JP2021012741A - Image processor, method for processing image, and program - Google Patents

Abstract

To allow a support for a worker so that the worker will be able to get to know which item is shown in which place even when the format of a document as a character recognition target is not known in advance.SOLUTION: An image processor includes: a character string recognition unit for recognizing a character string in a specific item in an image of a recognition target on the basis of a result of learning using a plurality of images including character strings; and an output unit for outputting the image of the recognition target and the character string recognition result so that the correspondence between the image of the recognition target and the character string recognition result showing the same specific item obtained by the character string recognition unit can be grasped.SELECTED DRAWING: Figure 3

Description

The present invention relates to an image processing apparatus, an image processing method and a program.

In relation to the form format, Patent Document 1 describes a method for efficiently selecting the optimum form format for the form. In this method, the form formats are divided into groups, and one representative form format is defined for each group. Then, in this method, any one group is selected based on the feature matching rate between the read form image and the representative form format. Further, in this method, among the form formats in the selected group, the form format having the highest feature matching rate with the read form image is selected.

Further, in relation to the correction of the character recognition result, in Patent Document 2, when an unreadable character (rejected character) occurs in the character recognition when a plurality of types of forms are mixed, the unreadable character is corrected. It is described that the correction processing of the same type of form is continuously performed by designating the type of the form to be processed and performing the correction processing of unreadable characters in the specified type of form.

Further, in relation to the correction of the character recognition result, in Patent Document 3, when the image input (character recognition) form is corrected / confirmed, the input content is displayed according to a predetermined format and the correction is accepted. It also states that when the user instructs the correction item, the specific image part corresponding to the received correction item is highlighted.

Japanese Unexamined Patent Publication No. 2016-408444 Japanese Unexamined Patent Publication No. 2004-118380 JP-A-2002-007951

When confirming and correcting the character recognition result, it is important for the confirmation and correction worker to be able to understand which item is shown where. In particular, when a worker confirms and corrects the reading result of a document in a format that he / she is not familiar with, it may take time to grasp where the item to be confirmed is described.
Therefore, it is preferable to be able to assist the worker in grasping which item is shown where. At that time, even if the format of the document to be recognized is not known in advance, it is preferable to be able to assist.

An object of the present invention is to provide an image processing apparatus, an image processing method, and a program capable of solving the above-mentioned problems.

According to the first aspect of the present invention, the image processing apparatus recognizes the character string of a specific item in the image to be recognized based on the result of learning using a plurality of images including the character string. The image of the recognition target and the character string recognition result are displayed in such a manner that the correspondence relationship indicating the same specific item in the unit, the image of the recognition target, and the character string recognition result by the character string recognition unit can be grasped. It includes an output unit for output.

According to the second aspect of the present invention, the image processing method includes a step of recognizing the character string of a specific item in the image to be recognized based on the result of learning using a plurality of images including a character string. It includes a step of outputting the image of the recognition target and the character string recognition result in a manner in which the correspondence relationship indicating the same specific item in the image of the recognition target and the character string recognition result can be grasped.

According to the third aspect of the present invention, the program recognizes the character string of a specific item in the image to be recognized based on the result of learning using a plurality of images including the character string on the computer. , The step of outputting the image of the recognition target and the character string recognition result is executed in a manner in which the correspondence relationship indicating the same specific item in the recognition target image and the character string recognition result can be grasped. It is a program for.

According to the present invention, even if the format of the document to be recognized for characters is not known in advance, it is possible to assist the operator in grasping which item is shown where.

It is a figure which shows the apparatus configuration example of the image processing system including the image processing apparatus which concerns on embodiment. It is a figure which shows the hardware configuration example of the image processing apparatus which concerns on embodiment. It is a schematic block diagram which shows the functional structure of the image processing apparatus which concerns on 1st Embodiment. It is a figure which shows an example of a document form. It is a figure which shows the outline of the record table stored in the database which concerns on 1st Embodiment. It is the first figure which shows the processing flow of the image processing apparatus which concerns on 1st Embodiment. It is a 2nd figure which shows the processing flow of the image processing apparatus which concerns on 1st Embodiment. FIG. 5 is a diagram showing an example of a display screen in which the display unit according to the first embodiment displays a display of a recorded character string and an image of a document form side by side. FIG. 5 is a diagram showing an example of a display screen in which the display unit according to the first embodiment displays the correspondence between the display of the recorded character string and the image of the document form. It is a schematic block diagram which shows the functional structure of the image processing apparatus which concerns on 2nd Embodiment. It is the first figure which shows the processing flow of the image processing apparatus which concerns on 2nd Embodiment. It is the 2nd figure which shows the processing flow of the image processing apparatus which concerns on 2nd Embodiment. It is a figure which shows the example of the structure of the image processing apparatus which concerns on embodiment.

Hereinafter, embodiments of the present invention will be described, but the following embodiments do not limit the inventions claimed. Also, not all combinations of features described in the embodiments are essential to the means of solving the invention.

FIG. 1 is a diagram showing an example of an apparatus configuration of an image processing system including the image processing apparatus according to the embodiment.
With the configuration shown in FIG. 1, the image processing system 100 is composed of an image processing device 1, an image reading device 2, a recording device 3, and a database 4.
The image processing device 1 is connected to the image reading device 2 by a communication cable. The image reading device 2 optically acquires image data such as a document form and outputs it to the image processing device 1. The image processing device 1 performs OCR processing on the image data of the document form and recognizes the characters. The image processing device 1 outputs the character recognition result to the recording device 3, and the recording device 3 records the character recognition result in the database. The document targeted by the image processing device 1 is not limited to a specific type. Various documents capable of OCR processing can be processed by the image processing apparatus 1.

