OCR target area position acquisition system and OCR target area position acquisition program

Abstract

To provide an OCR target area location acquisition system, an OCR target area location acquisition program, an actual document, an actual document generation system, and an actual document generation program that can improve accuracy of OCR processing.SOLUTION: An OCR system acquires a position of an image code in a document image (S121), acquires data indicated by the image code (S122), and acquires, on the basis of a position of the image code in the document image acquired in S121 and a position of the image code in a document included in the data acquired in S122, and a position of a handwriting input field in the document included in the data acquired in S122, a position of the handwriting input field in the document image (S123-S124).SELECTED DRAWING: Figure 9

Description

[Technical field]

[0001] The present invention relates to an OCR target area position acquisition system that acquires the position in a document image of an OCR target area as an area in which OCR (Optical Character Recognition) processing should be performed in a document image as an image of a document, an OCR target area position acquisition program, a real document, a real document generation system, and a real document generation program. [Background technology]

[0002] Conventionally, there is known a technique for performing OCR processing on an OCR target area in a document image (see Patent Document 1). [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2015-211428 A Summary of the Invention [Problem to be solved by the invention]

[0004] When a document image to be subjected to OCR processing is misaligned when it is read and generated from an actual document (hereinafter referred to as "actual document") made of a recording medium such as paper, the position of the document in the document image is misaligned relative to an ideal document image in which no misalignment occurs when it is read and generated from the actual document. When a document image to be subjected to OCR processing is misaligned relative to an ideal document image, the OCR target area in the document image to be subjected to OCR processing is also misaligned relative to the OCR target area in the ideal document image. In conventional techniques, when a position of the OCR target area in the document image to be subjected to OCR processing is misaligned relative to the OCR target area in the ideal document image, there is a problem in that the accuracy of the OCR processing is reduced.

[0005] Therefore, an object of the present invention is to provide an OCR target area position acquisition system, an OCR target area position acquisition program, a document real object, a document real object generation system, and a document real object generation program that can improve the accuracy of OCR processing. [Means for solving the problem]

[0006] The OCR target area position acquisition system of the present invention comprises an image code position acquisition unit that acquires the position of the image code in a document image, which is an image of a document to which data has been assigned by an image code, a data acquisition unit that acquires data indicated by the image code, and an OCR target area position acquisition unit that acquires the position in the document image of an OCR target area, which is an area in the document image to which OCR processing is to be performed, wherein the data assigned to the document by the image code includes image code position data including the position of the image code in the document, and OCR target area position data including the position of the OCR target area in the document, and the OCR target area position acquisition unit acquires the position of the OCR target area in the document image based on the position of the image code in the document image acquired by the image code position acquisition unit, the position of the image code in the document included in the image code position data acquired by the data acquisition unit, and the position of the OCR target area in the document included in the OCR target area position data acquired by the data acquisition unit.

[0007] With this configuration, the OCR target area position acquisition system of the present invention acquires the position of the OCR target area in a document image based on the position of the image code in the document image, the position of the image code in the document contained in the image code position data indicated by the image code in the document image, and the position of the OCR target area in the document contained in the OCR target area position data indicated by the image code in the document image, thereby being able to identify the OCR target area in the document image with high accuracy, and as a result, the accuracy of OCR processing can be improved.

[0008] The OCR target area position acquisition program of the present invention causes a computer to realize an image code position acquisition unit that acquires the position of the image code in a document image, which is an image of a document to which data has been assigned by an image code, a data acquisition unit that acquires data indicated by the image code, and an OCR target area position acquisition unit that acquires the position in the document image of an OCR target area, which is an area in the document image to which OCR processing is to be performed, wherein the data assigned to the document by the image code includes image code position data including the position of the image code in the document, and OCR target area position data including the position of the OCR target area in the document, and the OCR target area position acquisition unit acquires the position of the OCR target area in the document image based on the position of the image code in the document image acquired by the image code position acquisition unit, the position of the image code in the document included in the image code position data acquired by the data acquisition unit, and the position of the OCR target area in the document included in the OCR target area position data acquired by the data acquisition unit.

