JP6928401B1 - Information processing equipment, information processing methods and information processing programs - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an information processing device, an information processing method and an information processing program capable of improving a recognition rate. An information processing apparatus according to the present invention has a correction unit that corrects an input document image data by two or more correction patterns and generates two or more corrected image data corresponding to each correction pattern. It is provided with an OCR processing unit that reads information described in two or more corrected image data and converts the read information into text. [Selection diagram] Fig. 1

Description

The present invention relates to an information processing device, an information processing method, and an information processing program.

Conventionally, a technique called OCR (Optical Character Recognition) for recognizing characters contained in an image has been used. When the description location and size of the characters in the image are unified, the characters can be recognized with high accuracy by OCR and the character information can be extracted from the image (see, for example, Patent Document 1).

When the description location and size of the characters in the image are unified, the character information can be extracted from the image with high accuracy by the conventional OCR technology. However, when there are a plurality of items to be acquired or when there is a difference in the background color, it is difficult to properly recognize the characters by the conventional OCR technique.

Japanese Unexamined Patent Publication No. 2016-126356

The present invention has been made in view of the above problems, and an object of the present invention is to provide an information processing device, an information processing method, and an information processing program capable of improving the recognition rate.

In order to solve the above problems, the information processing apparatus of the present invention corrects the image data of the input document by two or more correction patterns, and generates two or more corrected image data corresponding to each correction pattern. It is provided with a unit and an OCR processing unit that reads information described in two or more corrected image data and converts the read information into text.

According to the present invention, it is possible to provide an information processing device, an information processing method, and an information processing program capable of improving the recognition rate.

It is a figure which shows an example of the schematic structure of the information processing system which concerns on embodiment. It is a figure which shows an example of the hardware configuration of the server which concerns on embodiment. It is a figure which shows an example of the database stored in the storage device of the server which concerns on embodiment. It is a figure which shows an example of the information stored in a database. It is a figure which shows an example of the functional structure of the server which concerns on embodiment. It is a figure which shows an example of the hardware configuration and the functional configuration of the user terminal which concerns on embodiment. It is a flowchart which shows an example of the character recognition processing of the server which concerns on embodiment. It is a figure which shows an example of character recognition by the OCR processing part of the server which concerns on embodiment.

Hereinafter, embodiments of the present invention will be described with reference to the drawings. In the present embodiment, the embodiment will be described by taking the My Number card as an example as a document. However, in the present invention, in addition to the My Number card, a document describing the My Number such as a My Number notification card, a driver's license, and a passport ( It can also be used for documents such as passports), basic resident register cards, residence cards, and resident cards. Further, in the present embodiment, the text includes characters, symbols, numerical values, and the like.

[Embodiment]
FIG. 1 is a diagram showing an example of an outline configuration of the information processing system 1 according to the embodiment. The information processing system 1 has a configuration in which a server 2 and a user terminal 3 are connected via a network 4. The number of servers 2 and user terminals 3 included in the information processing system 1 is arbitrary. Further, what kind of communication network the network 4 is configured with is arbitrary.

FIG. 2 is a diagram showing an example of a hardware configuration of the server 2 (information processing device) according to the present embodiment. As shown in FIG. 2, the server 2 includes a configuration in which a communication IF 200A, a storage device 200B, and a CPU 200C are connected via a bus.

The communication IF200A is an interface for communicating with an external terminal.

The storage device 200B is, for example, an HDD or a semiconductor storage device. The information processing program and various databases used by the server 2 are stored in the storage device 200B. In the present embodiment, the information processing program and various databases are stored in the storage device 200B of the server 2, but are stored in an external storage device such as a USB memory or an external server connected via a network, and are required. It may be configured to be referenceable or downloadable depending on the situation.

FIG. 3 is a diagram showing an example of a database stored in the storage device 200B of the server 2. As shown in FIG. 3, the storage device 200B includes a correction pattern database 1 (hereinafter, also referred to as a correction pattern DB1), an identification database 2 (hereinafter, also referred to as an identification DB2), and an item master database 3 (hereinafter, also referred to as an item master). A database 4 for notation change (hereinafter, also referred to as DB4 for notation change) is stored.