The database 4 is connected to the image processing device 1 and the recording device 3. The database 4 stores the correspondence between the image data of a plurality of document forms registered in the past from the recording device 3 and the recorded character string indicating the character string to be recorded among the character strings included in the image data. The character string indicated by the recorded character string is an important character string to be recorded and saved in the database 4 among the character strings described in the document form. The worker who uses the image processing system 100 registers in advance the image data of a plurality of document forms registered in the past using the recording device 3 and the recorded character string among the character strings included in the image data in the database 4. I will do it.
The worker is also referred to as a user of the image processing device 1 or simply a user.

Then, in the database 4, the correspondence between the image data of the document form and the information of the recorded character string indicating the character string to be recorded among the information of the character string included in the image data is sufficiently recorded for many document forms. It is assumed that it has been done. In such a state, the image processing device 1 performs processing.

FIG. 2 is a diagram showing a hardware configuration example of the image processing device.
In the configuration shown in FIG. 2, the image processing device 1 includes an image processing device main body 10, a display device 17, and an input device 18.
The image processing device main body 10 includes a CPU (Central Processing Unit) 11, an IF (Interface) 12, a communication module 13, a ROM (Read Only Memory) 14, a RAM (Random Access Memory) 15, an HDD (Hard Disk Drive) 16, and the like. It is a computer with a configuration. The communication module 13 may perform wireless communication with the image reading device 2, the recording device 3, and the database 4, or may perform wired communication, and may have these two functions. Good.
The display device 17 includes a display screen such as a liquid crystal panel or an LED (Light Emitting Diode) panel.
The input device 18 is a device that receives user operations, such as a keyboard and mouse, a touch sensor provided on the display screen of the display device 17 to form a touch panel, or a combination thereof.

<First Embodiment>
FIG. 3 is a schematic block diagram showing a functional configuration of the image processing device 1 according to the first embodiment.
The communication unit 110 is configured by using the communication module of FIG. 2 and communicates with other devices. In particular, the communication unit 110 communicates with the image reading device 2, the recording device 3, and the database 4.
The display unit 120 is configured by using the display device 17 of FIG. 2 and displays various images. In particular, the display unit 120 corresponds to the example of the output unit, and the correspondence information indicating the same specific item in the first document image and the second document image displayed corresponding to the first document image. Is output.

For example, the display unit 120 displays a document form image (a raw image instead of an OCR result) as a first document image, and a recorded character string which is a processing result of the image processing device 1 as a second document image. Display a GUI screen image to be displayed in a predetermined format. Then, the display unit 120 indicates which character string in the first document image the character string shown in the second document image is the reading result of, by drawing a line between both character strings.

Alternatively, when the correct character string of the recorded character string is obtained, the display unit 120 displays the OCR result of the document form as the first document image in the document form format, and displays it as the second document image. A GUI screen image that displays the correct character string in a predetermined format may be displayed. Then, the display unit 120 may indicate which character string in the first document image the character string shown in the second document image indicates the correct answer by drawing a line between the two character strings. Good.
However, the method of outputting the image and the related information by the output unit is not limited to the method of displaying the image and the related information. For example, the communication unit 110 may function as an output unit to transmit an image and related information to another device for display.

The operation input unit 130 is configured by using the input device shown in FIG. 2 and accepts user operations. In particular, the operation input unit 130 accepts a correction operation for the character string displayed by the display unit 120.
The storage unit 180 is configured by using the ROM 14, RAM 15 and HDD 16 of FIG. 2 and stores various data.
The control unit 190 is configured by the CPU 11 of FIG. 2 reading a program from the storage unit 180 (ROM 14, RAM 15 and HDD 16 of FIG. 2) and executing the program. The control unit 190 controls each unit of the image processing device 1 to execute various processes.
The acquisition unit 191 acquires the image data of the document form.

The feature amount extraction unit 192 extracts the feature amount of the character string from the recognition processing result of the image data of the document form. For example, the feature amount extraction unit 192 sets a first feature amount indicating the characteristics of the recorded character string included in the image data of the document form for each image data of the document form based on the recognition processing result of the image data of a plurality of document forms. Extract. Specifically, the feature amount extraction unit 192 specifies a recorded character string included in the character recognition-processed document image, and extracts the first feature amount in the document image of the specified recorded character string. The specification of the recorded character string referred to here is to determine any one of the character strings in the document form as one recorded character string. Extraction of features is also referred to as feature generation.

The recording unit 193 uses the feature amount of the character string in the image data of the new document form to extract and record the recorded character string from the information of the character string read from the image data of the new document form. In particular, the recording unit 193 corresponds to the example of the character string detection unit, and is pre-recorded based on the result of learning using a plurality of document images, and the character string of the item is recorded for each type of document image and for each specific item. Detects the character string of a specific item in the first document image based on the feature amount of the displayed first document image (document form image) among the feature amounts indicating the features (first feature amount). To do.

Further, the recording unit 193 detects a character string of the same item as the specific item in the first document image from the second document image. When the recording unit 193 is known about the format of the second document image, the recording unit 193 detects the recorded character string in the second document image by using the format information. On the other hand, when the recording unit 193 does not know the format of the second document image, the feature amount extraction unit 192 extracts the first feature amount also for the second document image, and the recording unit 193 obtains the first feature. The recorded character string may be specified by using the quantity.