[0009] With this configuration, a computer executing the OCR target area position acquisition program of the present invention acquires the position of the OCR target area in a document image based on the position of the image code in the document image, the position of the image code in the document contained in the image code position data indicated by the image code in the document image, and the position of the OCR target area in the document contained in the OCR target area position data indicated by the image code in the document image, thereby being able to identify the OCR target area in the document image with high accuracy, and as a result, the accuracy of OCR processing can be improved.

[0010] The actual document of the present invention is an actual document to which data has been assigned by an image code, and is characterized in that the data assigned to the document by the image code includes image code position data including a position of the image code in the document, and OCR target area position data including a position in the document of an OCR target area as an area in which OCR processing is to be performed in an image of the document.

[0011] With this configuration, the actual document of the present invention is assigned image code position data including the position of the image code in the document, and OCR target area position data including the position of the OCR target area in the document by means of an image code, so that the OCR target area position acquisition system can identify the OCR target area in the document image with high accuracy, thereby improving the accuracy of OCR processing.

[0012] The document actual generation system of the present invention comprises a document actual generation unit that generates a document actual as an actual document to which data has been assigned by an image code, and the data assigned to the document by the image code includes image code position data including a position of the image code in the document, and OCR target area position data including a position in the document of an OCR target area as an area in which OCR processing is to be performed in an image of the document.

[0013] With this configuration, the document actual generation system of the present invention generates a document actual in which image code position data, including the position of the image code in the document, and OCR target area position data, including the position of the OCR target area in the document, are assigned by an image code, thereby making it possible to generate a document actual that can improve the accuracy of OCR processing.

[0014] The document actual object generation program of the present invention is characterized in that a document actual object generation unit is implemented in a computer to generate a document actual object as an actual object of a document to which data has been assigned by an image code, and the data assigned to the document by the image code includes image code position data including a position of the image code in the document, and OCR target area position data including a position in the document of an OCR target area as an area in which OCR processing is to be performed in an image of the document.

[0015] With this configuration, a computer executing the document actual generation program of the present invention generates a document actual in which image code position data, including the position of the image code in the document, and OCR target area position data, including the position of the OCR target area in the document, are assigned by an image code, thereby making it possible to generate a document actual that can improve the accuracy of OCR processing. Effect of the Invention

[0016] The OCR target area position acquisition system, the OCR target area position acquisition program, the document real object, the document real object generation system, and the document real object generation program of the present invention can improve the accuracy of OCR processing. [Brief description of the drawings]

[0017] [Figure 1] FIG. 1 is a block diagram of a system according to an embodiment of the present invention. [Diagram 2] FIG. 2 is a block diagram of the OCR system shown in FIG. 1 when configured by one computer. [Diagram 3] FIG. 2 is a block diagram of the MFP shown in FIG. [Figure 4] FIG. 2 is a block diagram of a user terminal shown in FIG. [Diagram 5] 3 is a flowchart of the operation of the OCR system shown in FIG. 2 when causing the MFP to print a document. [Figure 6] FIG. 6 is a diagram showing an example of a document created in the operation shown in FIG. 5. [Figure 7]FIG. 7 is a diagram showing an example of the document shown in FIG. 6 to which an image code has been added. [Figure 8] FIG. 8 is a diagram showing an example of data indicated by the image code shown in FIG. 7. [Figure 9] 3 is a flow chart of the operation of the OCR system shown in FIG. 2 when extracting information from a document image. [Figure 10] Fig. 10(a) is a diagram showing an example of a handwritten input field and an image code in an actual document that is the target of the operation shown in Fig. 9. Fig. 10(b) is a diagram showing an example of a handwritten input field and an image code in a document image read from the actual document shown in Fig. 10(a). DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0018] Hereinafter, an embodiment of the present invention will be described with reference to the drawings.

[0019] First, the configuration of a system according to an embodiment of the present invention will be described.

[0020] FIG. 1 is a block diagram of a system 10 according to the present embodiment.

[0021] As shown in FIG. 1, the system 10 includes an OCR (Optical Character Recognition) system 20 that extracts information from a document image, an MFP (Multifunction Peripheral) 30 as an image reading device that reads a document image from an actual document composed of a recording medium such as paper, and a user terminal 40 used by a user.

[0022] The OCR system 20 may be configured with one computer, or may be configured with multiple computers.

[0023] The OCR system 20 and the MFP 30 can communicate with each other via a network such as a LAN (Local Area Network) or the Internet, or directly by wire or wirelessly without a network. Similarly, the OCR system 20 and the user terminal 40 can communicate with each other via a network such as a LAN or the Internet, or directly by wire or wirelessly without a network.

