JP2015201077A - Image processor, image processing method, and image processing program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an image processor capable of efficiently utilizing an OCR function while ensuring security.SOLUTION: The image processor performs a layout analysis (#3), and extracts an area on the basis of the result when receiving an input of a scan image (#1). The image processor determines the degree of a correlation between a plurality of areas (#4). In the case that the degree of the correlation between the areas is equal to or more than a prescribed degree, the image processor uses a different OCR function to each area (#5).

Description

  The present invention relates to an image processing device, an image processing method, and an image processing program, and more particularly to an image processing device, an image processing method, and an image processing program that can use an OCR (Optical Character Reader) function.

  One example of image processing is OCR (Optical Character Reader) processing on a scanned image. When the OCR process is performed, the scanned image is recognized, so that there is a risk of leakage of information on the original document depending on the reliability of the OCR function provider. Therefore, for example, when the reliability of the OCR function provider is not high, such as when using the OCR function provided by an external device, the scan image is divided and provided from different providers as a method for ensuring security. It is possible to use the OCR function.

JP 2008-293354 A JP 2006-260115 A JP 2010-102709 A JP 2007-323415 A JP 2005-007315 A

  However, if the scanned image is divided into small areas and OCR processing is performed separately, it becomes impossible to perform semantic analysis and word analysis between the divided small areas. Therefore, there is a problem that the OCR accuracy is reduced.

  In addition, since it is difficult to cut out characters one by one from the scanned image, it is conceivable to set the size of the small area to be small or to divide it into small areas by overlapping each portion. However, when doing so, there is also a problem that the amount of OCR processing increases.

  SUMMARY An advantage of some aspects of the invention is that it provides an image processing apparatus, an image processing method, and an image processing program that can efficiently use an OCR function while ensuring security. Yes.

  In order to achieve the above object, according to an aspect of the present invention, an image processing apparatus is an image processing apparatus capable of processing a scanned image, an input means for receiving an input of the scanned image, an OCR (Optical Character Reader) ) OCR processing means for acquiring text data obtained by performing OCR processing on the scanned image using the function, and control means for controlling use of the OCR function in the OCR processing means. . The control means includes a determination means for determining the degree of relevance between the first area and the second area extracted from the scan image based on the layout of the scan image, the first area and the second area When the degree of association with the area is equal to or greater than the predetermined degree, the control unit controls the OCR process to be performed on each of the first area and the second area using different OCR functions.

  Preferably, when the OCR processing means can use a plurality of OCR functions provided from different providers, the control means provides different providers for the first area and the second area, respectively. Control is performed to perform OCR processing using the OCR function provided from.

  Preferably, the control means controls the OCR process for each of the first area and the second area by using the same OCR function provided from the same provider at different timings.

  Preferably, the determination unit determines the degree of relevance based on keywords included in the first area and the second area.

  Preferably, the determination unit determines the degree of relevance based on the arrangement of the first area and the second area in the scan image.

  Preferably, when the determination unit cannot determine the relevance of the first area to the second area, the control unit further divides the first area into a plurality of areas, and uses OCR functions different from each other. Control to perform processing.

  Preferably, the image processing apparatus further includes a memory having a storage area for storing an OCR function to be used for each area included in the layout for each layout, and the control unit has a storage area for the layout of the scanned image. When the layout corresponds to the layout stored in the storage area, the OCR function stored in the storage area is used to perform OCR processing of the scanned image.

  Preferably, the control means performs OCR processing using different OCR functions for the first area and the second area when the security level set in the scanned image is within a predetermined range. Control as follows.

  More preferably, the input means includes setting means for setting a security level for the scanned image.

  Preferably, the control unit further uses an OCR used for the area based on a security level set for the area extracted from the scanned image and a security level of an OCR function available in the OCR processing unit. Determine function.

  Preferably, the scanned image is divided in units of pages, and the control unit uses an OCR function that is different for each of the first area and the second area based on the layout of each page of the scanned image. Control to perform processing.