Further, the combination of the feature amount extraction unit 192 and the recording unit 193 corresponds to the example of the correspondence learning unit, and machine learning the correspondence between the first document image and the second document image of the same specific item. ..
For example, when the recording unit 193 is known about the format of the second document image, the feature amount extraction unit 192 extracts and records the feature amount of the character string in the first document image (image of the document form) by machine learning. Part 193 identifies the recorded character string in the first document image based on the feature amount of the obtained character string. As a result, the recording unit 193 is known about the recorded character strings for both the first document image and the second document image, and has acquired the correspondence between the recorded character strings between these two document images. ..

Alternatively, when the recording unit 193 does not know the format of the second document image, the feature amount extraction unit 192 extracts the feature amount of the character string for the second document image in addition to the first document image. You may. In this case, the recording unit 193 specifies the recorded character string for each of the first document image and the second document image based on the feature amount of the obtained character string. As a result, the recording unit 193 is known about the recorded character strings for both the first document image and the second document image, and has acquired the correspondence between the recorded character strings between these two document images. ..

The correspondence learning unit (combination of the feature amount extraction unit 192 and the recording unit 193) uses the images after the character string of the second document image is determined, and uses the first document image and the second document. Machine learning may be performed on the correspondence between the image and the same specific item. When the character string of the second document image is confirmed, the confirmed character string is considered to be accurate. By performing machine learning using this accurate character string, the correspondence learning unit can learn the correspondence between the first document image and the second document image of the same specific item with relatively high accuracy. Is expected.

By such processing, the image processing apparatus 1 reduces the labor of recording the character string information to be recorded included in the image data of the new document form.

FIG. 4 is a diagram showing an example of a document form.
As shown in this figure, in the document form, the mark of the company that created the document, the date of creation, the person in charge of creation, and the content of the document are described in a format peculiar to the document form. For the document content, for example, if the document form is an order form, one or a plurality of sets of information such as the ordered product name and the ordered quantity are shown. Based on a certain document form, the worker records a specific character string (recording character string) to be recorded among the character strings described in the document form in the database 4 by using the recording device 3. Specifically, the worker inputs a recording character string to be recorded in the database 4 by the recording device 3 while looking at the document form. Further, the worker causes the image reading device 2 to read the image data of the document form. The document form is read by the image reading device 2 and output to the image processing device 1 based on the operation of the operator. Then, the recording device 3 associates the image data of one document form with the recorded character string among the character strings described in the document form based on the operation of the operator and the control of the image processing device 1. Record in database 4. In the example of FIG. 4, the date 51, the supplier 52, the product name 53, the quantity 54, and the amount 55 are the record character strings. Other information such as non-recorded character strings that are not recorded by the operator is also printed on the document form 5. The information includes, for example, the name 501 of the orderer who issued the document form, the emblem image 502 of the orderer, the title 503 of the document form, and the greeting text 504.

FIG. 5 is a diagram showing an outline of a recording table stored in the database.
As shown in FIG. 5, the database 4 stores the image data of the document form and the recorded character string among the character strings described in the document form in association with each other in the recording table.

FIG. 6 is a first diagram showing a processing flow of the image processing apparatus according to the first embodiment. FIG. 6 shows an example of a processing procedure in which the image processing apparatus 1 extracts the first feature amount.
Next, the processing flow of the image processing apparatus 1 will be described step by step.
First, in the database 4, a combination of image data of a certain document form and a recording character string described in the document form is recorded for a plurality of document forms of the same format. For example, it is assumed that a plurality of recorded character string information (information indicating the recorded character string) in the format of the document form 5 shown in FIG. 4 is recorded.
As a combination of these image data and the recorded character string information, for example, the image data of the document form handled in the past business and the recorded character string information can be used. When the required amount of image data and recorded character string information can be secured from past work, it is not necessary to separately prepare the image data and recorded character string information in order for the image processing apparatus to acquire the first feature amount.
In such a state, the operator activates the image processing device 1 and instructs the image processing device 1 to start processing.

The acquisition unit 191 of the image processing device 1 controls the communication unit 110 to read the image data of the document form and the information of the recorded character string corresponding to the image data from the database 4 (step S601). The acquisition unit 191 outputs the image data and the recorded character string to the feature amount extraction unit 192. The feature amount extraction unit 192 performs OCR processing on the image data to detect all the character strings in the image data and the coordinates in the image data indicating the range of the character strings (step S602). A character string is a group of characters composed of a plurality of characters. The feature amount extraction unit 192 analyzes the range of one group according to the distance from other characters, extracts one or more characters included in the range as a character string, and extracts the character string in the image data. Detects coordinates that indicate the range of. The characters included as a character string may include symbols such as ideographic characters and phonetic characters, marks, icon images, and the like.

The feature amount extraction unit 192 compares the character string extracted from the image data by the OCR process with the recorded character string read from the database 4 together with the image data. Among the character strings extracted from the image data by the OCR process, the feature amount extraction unit 192 includes the character strings in the image data that match the character information of the recorded character strings, the attributes of the characters included in the character strings, and the range thereof. The coordinates are specified (step S603).

The character attributes (character attributes) referred to here are information represented by numbers, alphabets, hiragana, kanji, number of characters, character height, fonts, and the like. The coordinates of the character string range are the coordinates indicating the position of the character string in the document form. For example, the coordinates of the range of the character string may be information indicating the coordinates of the first character and the coordinates of the end character included in the character string. Hereinafter, the attribute of the character included in the character string and the coordinates of the range of the character string are collectively referred to as a character string attribute or a character string attribute.

Further, the character information here may be only a character string or may include a character string attribute. That is, the feature amount extraction unit 192 may determine whether or not the recorded character string and the character string in the image data are the same as the character string. Alternatively, the feature amount extraction unit 192 may determine the identity of the character string attributes in addition to the identity of the characters.