[0024] FIG. 2 is a block diagram of an OCR system 20 configured using a single computer.

[0025] The OCR system 20 shown in Figure 2 includes an operation unit 21 which is an operation device such as a keyboard or mouse through which various operations are input, a display unit 22 which is a display device such as an LCD (Liquid Crystal Display) that displays various information, a communication unit 23 which is a communication device that communicates with external devices via a network such as a LAN or the Internet, or directly by wire or wirelessly without using a network, a memory unit 24 which is a non-volatile memory device such as a semiconductor memory or an HDD (Hard Disk Drive) that stores various information, and a control unit 25 which controls the entire OCR system 20.

[0026] The storage unit 24 stores an OCR program 24a for executing OCR processing. The OCR program 24a may be installed in the OCR system 20 during the manufacturing stage of the OCR system 20, or may be additionally installed in the OCR system 20 from an external storage medium such as a Universal Serial Bus (USB) memory, or may be additionally installed in the OCR system 20 from a network.

[0027] The control unit 25 includes, for example, a CPU (Central Processing Unit), a ROM (Read Only Memory) that stores programs and various data, and a RAM (Random Access Memory) that serves as a memory used as a working area for the CPU of the control unit 25. The CPU of the control unit 25 executes programs stored in the storage unit 24 or the ROM of the control unit 25.

[0028] The control unit 25 executes the OCR program 24a to realize a document real object generation unit 25a that generates a document real object to which data is added by an image code such as a one-dimensional code, a two-dimensional code, etc. Therefore, the OCR system 20 and the OCR program 24a respectively constitute the document real object generation system and the document real object generation program of the present invention.

[0029] By executing the OCR program 24a, the control unit 25 realizes an image code position acquisition unit 25b that acquires the position of an image code in a document image, a data acquisition unit 25c that acquires data indicated by the image code, and an OCR target area position acquisition unit 25d that acquires the position in the document image of a handwritten input field (described later) as an OCR target area where OCR processing should be performed in the document image. Therefore, the OCR system 20 and the OCR program 24a respectively constitute the OCR target area position acquisition system and the OCR target area position acquisition program of the present invention.

[0030] The control unit 25 executes the OCR program 24a to realize an OCR processing unit 25e that extracts information from a handwritten input field in a document image by OCR processing.

[0031] FIG. 3 is a block diagram of MFP 30.

[0032] The MFP 30 shown in FIG. 3 comprises an operation unit 31 which is an operation device such as buttons through which various operations are input, a display unit 32 which is a display device such as an LCD that displays various information, a printer 33 which is a printing device that prints images on a recording medium such as paper, a scanner 34 which is a reading device that reads images from a document, a fax communication unit 35 which is a fax device that performs fax communication with an external facsimile machine not shown via a communication line such as a public telephone line, a communication unit 36 ​​which is a communication device that communicates with an external device via a network such as a LAN or the Internet, or directly by wired or wireless means without going through a network, a memory unit 37 which is a non-volatile memory device such as a semiconductor memory or HDD that stores various information, and a control unit 38 which controls the entire MFP 30.

[0033] The control unit 38 includes, for example, a CPU, a ROM that stores programs and various data, and a RAM as a memory used as a working area for the CPU of the control unit 38. The CPU of the control unit 38 executes the programs stored in the storage unit 37 or the ROM of the control unit 38. Communication can be direct by wire or wirelessly.

[0034] FIG. 4 is a block diagram of the user terminal 40. As shown in FIG.

[0035] 4 includes an operation unit 41 which is an operation device such as a keyboard or mouse for inputting various operations, a display unit 42 which is a display device such as an LCD for displaying various information, a communication unit 43 which is a communication device for communicating with an external device via a network such as a LAN or the Internet, or directly by wire or wirelessly without using a network, a storage unit 44 which is a non-volatile storage device such as a semiconductor memory or HDD for storing various information, and a control unit 45 which controls the entire user terminal 40. The user terminal 40 is formed of, for example, a PC (Personal Computer).

[0036] The control unit 45 includes, for example, a CPU, a ROM that stores programs and various data, and a RAM as a memory used as a working area for the CPU of the control unit 45. The CPU of the control unit 45 executes the programs stored in the storage unit 44 or the ROM of the control unit 45.