  According to another aspect of the present invention, the image processing method is a scan image processing method, the step of receiving an input of the scan image, the first area extracted from the scan image based on the scan image layout, and the first area A step of determining the degree of relevance with the area 2 and a step of acquiring text data obtained by performing OCR processing on a scanned image using an OCR function, and acquiring text data Uses the different OCR function for each of the first area and the second area when the degree of association between the first area and the second area is a predetermined degree or more. Including performing steps.

  According to still another aspect of the present invention, the image processing program is a program that causes a computer to execute processing of a scanned image, and is extracted from the scanned image based on a step of receiving an input of the scanned image and a layout of the scanned image The step of determining the degree of relevance between the first area and the second area and the step of acquiring text data obtained by performing OCR processing on the scanned image using the OCR function The step of executing and acquiring the text data is different for each of the first area and the second area when the degree of association between the first area and the second area is equal to or greater than a predetermined degree. And a step of performing OCR processing using the OCR function.

  According to the present invention, in the image sub-processing apparatus, it is possible to efficiently use the OCR function while ensuring security.

1 is a block diagram illustrating a specific example of an apparatus configuration of an MFP (Multi-Functional Peripheral) as an example of an image processing apparatus according to an embodiment; FIG. 3 is a diagram illustrating an outline of operation in an MFP. It is a figure for demonstrating layout analysis. 2 is a block diagram illustrating a specific example of a functional configuration of an MFP. FIG. 3 is a flowchart showing an operation flow in the MFP.

  Embodiments of the present invention will be described below with reference to the drawings. In the following description, the same parts and components are denoted by the same reference numerals. Their names and functions are also the same. Therefore, these descriptions will not be repeated.

<Device configuration>
FIG. 1 is a block diagram showing a specific example of the apparatus configuration of an MFP (Multi-Functional Peripheral) 100 as an example of an image processing apparatus according to the present embodiment. Referring to FIG. 1, MFP 100 includes a CPU (Central Processing Unit) 10 for controlling the entire apparatus, a ROM (Read Only Memory) 11 that is a memory for storing a program executed by CPU 10, and a CPU 10. As an example of a RAM (Random Access Memory) 12 that is a memory for storing a calculation value or a work area when executing a program, and a large-capacity storage device for storing image data and the like An HDD (Hard Disk Drive) 13, a scanner 14, a printer 15, an operation panel 16, and a network controller 17 for communicating with other devices are included.

  MFP 100 can communicate with server 300 as an example of another device. The MFP 100 can use an OCR (Optical Character Reader) function provided by the server 300 by communicating with the server 300. The server 300 may provide a plurality of types of OCR functions. Alternatively, MFP 100 may be able to communicate with a plurality of servers 300 and use a plurality of OCR functions provided by each server 300.

<Overview of operation>
FIG. 2 is a diagram showing an outline of operation in MFP 100. Referring to FIG. 2, the user scans the document to be processed (step # 1). Preferably, the user sets a security level for the scanned image (step # 2). The security level setting may be added to the scanned image information as an attribute in the same manner as the resolution setting, or a predetermined image or code (for example, QR) indicating the security level, such as “confidential”. A code (registered trademark) or the like may be added to the scanned image itself.

  FIG. 3 is a diagram for explaining the layout analysis. As an example, a scan image obtained by scanning a document titled “Bill” is shown. As shown in FIG. 3, it is generally considered that there are many scanned images of a document created in a specific format such as an invoice or a resume as a target of OCR processing. MFP 100 can acquire character information (text) included in the document by performing OCR processing on the scanned image having such a format.

  Upon receiving the input of the scan image as shown in FIG. 3, the MFP 100 performs layout analysis on the scan image (step # 3). The layout analysis is a technique for analyzing and analyzing information included in an image from the feature amount of the image, and refers to a technique for detecting the arrangement of information in the image including characters. When the layout analysis of the scanned image of FIG. 3 is performed, for example, the MFP 100 extracts areas 501 to 517 as areas including (or possibly including) characters. The MFP 100 may extract an area including a table composed of ruled lines, a photograph, and other figures by using characters as areas other than the area. MFP 100 holds information representing the arrangement and contents of each area as layout information. MFP 100 may store layout information for each format.