If the feature amount extraction unit 192 cannot uniquely identify a character string whose character information matches the recorded character string, the image processing device 1 excludes the document image from the processing target (the extraction target of the first feature amount). You may do so. Alternatively, the image processing device 1 may display an image showing the range of each candidate of the recording character string in a frame on the display unit 120, and specify the recording character string in the character string selected by the operator. .. The candidate for the recorded character string here is a character string associated with the recorded character string determined to be not uniquely specified among the character strings whose character information matches the character information of the recorded character string. .. Further, the specification of the recorded character string referred to here is to determine any one of the character strings in the document form as one recorded character string.
When the feature amount extraction unit 192 determines that the character information of each of the plurality of character strings in the document form matches the character information of one recorded character string, these plurality of character strings are candidates for the recorded information. Become. The recorded character string is uniquely specified by the operator selecting any one of the plurality of character strings.

Next, the feature amount extraction unit 192 extracts the feature amount common to the document forms of the same format and for each recorded character string by using the character string attributes extracted for each document form and each recorded character string (step S604). ..
Specifically, the feature amount extraction unit 192 analyzes the character string attribute of the recorded character string in a plurality of document forms for each recorded character string, and extracts one feature amount for one recorded character string.
The method by which the feature amount extraction unit 192 extracts the feature amount common to the document forms of the same format and for each recorded character string is not limited to a specific method. For example, the feature amount extraction unit 192 describes the coordinates of the first character, the coordinates of the last character, the character type, the character height, the font type, etc. for a plurality of character string attributes obtained from a plurality of document forms. The mode may be calculated for each item. In addition, the feature amount extraction unit 192 determines the average value (Average) or median value (Median) for each item of the attributes indicated by numerical values such as the coordinates of the first character, the coordinates of the last character, the height of the characters, and the distance between the characters. ) May be obtained. Further, the feature amount extraction unit 192 uses a feature amount having a range, such as setting the maximum value and the minimum value as the feature amount, or a feature amount represented by a plurality of numerical values, for the item represented by the numerical value. It may be. Further, the feature amount extraction unit 192 may obtain the feature amount by digitizing attributes other than numerical values such as a character type and a font type. Further, the feature amount extraction unit 192 may extract the feature amount by using a known machine learning algorithm.
When the feature amount extraction unit 192 acquires a plurality of numerical values for one format of a document form and one recorded character string, the plurality of numerical values may be vectorized to extract the feature amount of one vector. ..

The feature amount extracted by the feature amount extraction unit 192, which is common to the document forms of the same format and for each recorded character string, is referred to as a first feature amount. The feature amount extraction unit 192 extracts the first feature amount of each recorded character string in the format by using a plurality of document forms of the same format. The first feature amount is a feature amount for extracting a recorded character string. The first feature quantity may include information indicating a character attribute, any of coordinates indicating a range of a character string, or a combination thereof.
The feature amount extraction unit 192 records the first feature amount obtained for each recorded character string in the database 4 in association with the identifier in the document form format (step S605).

For example, the feature quantity extraction unit 192 has a character attribute and a range of character strings for each of the date 51, the supplier 52, the product name 53, the quantity 54, and the amount 55, which are the record character strings included in the format of the document form 5 of FIG. Each first feature amount indicating the coordinates and the like indicating the above is recorded in the database 4 in association with the format identifier of the document form 5.
After step S605, the image processing apparatus 1 ends the process of FIG.

By the above processing, the image processing apparatus 1 can extract the information (first feature amount) used for reducing the labor of recording the recorded character string of the worker and store it in the database 4. As a result, the image processing device 1 can receive the input of the image data of the new document form and automatically record the recording character string included in the document form in the database 4. The process will be described with reference to FIG. 7.

FIG. 7 is a second diagram showing a processing flow of the image processing apparatus according to the first embodiment. FIG. 7 shows an example of a processing procedure in which the image processing device 1 extracts a recorded character string from the newly input image data.
The operator performs an operation of causing the image reading device 2 to read a new document form. As a result, the image reading device 2 generates image data of the document form and outputs (transmits) it to the image processing device 1. The acquisition unit 191 of the image processing device 1 acquires image data from the received data of the communication unit 110 (step S701). The acquisition unit 191 outputs the image data to the feature amount extraction unit 192. The feature amount extraction unit 192 performs OCR processing on the image data to obtain the character string, the feature (character attribute) of the character included in the character string, and the coordinates in the image data of the range of the character string for each character string. Detect (step S702). The feature amount extraction unit 192 extracts a third feature amount obtained by converting the detected information into a feature amount for each character string in the image data (step S703). That is, the third feature amount is information indicating the feature of the character string included in the document form of the newly read image data. After that, the feature amount extraction unit 192 reads out the first feature amount for each recorded character string from the database 4 (step S704). The feature amount extraction unit 192 outputs the third feature amount and the first feature amount to the recording unit 193.

The recording unit 193 acquires the third feature amount for each character string in the image data and the first feature amount for each recorded character string, and associates the first feature amount with the third feature amount (step S705). Specifically, the recording unit 193 associates each of the first feature amounts with one third feature amount that matches the feature amount or one closest third feature amount. By this association, the recorded character string is selected from the character strings obtained by OCR processing the image data of the document form.
Next, the feature amount extraction unit 192 controls the display unit 120 to display the recorded character string, which is the processing result of the image processing device 1, and the image of the document form side by side (step S706).