[0037] The operation of the system 10 will now be described.

[0038] First, the operation of the OCR system 20 when the MFP 30 is caused to print a document will be described.

[0039] FIG. 5 is a flowchart of the operation of the OCR system 20 when causing the MFP 30 to print a document.

[0040] A user of the OCR system 20 (hereinafter referred to as an "operator") can instruct the OCR system 20 to create a document via the operation unit 41 of the user terminal 40. Therefore, as shown in Fig. 5, the document actual generation unit 25a of the OCR system 20 creates a document according to the instruction from the operator (S101).

[0041] FIG. 6 is a diagram showing an example of the document 50 created in S101.

[0042] A document 50 shown in FIG. 6 includes various types of text 51, various types of ruled lines 52, various types of images 53, and a field 54 for inputting handwritten characters (hereinafter referred to as a "handwritten input field").

[0043] After instructing the OCR system 20 to create a document, the worker can instruct the OCR system 20 via the operation unit 41 of the user terminal 40 as to the data to be added to the document by the image code. Therefore, as shown in Fig. 5, after the process of S101, the document actual generation unit 25a of the OCR system 20 adds an image code corresponding to the data according to the instruction from the worker to the document created in S101 (S102). Here, for example, a QR code (registered trademark) can be adopted as the image code. In the following, the description will be given assuming that the image code is a QR code.

[0044] FIG. 7 is a diagram showing an example of a document 50 to which an image code 55 has been added in S102.

[0045] As shown in FIG. 7, a document 50 may be provided with a plurality of image codes 55 according to the size of the data in response to an instruction from an operator.

[0046] Fig. 8 is a diagram showing an example of data indicated by image code 55. The data shown in Fig. 8 is partially omitted.

[0047] As shown in FIG. 8, the data assigned to the document by the image code in S102 includes data (hereinafter referred to as "auto-index data") 61 for determining the storage destination of information extracted by OCR processing and for compiling various types of information, data relating to handwritten input field 54 (hereinafter referred to as "handwritten input field data") 62, data indicating the position of image code 55 in document 50 (hereinafter referred to as "image code position data") 63, data for reproducing text 51 (hereinafter referred to as "text data") 64, data for reproducing lines 52 (hereinafter referred to as "line data") 65, and data for reproducing image 53 (hereinafter referred to as "image data") 66.

[0048] The auto-index data 61 may include, for example, identification information and a value for each data. The value of the auto-index data 61 may be, for example, any of the texts 51. For example, "Data:CarLavel=Vehicle Number" indicates that the value of the data whose identification information is "CarLavel" is "Vehicle Number".

[0049] The handwriting input field data 62 may include, for example, identification information and the position and size in the document 50 for each handwriting input field. The handwriting input field data 62 includes the position in the document 50 of the handwriting input field as an OCR target area, and constitutes the OCR target area position data of the present invention. Note that, depending on the handwriting input field, the handwriting input field data 62 may include the type of characters used. For example, "InputArea:Name=(49,53,182,8),hint:[a-z0-9]" indicates that the position in the document 50 of the upper left corner of the handwriting input field having the identification information "Name" is 49 to the right and 53 down from the upper left corner of the document 50, the size of this character input field is 182 in the horizontal direction and 8 in the vertical direction, and the type of characters used in this character input field is only lowercase English letters and numbers.

[0050] The image code position data 63 may include, for example, the positions in the document 50 of the upper left corner, upper right corner, and lower left corner of the image code 55. Note that when multiple image codes 55 are added to the document 50, the image code position data 63 may be data indicating only the position of a specific image code 55, such as the position of the leftmost image code 55.

[0051] The data 64 for the text may include, for each piece of text, its location in the document 50, etc.

[0052] The ruled line data 65 may include the position of each ruled line in the document 50. For example, "Line: (5, 17)-(5, 134)" indicates a ruled line that connects a position 5 to the right and 17 down from the top left corner of the document 50 with a position 5 to the right and 134 down from the top left corner of the document 50.

[0053] Image data 66 may include, for example, identification information and a position in document 50 for each image. For example, "Image:xx=(218,8,8,8)" indicates that the position in document 50 of the upper left corner of an image with identification information "xx" is 218 to the right and 8 down from the upper left corner of document 50, and that the size of this image is 8 in the left-right direction and 8 in the up-down direction.