  The MFP 100 further determines the relationship between the extracted areas (step # 4), and determines an OCR function to be used for the OCR processing of each area based on the determination result (step # 5).

  In step # 4, MFP 100 performs a keyword search for each area, and classifies each area according to whether or not it includes a predefined keyword. For example, regarding areas 504, 507, and 517 including keywords such as “corporation” and “limited company”, MFP 100 assumes that the text included in areas 504, 507, and 517 represents the company name, and areas 504 and 507. , 517 can be “company name”. Similarly, for an area 506 that includes predetermined keywords such as “department”, “department manager”, and “section manager”, the MFP 100 classifies the area 506 as “person in charge” assuming that the text included in the area 506 represents the person in charge. ". For the area 515 that does not include any predefined keyword, the MFP 100 classifies the area 515 as a “special character string” that does not include the keyword.

  In step # 4, the MFP 100 determines the relevance of each area based on the classification result of each area. For example, the MFP 100 can identify the area 504 classified as “company name” and the area 506 classified as “person in charge” as a relevant combination. Preferably, MFP 100 further determines relevance based on the arrangement of areas. For example, the MFP 100 can specify an area 504 classified as “company name” and an area 506 classified as “person in charge” arranged in the vicinity thereof as a relevant combination.

  The MFP 100 groups related areas (in this example, area 504 and area 506) by adding tags or the like. Then, the MFP 100 determines the OCR function so that the OCR function used for the OCR of each of the plurality of areas that are related to each other is different (step # 5), and executes the OCR process (step # 6). ). For example, in the above example, MFP 100 determines that the OCR function used for the OCR processing in each of area 504 and area 506 is a different OCR function. The different OCR function is to use the OCR function provided by the first server and the OCR function provided by the second server which are different providers for the areas 504 and 506, respectively. Alternatively, the same OCR function provided by the first server may be used for the areas 504 and 506 at different timings.

  Thus, since MFP 100 uses different OCR functions for areas that are determined to be related, for example, different OCR functions are used for company names and persons in charge. Therefore, the relevance between a plurality of information is divided. That is, in the OCR function provided by the same server at the same opportunity, relevant information such as a company name and a person in charge is not subjected to OCR processing. Therefore, it is possible to effectively prevent leakage of the contents of the document.

  Note that the MFP 100 may determine not only the presence / absence of relevance but also the degree of relevance. That is, MFP 100 can store the degree of relevance for each classification in advance, and can determine the degree of relevance between areas according to the classification of each area as a result of layout analysis. In this case, MFP 100 may determine an OCR function to be used for each area according to the degree of association between the areas. That is, MFP 100 determines to use a different OCR function when the relationship between areas is greater than or equal to a predetermined level, and determines that the same OCR function may be used when the predetermined level is not reached. Also good. Alternatively, when a security level is set for the OCR function, the MFP 100 uses a different OCR function having a high security level for each area when the relationship between the areas is equal to or greater than a predetermined level. May be determined.

  Further, when a security level is set for each area of the scanned image, MFP 100 may determine an OCR function to be used in consideration of the security level of the area. In other words, MFP 100 preferably determines to use the OCR function with a high security level when the security level set for the area is high.

  The security level for each area may be set in advance by the document creator, or may be registered in the memory in association with the format, or the MFP 100 may store the security level corresponding to the keyword in advance, and You may judge based on the keyword detected from the area.

  On the other hand, for the area 504 classified as “special character string” by the layout analysis in step # 3, the MFP 100 further divides the area 504 into smaller areas, and determines the OCR function used for the processing of each small area as a different OCR function. . Alternatively, when the security level is set for the OCR function, the MFP 100 is more secure than the OCR function used for the area 504 classified as “special character string” for the area not classified as the special character string. It is decided to use a high-level OCR function.