FIG. 8 is a diagram showing an example of a display screen in which the display unit 120 displays the display of the recorded character string and the image of the document form side by side. In the example of FIG. 8, the display unit 120 displays the image W1 of the document form and the display window W2 of the recorded character string. The character strings of the OCR results of the character strings C1a and C1b in the image W1 of the document form are displayed as the recorded character strings C2a and C2b in the recorded character string display window W2, respectively. The "OK" button icon B1 is a button icon that the operator presses to confirm and correct the recorded character string displayed in the recorded character string display window W2. The button icon is pressed by, for example, a touch operation on the button icon or a mouse click.

Next, the recording unit 193 determines whether the correspondence between the display of the recorded character string and the image of the document form has been acquired (step S707). For example, the recording unit 193 determines that the correspondence relationship has been acquired when the number of times of learning for grasping the correspondence relationship between the display of the recorded character string and the image of the document form is equal to or more than a predetermined number of times. As described above, the learning for grasping the correspondence between the display of the recorded character string and the image of the document form may be the learning for specifying the recorded character string in the display of the recorded character string, or the document. It may be learning for specifying the recorded character string in the image of a form. Alternatively, it may be a combination of these learnings.

When it is determined that the correspondence between the display of the recorded character string and the image of the document form has been acquired (step S707: YES), the recording unit 193 controls the display unit 120 to display the recorded character string and display the document form. The correspondence with the image is displayed (step S708).
FIG. 9 is a diagram showing an example of a display screen in which the display unit 120 displays the correspondence between the display of the recorded character string and the image of the document form.
FIG. 9 shows an example in which the correspondence between the display of the recorded character string and the image of the document form is displayed on the display screen of FIG. The character strings C1a, C1b, C2a, and C2b are shown surrounded by rectangles F1a, F1b, F2a, and F2b, respectively. Then, it is shown by the line La that the character strings C1a and C2a correspond to each other. The line Lb indicates that the character strings C1b and C2b correspond to each other.
In this way, the display unit 120 displays the correspondence between the display of the recorded character string and the image of the document form, so that the worker can use either the format of the document form or the format of the display window of the recorded character string. Even if you are unfamiliar with both, you can easily and surely grasp the correspondence between character strings.

When the operator confirms and corrects the recorded character string and then performs a confirming operation, the image processing device 1 acquires the corrected recorded character string (step S709). If the recording unit 193 determines in step S707 that the correspondence between the display of the recorded character string and the image of the document form has not been acquired (step S707: NO), the process proceeds to step S709.
Then, the feature amount extraction unit 192 and the recording unit 193 as the correspondence learning unit learn the coordinates of the recorded character string in the display of the processing result (in the example of FIGS. 8 and 9, the recording character string display window W2). Step S710). For example, the feature amount extraction unit 192 extracts the first feature amount for the display of the processing result as in the case of the document form, and the recording unit 193 uses the first feature amount to determine the position of the recorded character string. Identify and remember.
If the recording unit 193 knows the format for displaying the processing result, the processing in step S710 is unnecessary.

The recording unit 193 records the recorded character string in the recording table in association with the identification information of the document form (step S711).
For example, it is assumed that the third feature amount a3, the third feature amount b3, the third feature amount c3, and the third feature amount d3 can be obtained from the image data of the document form. Then, the third feature amount a3 is recorded in the database in advance, the first feature amount a1, the third feature amount b3 is the first feature amount b1, the third feature amount c3 is the first feature amount c1, and the third feature amount. It is assumed that the feature amount d3 coincides with the first feature amount d1. In this case, the recording unit 193 records the character strings corresponding to the third feature amount a3, the third feature amount b3, the third feature amount c3, and the third feature amount d3 in the recording table of the document form as recording character strings. To do. The character string corresponding to the third feature amount referred to here is the character string from which the third feature amount is extracted. When the recorded character string is modified by the operator, the recording unit 193 records the modified recorded character string in the recording table of the document form.
After step S711, the image processing device 1 ends the process of FIG. 7.

The image processing device 1 may update the first feature amount in the document form by the processing of FIG. 7. For example, the image processing apparatus 1 may learn the first feature amount in the document form in addition to or instead of the machine learning of the first feature amount in the display of the processing result in step S710. In that case, the image processing apparatus 1 may re-execute the processing flow of FIG. Alternatively, the image processing device 1 may perform additional learning in step S710 to update the first feature amount of the data processed in FIG. 6 without performing processing again.
By updating the first feature amount in the process of FIG. 7, the image processing device 1 increases the number of sample data and improves the accuracy of the first feature amount, and the accuracy of the image processing device 1 extracting the recorded character string. Is expected to improve. Further, when the recorded character string is added in the process of FIG. 7, the image processing device 1 can also extract the newly added recorded character string from the image data, which saves the worker the trouble of inputting the character string. It is expected to be omitted.

According to the process shown in FIG. 7, the image processing device 1 uses the image data of the document form and the recorded character string recorded in advance by the operator to obtain the recorded character string in the image data of the newly input document form. It can be recorded automatically. Therefore, the image processing device 1 can reduce the labor of the operator for recording the recorded character string in the document form. Further, the operator can confirm and correct the recorded character string which is the processing result of the image processing device 1 by referring to the display of the display unit 120. At that time, the display unit 120 shows the correspondence relationship between the character string in the display of the processing result and the character string in the image of the document form, so that the worker can easily grasp the correspondence relationship.

<Second embodiment>
In the second embodiment, the case where the image processing device 1 corresponds to a plurality of formats of the document form will be described.
FIG. 10 is a schematic block diagram showing a functional configuration of the image processing apparatus according to the second embodiment.
As shown in FIG. 10, the image processing apparatus 1 according to the second embodiment further has the functions of the group classification unit 194 and the group identification unit 195 in addition to the functional units shown in FIG.