(Correction pattern DB1)
The correction pattern DB1 stores two or more correction patterns for correcting the image data of the My Number card. FIG. 4A is a diagram showing an example of information stored in the correction pattern DB1. As shown in FIG. 4A, each of the two or more correction patterns is configured by combining one or more correction filters. For example, the correction pattern 1 is configured by combining the correction filters 1 and 3. Further, the correction pattern 2 is configured by combining correction filters 1, 2 and 4. Further, the correction pattern 3 is configured by combining correction filters 1, 2 and 3. Further, the correction pattern 4 is configured by combining correction filters 1, 3 and 5. Further, the correction pattern 5 is configured by combining the correction filters 1 and 4. The number of correction filters included in each correction pattern is arbitrary.

Further, the correction filter 1 to the correction filter 5 include, for example, at least a filter that emphasizes the outline of the My Number card included in the image data and two or more filters that change the color tone of the image data. In addition, the correction filters 1 to 5 include filters such as perspective warp (trapezoidal correction), brightness correction, contrast correction, gauss correction, blur correction, and color tone correction. The combination of the correction filters of each correction pattern shown in FIG. 4A is only an example, and it is arbitrary what kind of correction filter is configured for each correction pattern. Further, the correction filter is not limited to the five correction filters 1 to 5.

(Identification DB2)
Information for identifying the My Number card is stored in the identification DB 2. FIG. 4B is a diagram showing an example of information stored in the identification DB 2. As shown in FIG. 4B, the identification DB 2 contains information for identifying the My Number card, for example, pattern matching data (image data or feature point data (for example, image data of a seal)) peculiar to the My Number card. And feature point data)) and keywords (KW) are stored. The pattern matching data and keywords shown in FIG. 4B are merely examples, and what kind of pattern matching data and keywords should be used as information for identifying the Individual Number Card is arbitrary. The identification unit 207, which will be described later, refers to the identification DB 2 and identifies that the image data is the image data of the Individual Number Card based on whether or not the image data includes pattern matching data or keywords. By using two or more pattern matching data and keywords, it can be expected to improve the accuracy of identifying that the image data is the image data of My Number Card.

(Item master DB3)
The item master DB 3 stores the item to be acquired and information on the position of the item (also referred to as regular format layout information). In the present embodiment, the item master DB 3 stores the item acquired from the My Number card and the information on the position of the item. My number card is created according to a predetermined format. For this reason, the format layout of My Number Card is fixed. The format layout information is information indicating which item of the Individual Number Card is arranged in which position (area). FIG. 4C is a diagram showing an example of information stored in the item master DB 3. As shown in FIG. 4C, the item master DB 3 stores information on items and positions acquired from the Individual Number Card. The position information is represented by XY coordinates with the upper left of the My Number card as the zero point, and the horizontal direction is the X axis (right direction is positive) and the vertical direction is the Y axis (down direction is positive) toward the My Number card. It has become. The number of pixels may be used as a numerical value representing the coordinates. In addition, which position of the My Number card is set as the zero point is arbitrary. Further, a numerical value other than the number of pixels may be used as the numerical value representing the coordinates. Furthermore, what kind of item information is acquired from My Number Card is arbitrary.

(DB4 for notation change)
The notation change DB 4 stores information for changing the notation of the information described in the image data of the My Number card acquired by the server 2 to a predetermined notation. Specifically, the notation change DB 4 stores information for logic processing and information for first and second master processing.

The information for logic processing is information that associates the information described in the image data of the Individual Number Card, which is the target for changing the notation (hereinafter, also referred to as the processing description target), with the processing rule for changing the notation of the information. be. The information for logic processing is information associated with each item for acquiring the processing description target and the processing rule for the processing description target, and by referring to the information for logic processing, the notation is unified as a predetermined value. It can be changed to the written notation.

The information for the first master processing is information associated with each item for acquiring the information described in the image data of the My Number card, which is the processing description target, and the content of the overwriting processing by the dictionary data of the information. The information for the first master processing includes various patterns of overwrite processing contents by dictionary data in the processing description target. In this way, the information for the first master processing is the information associated with each item acquired by the notation that is an error (for example, an error that is assumed in advance such as an error due to a conversion error) and the correct notation for the error. By referring to the information for the first master processing, the typographical error can be corrected to the correct notation.

Further, the information for the second master processing is information for deleting unnecessary character data included in the information described in the image data of the My Number card. Specifically, the information for the second master processing is the information described in the image data of the My Number card and the character information included in the information. As described above, the information for the second master processing is the information described in the image data of the My Number card and the information of the characters included in the information associated with each item to be acquired. By referring to the information for 2 master processing, it is possible to recognize unnecessary character data other than the characters included in the information described in the image data of the My Number card and delete the unnecessary character data.