  Since the contents described in the area classified as “special character string” are unknown in the layout analysis, the relevance to other areas cannot be determined. By determining the OCR function to be used as described above for the area classified as “special character string” by the MFP 100, it is possible to more effectively prevent leakage of the contents of the area.

  As another example, the areas 505 and 508 classified as “address” as described above are further classified into smaller areas by the MFP 100, and the OCR function used for processing of each small area is changed to a different OCR function. It may be determined. That is, MFP 100 may further classify into a small area for a specific classification, and determine the OCR function used for the processing of each small area as a different OCR function. By doing in this way, the leakage of the content of the said area can be prevented more effectively.

  The MFP 100 receives the text information that is the result of the OCR processing from each OCR function and integrates them (step # 7), thereby obtaining the contents of the document.

  When the security level is set for the OCR function, the MFP 100 may determine the OCR function by further considering the security level set for the scanned image. That is, when a high security level is set for the scanned image, MFP 100 can determine to use the OCR function for which the highest security level is set. As the OCR function having the highest security level, an OCR function installed in the MFP 100 itself can be cited. On the other hand, when the security level set for the scanned image is low, the MFP 100 does not consider the degree of association between the areas as described above, and any one of the OCR functions for the entire scanned image. You may decide to use. By doing so, the processing can be made more efficient and faster. Therefore, the FMP 100 uses an OCR function that is used according to the degree of relevance between areas extracted as a result of layout analysis as described above only when the security level set for the scanned image is within a predetermined range. It may be determined.

  Preferably, MFP 100 stores the OCR function to be used determined for each area as OCR information in association with a document format in a memory. For example, in the example of FIG. 3, the MFP 100 stores the classification of each of the areas 501 to 517 and the OCR function to be used in a memory in association with the document format “invoice”. If it is determined that the scanned image to be processed is of the format “invoice” as a result of layout analysis in the next process, the MFP 100 reads out the OCR information associated with the format “invoice” from the memory. Applied to the scanned image to be processed. By doing in this way, the process which determines the OCR function utilized in the process after the 2nd time can be sped up. Therefore, preferably, when the scanned image is divided into pages, that is, when the document is composed of a plurality of pages, MFP 100 performs layout analysis for each page and specifies the format of the page. Then, it is confirmed whether or not the OCR information associated with the format is registered in the memory.

<Functional configuration>
FIG. 4 is a block diagram showing a specific example of a functional configuration for causing MFP 100 to perform the above-described operation. Each function of FIG. 4 is realized mainly by the CPU 10 when the CPU 10 of the MFP 100 reads the program stored in the ROM 11 onto the RAM 12 and executes it. However, at least a part of the functions may be realized by another configuration shown in FIG. 1 or another hardware configuration such as an electric circuit (not shown).

  Referring to FIG. 4, HDD 13 as an example of memory is OCR information which is a storage area for storing OCR information which is information representing an OCR function determined to be used for each area in association with a document format. A storage unit 131 is included.

  Further, referring to FIG. 4, the CPU 10 performs OCR processing on the input scan image and the input unit 101 for receiving the input of the scan image from another device via the scanner 14 or the network controller 17. An OCR processing unit 102 for acquiring obtained text data, an OCR execution unit 103 for executing OCR processing on a scanned image, and a control unit for controlling use of the OCR function in the OCR processing unit 102 104. Preferably, the input unit 101 includes a setting unit 109 for setting a security level for the scanned image.

  The control unit 104 is extracted with an analysis unit 105 for performing layout analysis of the scanned image, an extraction unit 106 for extracting an area including characters (or character strings) from the scanned image based on the analyzed layout. A determination unit 107 for determining the degree of relevance between areas (including the presence or absence of relevance), and an OCR function used for OCR processing of each area based on the determined degree of relevance between areas And a determination unit 108 for determination.

  The OCR function used for each area determined by the determination unit 108 is stored in the OCR information storage unit 131 as OCR information in association with the layout of the document. When the OCR information associated with the document format specified from the layout analyzed by the analysis unit 105 is stored in the OCR information storage unit 131, the control unit 104 stores the OCR information on the street from the OCR information storage unit 131. Is read. In this case, the determination unit 108 determines an OCR function to be used for each area based on the OCR information.