[0054] 5, after the process of S102, the document actual generation unit 25a of the OCR system 20 instructs the MFP 30 via the communication unit 23 of the OCR system 20 to print the document to which the image code was added in S102 (S103). When the control unit 38 of the MFP 30 receives the instruction to print the document from the OCR system 20 via the communication unit 36, it causes the printer 33 to print the document according to the received instruction.

[0055] After the process of S103, the document-entity generation unit 25a ends the operation shown in FIG.

[0056] In the above, the worker executes various instructions to the OCR system 20 via the operation unit 41 of the user terminal 40, but instead of the operation unit 41 of the user terminal 40, the worker may execute various instructions to the OCR system 20 via the operation unit 31 of the MFP 30.

[0057] The worker distributes the actual document printed by the operation shown in Fig. 5 to, for example, multiple people, and asks the recipients to handwrite appropriate characters in the handwriting input field of the actual document. After the recipients handwrite appropriate characters in the handwriting input field of the distributed actual document, they return the actual document with the appropriate characters handwritten in the handwriting input field to the worker.

[0058] The operation of OCR system 20 in extracting information from a document image will now be described.

[0059] FIG. 9 is a flow chart of the operation of OCR system 20 in extracting information from a document image.

[0060] The worker sets the actual document returned by the recipient of the actual document in the scanner 34 of the MFP 30, and can instruct the MFP 30, for example, via the operation unit 31 of the MFP 30, to extract information from the actual document. When the control unit 38 of the MFP 30 is instructed to extract information from the actual document, the control unit 38 reads a document image from the actual document set in the scanner 34 with the scanner 34, and instructs the OCR system 20, via the communication unit 36 ​​of the MFP 30, to extract information from the document image. Here, the instruction to extract information from the document image (hereinafter referred to as an "information extraction instruction") includes the document image that is the target of the information extraction instruction. When the control unit 25 of the OCR system 20 receives the information extraction instruction via the communication unit 23, it executes the operation shown in FIG. 9.

[0061] As shown in FIG. 9, the image code position acquiring unit 25b acquires the position of an image code in a document image that is the target of an information extraction instruction (S121).

[0062] After the process of S121, the data acquisition unit 25c acquires various pieces of information indicated by image codes included in the document image that is the target of the information extraction instruction (S122).

[0063] After the process of S122, the OCR target area position acquisition unit 25d calculates a transformation matrix M from the actual document to the document image that is the target of the information extraction instruction (S123).

[0064] The method for calculating the transformation matrix M in S123 will be described below.

[0065] Fig. 10(a) is a diagram showing an example of a handwritten input field and an image code in an actual document 70 that is the target of the operation shown in Fig. 9. Fig. 10(b) is a diagram showing an example of a handwritten input field and an image code in a document image read from the actual document 70 shown in Fig. 10(a).

[0066] 10(a), the origin may be any point for the following calculations, but in the following explanation, it is assumed to be the upper left corner. The upper left corner, upper right corner, and lower left corner of the handwriting input field 71 in the actual document 70 are respectively designated as Q0, Q1, and Q2, and the upper left corner, upper right corner, and lower left corner of the image code 72 in the actual document 70 are respectively designated as P0, P1, and P2.

[0067] 10(b), the origin is the upper left corner. The upper left corner, upper right corner, and lower left corner of the handwriting input field 71 in the document image 80 are respectively Q0', Q1', and Q2', and the upper left corner, upper right corner, and lower left corner of the image code 72 in the document image 80 are respectively P0', P1', and P2'.

