related to vouchers;
Determining whether or not the first evidenced information accumulated by the accumulation unit and the second evidenced information obtained by character recognition of the fifth image or the sixth image are the same evidenced information. a determination unit for
a transmission unit that transmits an instruction to present warning information when it is determined that the first evidenced information and the second evidenced information are the same evidenced information,
A voucher information processing apparatus according to any one of claims 6 to 8. The first evidence information includes first evidence identification information that identifies the evidence,
The second evidence information includes second evidence identification information that identifies the evidence,
When the first evidenced document identification information and the second evidenced evidence identification information are the same, the determination unit determines that the first evidenced evidenced information and the second evidenced evidenced information are the same evidenced evidenced information. ,
The voucher information processing apparatus according to claim 9 or 10. obtaining a first image containing the voucher;
detecting a plurality of evidences including a first evidence and a second evidence in the first image;
cropping a second image containing the first voucher from the first image;
If the first voucher and the second voucher overlap in the second image, processing the area of the second voucher from the second image so that the characters cannot be recognized;
outputting the processed second image;
Evidence information processing method having A program for causing a computer to execute the voucher information processing method according to claim 12.

Images