  The determination unit 107 determines the degree of relevance based on the keywords included in each area. Preferably, the determination unit 107 further determines the degree of relevance in consideration of the arrangement of each area in the scan image.

  The determination unit 108 determines to use a different OCR function for each area when the degree of association between the areas is equal to or greater than a predetermined degree. Different OCR functions may be OCR functions provided from different providers, or may use the same OCR function at different timings.

  For areas where the degree of relevance cannot be determined by the determination unit 107, the determination unit 108 preferably determines that the area is further divided into a plurality of areas and uses different OCR functions.

<Operation flow>
FIG. 5 is a flowchart showing an operation flow in MFP 100. The operation shown in the flowchart of FIG. 5 is realized by causing the CPU 10 of the MFP 100 to read out and execute a program stored in the ROM 11 on the RAM 12 and exhibit the functions shown in FIG.

  Referring to FIG. 5, when receiving an input of a scanned image (YES in step S101), CPU 10 checks whether or not a security level is set for the scanned image. When a high security level or a low security level is set for the input scanned image, or when no security level is set (NO in step S103), the CPU 10 performs different OCR for each area thereafter. It is determined that the predetermined OCR function is used without performing the process for using the function (step S125), and the OCR process is executed (step S127). That is, when the security level of the scanned image is higher than the predetermined level, the CPU 10 uses an OCR function with a high security level such as an OCR function installed in the MFP 100 for all areas of the scanned image. Then decide. When the security level of the scanned image is lower than the predetermined level, or when the security level is not set, the CPU 10 determines to use a predetermined OCR function for all areas of the scanned image. .

  If the security level set for the input scan image is within a predetermined range (YES in step S103), the CPU 10 performs a layout analysis on the scan image and specifies the original document layout of the scan image. Then, an area including characters is extracted (step S105).

  When the OCR information is stored in the memory in association with the format obtained from the specified layout (YES in step S107), the CPU 10 reads the OCR information on the street from the memory instead of the following process (step S107). In step S109, an OCR function to be used for each area is determined (step S125), and an OCR process is executed (step S127).

  When the OCR information is not registered in the memory (NO in step S107), the CPU 10 executes a process for determining an OCR function to be used for each area. That is, the CPU 10 executes OCR processing for each area extracted by the layout analysis in step S105 (step S111), and searches whether or not a predetermined keyword is included in the area (step S113). . As a result, when a keyword is included in the area (YES in step S115), the CPU 10 classifies the area according to the included keyword (step S117). If none of the keywords are included in the area (NO in step S115), the CPU 10 classifies the area as a “special character string” (step S119).

  CPU10 performs the process of the above step S111-S119 with respect to all the extracted areas. When the processing is completed for all areas (NO in step S121), the CPU 10 determines the relevance of each area based on the classification result of each area. That is, the CPU 10 identifies a group of related areas according to the classification result of each area, and groups them by adding a tag or the like (step S123). At this time, the CPU 10 may specify a group of related areas in consideration of the arrangement of each area.

  The CPU 10 determines the OCR function to be used for each area so that the OCR function to be used for each of the plurality of grouped areas is different if there is a relationship (step S125). In addition, for the area classified as “special character string” in step S119, the CPU 10 further divides the area into small areas and uses different OCR functions for each small area. The OCR function to be used for each small area is determined.

  The CPU 10 passes only the target area of the scanned image to the server 300 that provides the OCR function determined to be used for the OCR process of the area, and executes the OCR process (step S127).

  For each used OCR function, the CPU 10 accepts a text as a processing result from the server 300 that provides the OCR function. Then, by integrating them based on the arrangement of the areas (step S129), the contents of the scanned image are obtained.