The CPU 200C controls the server 2 and includes a ROM (Read Only Memory) and a RAM (Random Access Memory) (not shown).

As shown in FIG. 5, the server 2 includes a receiving unit 201, a transmitting unit 202, a storage device control unit 203, a correction unit 204, a removing unit 205, an OCR processing unit 206, an identification unit 207, an acquisition unit 208, and a notation changing unit 209. It has functions such as. The function shown in FIG. 5 is realized by the CPU 200C executing an information processing program stored in the ROM (not shown) of the server 2.

The receiving unit 201 receives information transmitted from the outside, for example, image data information.

The transmission unit 202 transmits information to the outside.

The storage device control unit 203 controls the storage device 200B. Specifically, the storage device control unit 203 controls the storage device 200B to write and read information.

The correction unit 204 corrects the input image data of the My Number card by two or more correction patterns stored in the correction pattern DB1 (each correction pattern includes one or more different correction filters). , Generates two or more corrected image data corresponding to each correction pattern. Specifically, the correction unit 204 generates the corrected image data 1 in which the image data is corrected by the correction pattern 1. Further, the correction unit 204 generates the corrected image data 2 in which the image data is corrected by the correction pattern 2. Further, the correction unit 204 generates the corrected image data 3 in which the image data is corrected by the correction pattern 3. Further, the correction unit 204 generates the corrected image data 4 in which the image data is corrected by the correction pattern 4. Further, the correction unit 204 generates the corrected image data 5 in which the image data is corrected by the correction pattern 5.

The removal unit 205 removes the second area other than the first area in which the items to be acquired from the My Number card are arranged from the image data based on the regular format layout information stored in the item master DB3.

The OCR processing unit 206 reads the information described in the two or more corrected image data based on the regular format layout information of the My Number card, and converts the read information into text. Specifically, the OCR processing unit 206 reads the information described in the image data after two or more corrections for each predetermined item based on the regular format layout information of the My Number card, and corresponds to each correction pattern2. Among the information of each item in which the above corrected image data is converted into text, the most abundant information is selected as the formal textual information of each item.

The identification unit 207 refers to the identification DB 2 and identifies the image data of the My Number card. Specifically, the identification unit 207 refers to the identification DB 2 and identifies that the image data is the image data of the My Number card depending on whether or not the pattern matching data or the keyword exists. As described above, the identification unit 207 may identify whether the image data is the image data of the Individual Number Card by using two or more pattern matching data or keywords. By using two or more pattern matching data and keywords, it can be expected to improve the accuracy of identifying that the image data is the image data of My Number Card.

The acquisition unit 208 refers to the item master DB3 and acquires the character corresponding to the item for each item. For example, the acquisition unit 208 corresponds to the character up to the next item (next item) or the character up to the line feed existing on the first side (right side in this embodiment) of the item converted into text by the OCR processing unit 206. Get as a character. Further, when the character (excluding the character constituting the item) does not exist within the predetermined range of the first side (right side in this embodiment) of the item, the acquisition unit 208 has a second side different from the first side of the item. The characters up to the next item (next item) or the characters up to the line feed existing in (lower side in this embodiment) are acquired as the information described in the image data of the My Number card. The range (from where to where) for the acquisition unit 208 to acquire characters may be determined based on the number of pixels. In this case, it is advisable to define the range for acquiring characters for each item.

The notation change unit 209 changes the notation of the information described in the image data of the My Number card acquired by the acquisition unit 208 to a predetermined notation. Specifically, the notation change unit 209 refers to the notation change DB 4, and based on the logic processing information stored in the notation change DB 4, the image data of the My Number card acquired by the acquisition unit 208 is used. Change the notation of the described information to the prescribed notation. Further, the notation change unit 209 refers to the notation change DB4 and corrects the erroneous notation to the correct notation based on the information for the first and second master processing stored in the notation change DB4, or is unnecessary. The character data is recognized and the unnecessary character data is deleted.