FIG. 11 is a first diagram showing a processing flow of the image processing apparatus according to the second embodiment.
Next, the processing flow of the image processing apparatus 1 according to the second embodiment will be described step by step.
In the database 4, a large number of combinations of image data of a plurality of document forms having different formats and record character strings described in each document form are recorded for each document form. In such a state, the operator activates the image processing device 1 and instructs the image processing device 1 to start processing.

The acquisition unit 191 of the image processing device 1 determines whether or not all the image data of the document form and the information of the recorded character string corresponding to the image data have been read from the database 4 (step S901). If NO, the acquisition unit 191 reads the image data of the document form and the information of the recorded character string corresponding to the image data from the database 4 (step S902). The acquisition unit 191 outputs the image data and the recorded character string to the feature amount extraction unit 192. The feature amount extraction unit 192 performs OCR processing on the image data to detect all the character strings in the image data and the coordinates in the image data (step S903). A character string is a group of characters composed of a plurality of characters. The feature amount extraction unit 192 analyzes the range of one group according to the distance from other characters, extracts one or more characters included in the range as a character string, and extracts the character string in the image data. Detects coordinates that indicate the range of. The characters included as a character string may include symbols such as ideographic characters and phonetic characters, marks, icon images, and the like.

The feature amount extraction unit 192 compares the character string extracted from the image data by the OCR process with the recorded character string read from the database 4 together with the image data. Among the character strings extracted from the image data by the OCR process, the feature amount extraction unit 192 includes the character strings in the image data that match the character information of the recorded character strings, the attributes of the characters included in the character strings, and the range thereof. The coordinates are specified (step S904).

As described in the first embodiment, the character attributes are information represented by numbers, alphabets, hiragana, kanji, number of characters, character height, fonts, and the like. The coordinates of the character string range are the coordinates indicating the position of the character string in the document form. For example, the coordinates of the range of the character string may be information indicating the coordinates of the first character and the coordinates of the end character included in the character string. The attributes of the characters included in the character string and the coordinates of the range of the character string are collectively referred to as the character string attribute or the character string attribute.

As in the case of the first embodiment, the character information here may be only a character string or may include a character string attribute. That is, the feature amount extraction unit 192 may determine whether or not the recorded character string and the character string in the image data are the same as the character string. Alternatively, the feature amount extraction unit 192 may determine the identity of the character string attributes in addition to the identity of the characters.

If the feature amount extraction unit 192 cannot uniquely identify a character string whose character information matches the recorded character string, the image processing device 1 excludes the document image from the processing target (the extraction target of the first feature amount). You may do so. Alternatively, the image processing device 1 may display an image showing the range of each candidate of the recording character string in a frame on the display unit 120, and specify the recording character string in the character string selected by the operator. .. As described in the first embodiment, the record character string candidate referred to here is a record character string determined that the character information is not uniquely specified among the character strings whose character information matches the character information of the record character string. It is a character string associated with. Further, the specification of the recorded character string referred to here is to determine any one of the character strings in the document form as one recorded character string.
When the feature amount extraction unit 192 determines that the character information of each of the plurality of character strings in the document form matches the character information of one recorded character string, these plurality of character strings are candidates for the recorded information. Become. The recorded character string is uniquely specified by the operator selecting any one of the plurality of character strings.

Next, the feature amount extraction unit 192 extracts the feature amount for each document form and each recorded character string by using the character string attribute extracted for each document form and each recorded character string (step S905). Specifically, the feature amount extraction unit 192 characterizes the character string attribute of the character string associated with the recorded character string in step S904. Since a plurality of types of formats are targeted in the second embodiment, at the time of step S905 in which the document forms are not grouped by format, the first feature amount is directly extracted unlike the case of step S604 of FIG. Can't. Therefore, the feature amount extraction unit 192 extracts the feature amount for each document form and each recorded character string in preparation for extracting the first feature amount for each group. The feature amount for each document form and each recorded character string is referred to as an individual first feature amount.
The feature amount extraction unit 192 records the obtained individual first feature amount in the database 4 in association with the identifier of the document form and the identifier of the recorded character string (step S906). As the identifier of the recorded character string, for example, a coordinate value indicating the position of the recorded character string can be used.

For example, the feature amount extraction unit 192 has a range of character attributes and character strings for each of the date 51, the supplier 52, the product name 53, the quantity 54, and the amount 55, which are the record character strings included in the format of the document form 5 of FIG. The individual first feature amount indicating the coordinates and the like indicating the above is recorded in the database 4 in association with the identifier of the document form 5 and the identifier of the recorded character string for each document form and each recorded character string.

The feature amount extraction unit 192 also extracts a non-recorded character string in the image data that does not match the character information included in the recorded character string, and a character string attribute of the non-recorded character string (step S907). As described above, the non-recorded character string is a character string that is not recorded by the operator, that is, a character string other than the recorded character string. The character string attribute may include information indicating the attribute of the character included in the character string, information indicating the coordinates of the range of the character string, or both.

The feature amount extraction unit 192 extracts the feature amount for each document form and each non-recorded character string by using the character string attribute extracted for each document form and each non-recorded character string (step S908).
Specifically, the feature amount extraction unit 192 characterizes the attributes (character string attributes) of the character strings for each of the character strings that are not associated with any of the recorded character strings in step S904. As in the case of the first feature amount, at the time of step S908 in which the document forms are not grouped by format, it is not possible to generate a common feature amount for the document forms having the same format. Therefore, the feature amount extraction unit 192 extracts the feature amount for each document form and each non-recorded character string in preparation for extracting the second feature amount for each group. The feature amount for each document form and each non-recorded character string is referred to as an individual second feature amount.
The feature amount extraction unit 192 may generate an individual second feature amount for each document form and a collection of a plurality of non-recorded character strings. For example, the feature amount extraction unit 192 may generate one individual second feature amount for one document form.