  If the scanned image input in step S101 is divided into pages, the CPU 10 repeats steps S103 to S129 for each page. As a result, the CPU 10 performs layout analysis in units of pages, and when the OCR information associated with the format obtained from the layout is registered in the memory (the process of determining the OCR function for each subsequent area). It is possible to determine the OCR function using the registered OCR information and execute the process (without performing the process).

<Effect of Embodiment>
By executing the above processing in MFP 100, MFP 100 can also use the OCR function provided by the external device while ensuring security. In other words, when MFP 100 according to the present embodiment is used, OCR processing is performed using OCR functions in which areas including highly relevant contents such as address and company name, product name and unit price are different. Therefore, even if the content of the area that is the result of the OCR processing is leaked through the server 300 that provides the OCR function, since the relationship is broken, the leakage of the meaning of the content is prevented. The In other words, in the above example, even if only the product name or only the unit price is leaked to a third party, it will not be leaked to a third party with a content that constitutes a meaning as “unit price of the product”. . Therefore, OCR processing can be performed more efficiently than when all of the OCR functions are used. That is, MFP 100 according to the present embodiment can efficiently use the OCR function while ensuring security.

  Further, in MFP 100 according to the present embodiment, once an OCR function to be used for each area is determined, the result is stored as OCR information in a memory in association with a format. After the next time, when the target scan image is subjected to layout analysis and the format is specified, the OCR function that reads and uses the OCR information stored in association with the format is specified. In general, scenes used for OCR processing of documents are often in an office environment, and in many cases, documents created in accordance with the format such as “invoice” and “invoice” are processed. . Therefore, by registering in advance the OCR function to be used for each area for each format, it is not necessary to repeat the process for determining the OCR function for each process after the second time, and the process can be speeded up. .

<Other example 1>
In the above example, the MFP 100 specifies and classifies the keywords included in the area using the OCR function installed in the MFP 100. The area classification is not limited to the method using the OCR function. As another example, a specific user such as an administrator or a document creator may classify each area in advance, and a tag representing the classification may be given. Furthermore, OCR information for each format may be registered in advance by a specific user such as an administrator or a document creator. Further, the OCR information may be registered not only in the memory of MFP 100 but also in another storage device (such as server 300), and MFP 100 may access the storage device and read the OCR information. Alternatively, OCR information may be added to the scanned image.

<Other example 2>
In the above example, the image processing apparatus according to the present embodiment is an MFP, and the MFP itself is also equipped with an OCR function. However, the image processing apparatus is not limited to the MFP. As another example of the image processing apparatus, for example, a terminal device called a general-purpose computer or a so-called tablet may be used. That is, any device can be used as long as it can communicate with the server 300 that provides the OCR function and can handle image data as data to be processed.

  When the image processing apparatus itself does not have the OCR function as described above, the image processing apparatus performs the processing for the area based on the information indicating the classification set in advance for the area as in the above example. The OCR function to be used can be determined.

<Other example 3>
Furthermore, it is possible to provide a program for causing the MFP 100 or an image processing apparatus, which is another apparatus, to execute the above-described processing. By providing the program, a general-purpose MFP or computer can function as the image processing apparatus according to the present embodiment.

  Such a program is recorded on a computer-readable recording medium such as a flexible disk attached to the computer, a CD-ROM (Compact Disk-Read Only Memory), a ROM, a RAM, and a memory card, and provided as a program product. You can also Alternatively, the program can be provided by being recorded on a recording medium such as a hard disk built in the computer. A program can also be provided by downloading via a network.

  A program according to the present invention is a program module that is provided as a part of an operating system (OS) of a computer and that executes necessary processes by calling necessary modules in a predetermined arrangement at a predetermined timing. Also good. In that case, the program itself does not include the module, and the process is executed in cooperation with the OS. A program that does not include such a module can also be included in the program according to the present invention.

  The program according to the present invention may be provided by being incorporated in a part of another program. Even in this case, the program itself does not include the module included in the other program, and the process is executed in cooperation with the other program. Such a program incorporated in another program can also be included in the program according to the present invention.

  The provided program product is installed in a program storage unit such as a hard disk and executed. The program product includes the program itself and a recording medium on which the program is recorded.