(User terminal 3)
FIG. 6 is a diagram showing an example of a hardware configuration and a functional configuration of the user terminal 3 according to the embodiment. FIG. 6A is a diagram showing an example of the hardware configuration of the user terminal 3, and FIG. 6B is a diagram showing an example of the functional configuration of the user terminal 3. The user terminal 3 is a PC (Personal Computer), a mobile terminal (for example, a tablet terminal), or the like. As shown in FIG. 6A, the user terminal 3 includes a communication IF 300A, a storage device 300B, an input device 300C, a display device 300D, a CPU 300E, and the like.

The communication IF 300A is an interface for communicating with another device (server 2 in the embodiment).

The storage device 300B is, for example, an HDD (Hard Disk Drive) or a semiconductor storage device (SSD (Solid State Drive)). The storage device 300B stores an identifier (ID) of the user terminal 3, an information processing program, and the like. The identifier may be newly assigned by the server 2 to the user terminal 3, or may use an IP (Internet Protocol) address, a MAC (Media Access Control) address, or the like.

The input device 300C is, for example, a keyboard, a touch panel, or the like, and the input device 300C can be operated to input information (for example, image data of a My Number card) necessary for using the information processing system 1.

The display device 300D is, for example, a liquid crystal monitor or an organic EL monitor. The display device 300D displays a screen (for example, a screen for inputting image data of the My Number card) necessary for using the information processing system 1.

The CPU 300E controls the user terminal 3 and includes a ROM and a RAM (not shown).

As shown in FIG. 6B, the user terminal 3 has functions such as a receiving unit 301, a transmitting unit 302, a storage device control unit 303, an operation receiving unit 304, and a display device control unit 305. The function shown in FIG. 6B is realized by the CPU 300E executing an information processing program stored in the storage device 300B.

The receiving unit 301 receives the information transmitted from the server 2.

The transmission unit 302 assigns an identifier to the information input using the input device 300C and transmits the information to the server 2. By assigning an identifier to the information transmitted from the user terminal 3, the server 2 can recognize which user terminal 3 the received information is transmitted from.

The storage device control unit 303 controls the storage device 300B. Specifically, the storage device control unit 303 controls the storage device 300B to write and read information.

The operation reception unit 304 receives an input operation on the input device 300C. For example, it accepts an input operation of image data of My Number Card.

The display device control unit 305 controls the display device 300D. Specifically, the display device control unit 305 controls the display device 300D to display a screen (for example, a screen for inputting image data of the Individual Number Card) necessary for using the information processing system 1 according to the embodiment. Display it.

(Information processing method)
FIG. 7 is a flowchart showing an example of the character recognition process of the server according to the embodiment.

(Step S101)
The correction unit 204 of the server 2 refers to the correction pattern DB1 and corrects the image data of the document. Specifically, the correction unit 204 uses two or more correction patterns stored in the correction pattern DB1 (each correction pattern includes one or more different corrections) to input the image data of the Individual Number Card. It is corrected and two or more corrected images corresponding to each correction pattern are generated.

(Step S102)
The OCR processing unit 206 of the server 2 reads the information described in the two or more corrected image data based on the regular format layout information of the My Number card, and converts the read information into text. Specifically, the OCR processing unit 206 reads the information described in the image data after two or more corrections for each predetermined item based on the regular format layout information of the My Number card, and corresponds to each correction pattern2. Among the information of each item in which the above corrected image data is converted into text, the most abundant information is selected as the formal textual information of each item.

FIG. 8 is a diagram showing an example of character recognition by the OCR processing unit 206. As shown in FIG. 8, the OCR processing unit 206 recognizes characters from two or more corrected images corresponding to each correction pattern generated by the correction unit 204. In the example shown in FIG. 8, in the correction patterns 1, 3 and 5, the recognition result is “Taro Yamada”. Further, in the correction pattern 2, the recognition result is "Taro Yamada". Further, in the correction pattern 3, the recognition result is "unrecognizable", that is, the character cannot be recognized. The OCR processing unit 206 selects the most recognized characters from the two or more corrected image data, "Taro Yamada" in the example shown in FIG. 8, and determines the recognized characters. If there is no recognized character with the largest number (for example, when the determination results of the correction patterns 1 to 5 are 2, 2, 2, 2, and 1, respectively), the process of step S102 may be performed again. Alternatively, the correction pattern may be changed to perform the process of step S102, or a notification (an error may be output) may be given as if the characters could not be read.