The feature amount extraction unit 192 records the obtained individual second feature amount in the database 4 in association with the identifier of the document form and the identifier of the non-recorded character string (step S909). As the identifier of the non-recorded character string, for example, a coordinate value indicating the position of the non-recorded character string can be used.
For example, the feature amount extraction unit 192 is an individual indicating an orderer's name 501, an orderer's emblem image, a document form title 503, a greeting text 504, etc., which are non-recorded character strings included in the format of the document form 5 of FIG. The second feature amount is recorded in the database 4 in association with the identifier of the document form 5 and the identifier of the non-recorded character string.

In the database 4, for each of the plurality of formats of the document form, the image data of the document form in that format and the information of the recording character string corresponding to the image data are recorded. The acquisition unit 191 of the image processing device 1 repeats the processes of steps S901 to S909 until the image data and the recorded character string information for all the document forms are read.

When it is determined in step S901 that the acquisition unit 191 has read all the image data of the document form and the information of the recorded character string corresponding to the image data from the database 4 (step S901: YES), the group classification unit 194 documents. The forms are grouped (step S921). The group classification unit 194 divides the document form into groups based on the individual second feature amount included in the image data of the document form. For example, the group classification unit 194 groups each document form based on the degree of matching of non-recorded character strings indicated by the individual second feature, the degree of matching of emblem images, the degree of matching of the coordinate range of non-recorded character strings, and the like. To do. The group classification unit 194 determines the group identifier of the document form in this grouping process. The group classification unit 194 determines whether or not the grouping of all the document forms has been completed (step S922).

The group classification unit 194 repeats the process of step S921 when the grouping of all the document forms is not completed. Specifically, when the group classification unit 194 determines in step S922 that there is a document form for which grouping has not been completed (step S922: NO), the process returns to step S921.
When the grouping of all the document forms is completed (step S922: YES), the group classification unit 194 associates the identifier of the document form with the group identifier given to the document form and sets the group table (record) of the database 4. (Table) (step S923).

Then, the feature amount extraction unit 192 reads each individual first feature amount and each individual second feature amount of one or a plurality of document forms belonging to a certain group from the database 4, and each individual first feature amount of the document form belonging to the group. Each group first feature amount and each group second feature amount corresponding to the amount and each individual second feature amount are extracted (step S924). The first feature amount of each group may be a value such as the average of each individual first feature amount of the document forms belonging to the group. Similarly, the second feature amount of each group may be a value such as the average of each individual second feature amount of the document form belonging to the group. The first feature of each group and the second feature of each group are not the average of each individual first feature or the average of each individual second feature, but by using a method such as predetermined statistical processing or machine learning. As long as the feature amount is calculated so that the recorded character string or non-recorded character string of one or more document forms belonging to the group can be specified, any method is used for each group first feature amount and each group first. (Ii) The feature amount may be extracted.

For example, even if the feature amount extraction unit 192 extracts the character string attribute for each recorded character string from a plurality of document forms belonging to the same group in step S924, the group first feature amount is generated (directly). Good. In this case, the feature amount extraction unit 192 passes the extraction and recording of the individual first feature amount in steps S905 and S906 (no particular processing is performed).
On the other hand, the individual second feature amount is extracted by the feature amount extraction unit 192 in step S908 so that it can be used in the grouping in step S921. However, in step 921, the group classification unit 194 may group the document forms by using the non-recorded character string without using the individual second feature amount. In this case, the feature amount extraction unit 192 may extract the group second feature amount (directly) from the character string attributes of the non-recorded character strings of a plurality of document forms belonging to the same group in step S924. In this case, the feature amount extraction unit 192 does not perform any particular processing in steps S907 to S909.
The feature amount extraction unit 192 calculates each group first feature amount and each group second feature amount for each group, and records them in the database 4 in association with the group identifier (step S925).
After step S925, the image processing apparatus 1 ends the process of FIG.

Through the above processing, the image processing device 1 extracts the information (group first feature amount and group second feature amount) necessary for reducing the labor of recording the recorded character string of the worker for each group of the document form. Can be stored in the database 4. As a result, the image processing device 1 can receive the input of the image data of the new document form and automatically record the recording character string included in the document form in the database 4. The process will be described with reference to FIG.

FIG. 12 is a second diagram showing a processing flow of the image processing apparatus according to the second embodiment. FIG. 12 shows an example of a processing procedure in which the image processing device 1 extracts a recorded character string from the newly input image data.
The operator performs an operation of causing the image reading device 2 to read a new document form. As a result, the image reading device 2 generates image data of the document form and outputs (transmits) it to the image processing device 1. The acquisition unit 191 of the image processing device 1 acquires image data from the received data of the communication unit 110 (step S1001). The acquisition unit 191 outputs the image data to the feature amount extraction unit 192. The feature amount extraction unit 192 performs OCR processing on the image data to obtain the character string, the feature (character attribute) of the character included in the character string, and the coordinates in the image data of the range of the character string for each character string. Detect (step S1002). The feature amount extraction unit 192 extracts a third feature amount obtained by converting the detected information into a feature amount for each character string in the image data (step S1003). The third feature amount is information indicating the features of the character string included in the document form of the newly read image data.