  The embodiment disclosed this time should be considered as illustrative in all points and not restrictive. The scope of the present invention is defined by the terms of the claims, rather than the description above, and is intended to include any modifications within the scope and meaning equivalent to the terms of the claims.

  10 CPU, 11 ROM, 12 RAM, 13 HDD, 14 scanner, 15 printer, 16 operation panel, 17 network controller, 100 MFP, 101 input unit, 102 OCR processing unit, 103 OCR execution unit, 104 control unit, 105 analysis unit 106 extraction unit 107 determination unit 108 determination unit 109 setting unit 131 OCR information storage unit 300 server 501-517 area.

Claims (13)

An image processing apparatus capable of processing a scanned image,
Input means for receiving input of the scanned image;
OCR processing means for acquiring text data obtained by performing OCR processing on the scanned image using an OCR (Optical Character Reader) function;
Control means for controlling the use of the OCR function in the OCR processing means,
The control means includes determination means for determining the degree of relevance between the first area and the second area extracted from the scan image based on the layout of the scan image,
When the degree of relevance between the first area and the second area is equal to or greater than a predetermined degree, the control means performs different OCR for each of the first area and the second area. An image processing apparatus that controls to perform OCR processing using a function.   When the OCR processing means can use a plurality of OCR functions provided by different providers, the control means provides different provisions for each of the first area and the second area. The image processing apparatus according to claim 1, wherein control is performed so that OCR processing is performed using an OCR function provided from the beginning.   The control means controls to perform OCR processing for each of the first area and the second area by using the same OCR function provided from the same provider at different timings. The image processing apparatus according to claim 1.   The image processing apparatus according to claim 1, wherein the determination unit determines the degree of relevance based on keywords included in the first area and the second area.   The image processing apparatus according to claim 1, wherein the determination unit determines the degree of relevance based on the arrangement of the first area and the second area in the scanned image. .   When the determination means cannot determine the relevance of the first area to the second area, the control means further divides the first area into a plurality of areas and uses different OCR functions. The image processing apparatus according to claim 1, wherein the image processing apparatus controls to perform OCR processing. Each layout further includes a memory having a storage area for storing an OCR function used for each area included in the layout,
When the layout of the scanned image corresponds to the layout stored in the storage area, the control means performs OCR processing of the scanned image using the OCR function stored in the storage area. The image processing apparatus according to claim 1, which is controlled.   The control means performs OCR processing using different OCR functions for the first area and the second area when the security level set for the scanned image is within a predetermined range. The image processing apparatus according to claim 1, wherein the image processing apparatus is controlled to perform.   The image processing apparatus according to claim 8, wherein the input unit includes a setting unit for setting a security level for the scanned image.   The control means is further used for the area based on the security level set for the area extracted from the scanned image and the security level of the OCR function available in the OCR processing means. The image processing apparatus according to claim 1, wherein an OCR function is determined. The scanned image is divided into pages,
The control means controls to perform OCR processing using a different OCR function for each of the first area and the second area based on a layout of each page of the scanned image. Item 11. The image processing apparatus according to any one of Items 1 to 10. A method of processing a scanned image,
Receiving an input of the scanned image;
Determining the degree of relevance between the first area and the second area extracted from the scanned image based on the layout of the scanned image;
Using the OCR function to obtain text data obtained by performing OCR processing on the scanned image,
The step of acquiring the text data is performed for each of the first area and the second area when the degree of association between the first area and the second area is a predetermined degree or more. An image processing method including a step of performing OCR processing using different OCR functions. A program that causes a computer to process scanned images,
Receiving an input of the scanned image;
Determining the degree of relevance between the first area and the second area extracted from the scanned image based on the layout of the scanned image;
Using the OCR function to cause the computer to execute text data obtained by performing OCR processing on the scanned image;
The step of acquiring the text data is performed for each of the first area and the second area when the degree of association between the first area and the second area is a predetermined degree or more. An image processing program including a step of performing OCR processing using different OCR functions.

Images