(Step S103)
The identification unit 207 of the server 2 refers to the identification DB 2 and identifies the image data. Specifically, the identification unit 207 refers to the identification DB 2 and determines whether or not pattern matching data or keywords exist in the characters recognized by the OCR processing unit 206. When the characters recognized by the OCR processing unit 206 include any one of the pattern matching data or the keyword, the identification unit 207 identifies that the image data is an image of the My Number card.

(Step S104)
The acquisition unit 208 of the server 2 refers to the item master DB 3 and acquires the character corresponding to the item for each item. For example, the acquisition unit 208 corresponds to the character up to the next item (next item) or the character up to the line feed existing on the first side (right side in this embodiment) of the item converted into text by the OCR processing unit 206. Get as a character. Further, when the character (excluding the character constituting the item) does not exist within the predetermined range of the first side (right side in this embodiment) of the item, the acquisition unit 208 has a second side different from the first side of the item. The characters up to the next item (next item) or the characters up to the line feed existing in (lower side in this embodiment) are acquired as the information described in the image data of the My Number card. The range (from where to where) for the acquisition unit 208 to acquire characters may be determined based on the number of pixels. In this case, it is advisable to define the range for acquiring characters for each item.

(Step S105)
The notation changing unit 209 changes the notation of the information described in the image data of the My Number card acquired by the acquisition unit 208 to a predetermined notation. Specifically, the notation change unit 209 refers to the notation change DB 4, and is an acquisition unit based on the information for logic processing and the information for the first and second master processes stored in the notation change DB 4. The notation of the information described in the image data of the My Number card acquired in 208 is changed to a predetermined notation. Further, the notation change unit 209 refers to the notation change DB4 and corrects the typographical error to the correct notation based on the information for the first and second master processing stored in the notation change DB4, or is unnecessary. The character data is recognized and the unnecessary character data is deleted.

As described above, the server 2 (information processing device) according to the present embodiment corrects the image data of the input document (My Number Card) by two or more correction patterns, and two or more corrections corresponding to each correction pattern. It includes a correction unit 204 that generates later image data, and an OCR processing unit 206 that reads information described in two or more corrected image data and converts the read information into text. In this way, the image data is corrected by two or more correction patterns, the information described in the corrected image data is read, and the read information is converted into text, so that the reading accuracy of characters can be expected to be improved.

Further, in the present embodiment, the correction pattern includes a filter that emphasizes the outline of the My Number card included in the image data. Therefore, the position of the My Number card included in the image data can be recognized, and it can be expected that the reading position of the item can be recognized more accurately.

Further, in the present embodiment, the correction pattern includes two or more filters that change the color tone of the image data. Therefore, it can be expected that the character recognition accuracy will be improved even when regions having different background colors exist.

Further, the server 2 according to the present embodiment includes a removing unit 207 that removes a second area other than the first area in which the information read by the OCR processing unit 206 is described from the image data. It is expected that the reading accuracy will be improved because the text is converted after removing the unnecessary area in this way.

Further, the OCR processing unit 206 of the server 2 according to the present embodiment reads the information described in the two or more corrected image data for each predetermined item, and the two or more corrected images corresponding to each correction pattern. Of the information of each item in which the data is converted into text, the most information is selected as the official textual information of each item. Since the data corrected by the correction pattern of 2 or more is selected as the formal textual information of each item by the majority voting method, the reading accuracy can be expected to be improved.

Further, the server 2 according to the present embodiment refers to the information in which the information described in the image data is associated with the processing rule for changing the notation of the information, and the information converted into text based on the processing rule. A notation changing unit 209 for changing the notation to a predetermined notation is provided. Since the notation of the information converted into text is changed to a predetermined notation, it can be expected that the characters are recognized more accurately.

Further, the notation changing unit 209 of the server 2 according to the present embodiment refers to the information in which the information described in the image data and the overwriting processing content by the dictionary data of the information are associated with each other, and text based on the processing content. Change the notation of the converted information to the specified notation. Since the notation of the information converted into text is changed to a predetermined notation, it can be expected that the characters are recognized more accurately.

Further, the correction pattern of the present embodiment includes one or more different correction filters. As described above, since each correction pattern includes one or more different correction filters, the image data is processed by various filters, and it can be expected that the characters are recognized more accurately.

Further, the OCR processing unit 206 of the server 2 according to the present embodiment reads the information described in the two or more corrected image data based on the regular format layout information of the My Number card, and converts the read information into text. .. In this way, since the information is read based on the regular format layout information of the Individual Number Card, it can be expected that the characters will be recognized more accurately.