[0068] The coordinates of P0 in the left and right direction are P 0x Let P0 be the vertical coordinate of P 0y Then, the coordinates of P0 can be expressed as shown in Expression 1. Similarly, the coordinates of P1, P2, P0', P1', and P2' can be expressed as shown in Expression 1. Note that the coordinates shown in Expression 1 are a homogeneous coordinate system used in affine transformation.

number

[0069] If the transformation from the handwritten input field 71 and image code 72 in the actual document 70 shown in FIG. 10(a) to the handwritten input field 71 and image code 72 in the document image 80 shown in FIG. 10(b) is a transformation that combines rotation, enlargement, and translation, the transformation matrix M can be expressed as shown in Equation 2.

number

[0070] Here, P0', P1', and P2' can be expressed as shown in Equation 3 using P0, P1, P2, and M.

number

[0071] From the formulas 1 to 3, the formula shown in the formula 4 is established.

number

[0072] By summarizing the equation shown in Equation 4 for each x and y, the equation shown in Equation 5 can be obtained.

number

[0073] From the formula 5, the formula 6 can be obtained.

number

[0074] From the formula 6, the formula 7 can be obtained. In the formula 7, -1 in the upper right corner of the matrix means an inverse matrix. In the formula 7, cases where an inverse matrix does not exist are not anticipated.

number

[0075] Based on the equation shown in Equation 2 and the equation shown in Equation 7, the transformation matrix M can be expressed as shown in Equation 8. In Equation 8, T at the top right of the matrix means a transposed matrix. In Equation 8, −1 at the top right of the matrix means an inverse matrix. In Equation 8, cases where an inverse matrix does not exist are not anticipated. Here, P 0x , P 0y , P 1x , P 1y , P 2x , P 2y is indicated by the image code position data obtained by the process of S122. 0x ´, P 0y ´, P 1x ´, P 1y ´, P 2x´, P 2y ' is obtained by the process of S121.

number

[0076] As shown in FIG. 9, after processing of S123, the OCR target area position acquisition unit 25d calculates the position of the handwriting input field in the document image that is the target of the information extraction instruction, based on the transformation matrix M calculated in S123 and the position of the handwriting input field in the actual document (S124).

[0077] That is, the positions Q0', Q1', and Q2' of the handwriting input fields in the document image that is the target of the information extraction instruction can be expressed as shown in Equation 9 using the transformation matrix M calculated in S123 and the positions Q0, Q1, and Q2 of the handwriting input fields in the actual document. Here, the coordinates of Q0, Q1, and Q2 are indicated by the handwriting input field data acquired by the process of S122.

number

[0078] 9, after the process of S124, the OCR processing unit 25e extracts information from the handwriting input field in the document image that is the target of the information extraction instruction by OCR processing based on the position calculated in S124 (S125). Here, the OCR processing unit 25e may use the "type of characters used" indicated by the handwriting input field data acquired by the process of S122 in the OCR processing.

[0079] After the process of S125, the OCR processing unit 25e stores the information extracted in S125, for example, in the storage unit 24 (S126). Here, the OCR processing unit 25e may store at least one of the data acquired by the process of S122 together with the information extracted in S125. For example, the OCR processing unit 25e stores the text data, the line data, and the image data acquired by the process of S122 together with the information extracted in S125, thereby making it possible to reproduce the actual document based on the stored data. In addition, the OCR processing unit 25e may adopt a storage destination according to the information indicated by the auto-index data acquired by the process of S122 as the storage destination of the information in S126.

[0080] After the process of S126, the control unit 25 ends the operation shown in FIG.

[0081] As described above, OCR system 20 acquires the position of the handwriting input field in the document image based on the position of the image code in the document image, the position of the image code in the document included in the image code position data indicated by the image code in the document image, and the position of the handwriting input field in the document included in the handwriting input field data indicated by the image code in the document image (S123-S124), so that it is possible to specify the handwriting input field as the OCR target area in the document image with high accuracy, thereby improving the accuracy of the OCR process. Therefore, OCR system 20 can, for example, improve the efficiency of data entry work in which information handwritten on a document is entered as data into a computer.

[0082] Since the actual document 70 is provided with image code position data including the position of the image code in the document and handwritten input field data including the position of the handwritten input field in the document by image codes, the OCR system 20 can identify the handwritten input field in the document image with high accuracy, thereby improving the accuracy of the OCR processing.

[0083] Image code position data including the position of the image code in the document and handwriting input field data including the position of the handwriting input field in the document are assigned to each actual document by an image code. Therefore, even when extracting information continuously from a plurality of document images generated from a plurality of types of actual documents, such as actual documents with a plurality of layouts, the OCR system 20 can identify the handwriting input field in each document image with high accuracy.

[0084] The OCR system 20 generates an actual document in which image code position data including the position of the image code in the document and handwriting input field data including the position of the handwriting input field in the document are assigned by an image code (S101 to S103), and therefore it is possible to generate an actual document that can improve the accuracy of OCR processing.

[0085] In the above, a method of reading a document image from an actual document using a scanner has been described as a method of acquiring a document image by an MFP. However, a method of acquiring a document image by an MFP other than a method of reading a document image from an actual document using a scanner may be adopted. For example, the MFP may acquire a document image by receiving the document image via a fax communication unit.

[0086] In the above, an MFP has been described as an example of an image reading device. However, the image reading device may be a device other than an MFP equipped with a scanner, such as a dedicated scanner, or may be a device other than a device equipped with a scanner. For example, the image reading device may be a device such as a mobile terminal equipped with a camera that captures an actual document and generates a document image. Here, a document image generated from an actual document by a device equipped with a camera is more likely to have a greater deviation in the position of the document in the document image relative to an ideal document image, compared to a document image generated from an actual document by a device equipped with a scanner. Therefore, the present invention is more likely to be needed when a document image is generated from an actual document by a device equipped with a camera, compared to a document image generated from an actual document by a device equipped with a scanner.

[0087] In the above, the OCR system and the image reading device exist separately, but for example, the image reading device may have an OCR system built therein.

[0088] In the above, the OCR system and the user terminal exist separately. However, for example, the user terminal may have an OCR system built in it.

[0089] In the above, the OCR system performs the OCR process by itself. However, the OCR system may request the OCR process from an external service such as a cloud service.

[0090] The present invention may be adopted by companies that have introduced, for example, ECM (Enterprise Content Management), RPA (Robotic Process Automation), and the like. [Explanation of symbols]

[0091] 20 OCR system (OCR target area position acquisition system, document real-world generation system) 24a OCR program (OCR target area position acquisition program, document actual object generation program) 25a Document Reality Generation Unit 25b Image code position acquisition unit 25c Data Acquisition Section 25d OCR target area position acquisition unit 50 Documents 54 Handwritten input field (OCR target area) 55 Image Code 62 Handwritten input field data (OCR target area position data) 63 Image code position data 70 Original Documents 71 Handwritten input field (OCR target area) 72 Image Code 80 Document Images

Claims

1. an image code position acquisition unit that acquires a position of the image code in a document image, which is an image of a document to which data is added by the image code; a data acquisition unit that acquires data represented by the image code; an OCR target area position acquisition unit that acquires a position in the document image of an OCR target area, which is an area in the document image where OCR processing is to be performed; Equipped with The data provided to the document by the image code is image code location data including a location of the image code in the document; OCR target region location data including a location of the OCR target region in the document; and Including, The OCR target area position acquisition unit a position of the image code in the document image acquired by the image code position acquisition unit; the position of the image code in the document, which is included in the image code position data acquired by the data acquisition unit; Calculating a transformation matrix from the document to the document image based on The OCR target area position acquisition unit The transformation matrix; the position of the OCR target area in the document, which is included in the OCR target area position data acquired by the data acquisition unit; and The OCR target area position acquisition system according to claim 1, further comprising:

2. The OCR target area position acquisition unit, three positions of the image code in the document image acquired by the image code position acquisition unit; three positions of the image code in the document, the three positions being included in the image code position data acquired by the data acquisition unit; Calculate the transformation matrix based on 2. The OCR target area position acquisition system according to claim 1, wherein the remaining position of the image code in the document is not present on a straight line passing through two of the three positions of the image code.

3. An image code position acquisition unit that acquires a position of the image code in a document image, which is an image of a document to which data is assigned by an image code; a data acquisition unit that acquires data represented by the image code; an OCR target area position acquisition unit that acquires a position in the document image of an OCR target area, which is an area in the document image where OCR processing is to be performed; By implementing this on a computer, The data provided to the document by the image code is image code location data including a location of the image code in the document; OCR target region location data including a location of the OCR target region in the document; and Including, The OCR target area position acquisition unit a position of the image code in the document image acquired by the image code position acquisition unit; the position of the image code in the document, which is included in the image code position data acquired by the data acquisition unit; Calculating a transformation matrix from the document to the document image based on The OCR target area position acquisition unit The transformation matrix; the position of the OCR target area in the document, which is included in the OCR target area position data acquired by the data acquisition unit; and and acquiring a position of the OCR target area in the document image based on the position of the OCR target area.

Images