[Modification 1 of the embodiment]
In the above embodiment, the identification unit 207 refers to the identification DB 2 and determines whether or not the characters recognized by the OCR processing unit 206 have pattern matching data or keywords peculiar to the My Number card, and documents. The image data of is identified as the image data of My Number Card.

However, the identification DB2 stores pattern match data (image data, feature point data (for example, seal image data, feature point data, etc.)) and keywords (KW) that should not be included in the My Number card. When the characters recognized by the OCR processing unit 206 include pattern matching data or keywords that should not be included, the image data may be identified as not the image data of the My Number card.

[Modification 2 of the embodiment]
Further, in the above embodiment, different processing may be performed depending on the type of document. For example, the correction pattern according to the type of the document may be stored in the correction pattern DB1, and the correction unit 204 may correct the image data of the document by the correction pattern according to the type of the document. Further, not only the document type but also the correction pattern for each document type and document item is stored in the correction pattern DB1, and the correction unit 204 uses the correction pattern according to the document type and the document item to obtain the image data of the document. May be corrected for each item. For example, if the document is an Individual Number Card, the background color of the name and address items is "pink", the background color of the remarks items such as transference is "blue", and the background color of the My Number item is "gray". However, in such a case, a correction pattern considering the color of the pond may be prepared, and the image data of the document may be corrected for each item by the correction pattern according to the type of the document and the item of the document.

1 Information processing system 2 Server (information processing device)
200A communication IF
200B storage device 200C CPU
201 Reception unit 202 Transmission unit 203 Storage device control unit 204 Correction unit 205 Removal unit 206 OCR processing unit 207 Identification unit 208 Acquisition unit 209 Notation change unit 3 User terminal 300A Communication IF
300B storage device 300C input device 300D display device 300E CPU
301 Reception unit 302 Transmission unit 303 Storage device control unit 304 Operation reception unit 305 Display device control unit 4 Network DB1 Correction pattern database DB2 Identification database DB3 Item master database DB4 Notation change database

Claims (4)

A correction unit that corrects the image data of the input document by two or more correction patterns composed of a combination of different correction filters, and generates two or more corrected image data corresponding to each correction pattern.
An OCR processing unit that reads the information described in the two or more corrected image data and converts the read information into text.
Referencing the information associated with the notation that is an error and the correct notation for the error, and if the textualized information contains an error, the notation change part that changes the error to the correct notation. Prepare
The notation change part
Referencing the information in which the information described in the image data of the document and the information of the characters included in the information are associated with each other, and when the information described in the image data of the document contains unnecessary characters, the said. Delete unnecessary character data,
An information processing device characterized by this. The OCR processing unit
Based on the regular format layout information of the document, the information described in the two or more corrected image data is read, and the read information is converted into text.
The information processing apparatus according to claim 1. It is provided with an acquisition unit that acquires characters corresponding to items for each item from the information converted into text by the OCR processing unit.
The acquisition unit
The character up to the next item or the character up to the line feed existing on the first side of the predetermined item converted into text by the OCR processing unit is acquired as the character corresponding to the predetermined item, and is within the predetermined range on the first side of the predetermined item. If the character does not exist in, the character up to the next item or the character up to the line feed existing on the second side different from the first side of the predetermined item is used as the information corresponding to the predetermined item described in the image data. get,
The information processing apparatus according to claim 2. A correction unit that corrects the image data of the input document by two or more correction patterns composed of a combination of different correction filters, and generates two or more corrected image data corresponding to each correction pattern.
An OCR processing unit that reads the information described in the two or more corrected image data and converts the read information into text.
It is provided with an acquisition unit that acquires characters corresponding to items for each item from the information converted into text by the OCR processing unit.
The OCR processing unit
Based on the regular format layout information of the document, the information described in the two or more corrected image data is read, and the read information is converted into text.
The acquisition unit
The character up to the next item or the character up to the line feed existing on the first side of the predetermined item converted into text by the OCR processing unit is acquired as the character corresponding to the predetermined item, and is within the predetermined range on the first side of the predetermined item. If the character does not exist in, the character up to the next item or the character up to the line feed existing on the second side different from the first side of the predetermined item is used as the information corresponding to the predetermined item described in the image data. get,
An information processing device characterized by this.

Images