Next, the group identification unit 195 reads the group second feature amount used for group identification of a new document form among the group second feature amounts stored in the database 4. The group second feature amount may be, for example, a feature amount corresponding to the orderer's emblem image 502 displayed in the image data of the document form. The group specifying unit 195 determines whether or not the information shown in the second feature amount of a certain group can be specified from the image data of the document form acquired in step S1001. The group identification unit 195 performs the same processing using the group second feature amount for all groups. When the information matching the group second feature amount read from the database 4 can be specified from the image data of the newly read document form, the group identification unit 195 newly reads the group having the group second feature amount. It is specified as a group of image data of a document form (step S1004). After that, the group identification unit 195 reads out one or a plurality of group first feature quantities for the group from the database 4 (step S1005). The group identification unit 195 outputs the third feature amount and one or a plurality of group first feature amounts to the recording unit 193. The group first feature amount is a feature amount for specifying one or more recorded character strings in the document form belonging to the group.

The recording unit 193 acquires the third feature amount for each character string in the image data and the group first feature amount for each recorded character string, and associates the group first feature amount with the third feature amount (step S1006). .. Specifically, the recording unit 193 associates each of the group first feature amounts with one third feature amount that matches the feature amount or one closest third feature amount. By this association, the recorded character string is selected from the character strings obtained by OCR processing the image data of the document form.
Next, the feature amount extraction unit 192 controls the display unit 120 to display the recorded character string, which is the processing result of the image processing device 1, and the image of the document form side by side (step S1007). Step S1007 is the same as step S706 in FIG.

Next, the recording unit 193 determines whether or not the correspondence between the display of the recorded character string and the image of the document form has been acquired (step S1008). For example, the recording unit 193 determines that the correspondence relationship has been acquired when the number of times of learning for grasping the correspondence relationship between the display of the recorded character string and the image of the document form is equal to or more than a predetermined number of times. As described above, the learning for grasping the correspondence between the display of the recorded character string and the image of the document form may be the learning for specifying the recorded character string in the display of the recorded character string, or the document. It may be learning for specifying the recorded character string in the image of a form. Alternatively, it may be a combination of these learnings.

When it is determined that the correspondence between the display of the recorded character string and the image of the document form has been acquired (step S1008: YES), the recording unit 193 controls the display unit 120 to display the recorded character string and display the document form. The correspondence with the image is displayed (step S1009). Step S1009 is the same as step S708 of FIG.
When the operator confirms and corrects the recorded character string and then performs a confirming operation, the image processing device 1 acquires the corrected recorded character string (step S1010). If the recording unit 193 determines in step S1008 that the correspondence between the display of the recorded character string and the image of the document form has not been acquired (step S1008: NO), the process proceeds to step S1010.

Then, the feature amount extraction unit 192 and the recording unit 193 as the correspondence learning unit learn the coordinates of the recorded character string in the display of the processing result (in the example of FIGS. 8 and 9, the recording character string display window W2). Step S1011). For example, the feature amount extraction unit 192 extracts the first feature amount for the display of the processing result as in the case of the document form, and the recording unit 193 uses the first feature amount to determine the position of the recorded character string. Identify and remember.
If the recording unit 193 knows the format for displaying the processing result, the processing in step S1011 is unnecessary.

The recording unit 193 records the recorded character string in the recording table in association with the identification information of the document form (step S1012).
For example, it is assumed that the third feature amount a3, the third feature amount b3, the third feature amount c3, and the third feature amount d3 can be obtained from the image data of the document form. Then, the third feature amount a3 is recorded in the database in advance, the first feature amount a1, the third feature amount b3 is the first feature amount b1, the third feature amount c3 is the first feature amount c1, and the third feature amount. It is assumed that the feature amount d3 coincides with the first feature amount d1. In this case, the recording unit 193 records the character strings corresponding to the third feature amount a3, the third feature amount b3, the third feature amount c3, and the third feature amount d3 in the recording table of the document form as recording character strings. To do. The character string corresponding to the third feature amount referred to here is the character string from which the third feature amount is extracted. When the recorded character string is modified by the operator, the recording unit 193 records the modified recorded character string in the recording table of the document form.
After step S1012, the image processing apparatus 1 ends the process of FIG.

The image processing device 1 may update the first feature amount in the document form by the processing of FIG. For example, the image processing apparatus 1 may learn the first feature amount in the document form in addition to or instead of the machine learning of the first feature amount in the display of the processing result in step S1011. In that case, the image processing apparatus 1 may re-execute the processing flow of FIG. Alternatively, the image processing device 1 may perform additional learning in step S1011 to update the first feature amount of the data processed in FIG. 6 without performing processing again.
By updating the first feature amount in the process of FIG. 12, the image processing device 1 increases the number of sample data and improves the accuracy of the first feature amount, and the accuracy of the image processing device 1 extracting the recorded character string. Is expected to improve. Further, when the recorded character string is added in the process of FIG. 12, the image processing device 1 can also extract the newly added recorded character string from the image data, which saves the worker the trouble of inputting the character string. It is expected to be omitted.

According to the process shown in FIG. 12, the image processing device 1 sets the type of the newly input document form by the image data and the recorded character string of the document form of each of the plurality of formats recorded in advance by the operator. Therefore, the recorded character string in the image data of the document form can be automatically recorded. Therefore, the image processing device 1 can reduce the labor of the operator for recording the recorded character string in the document form. Further, the operator can confirm and correct the recorded character string which is the processing result of the image processing device 1 by referring to the display of the display unit 120. At that time, the display unit 120 shows the correspondence relationship between the character string in the display of the processing result and the character string in the image of the document form, so that the worker can easily grasp the correspondence relationship.

<Third Embodiment>
As another example of the processing of the image processing device 1, the worker may register a group of document forms in the image processing device 1 in advance. For example, when the worker registers the image data of the document form in the past, he / she inputs the group identifier according to the type of the document form and registers it in the database 4 in association with the image data of the document form. As a result, different forms are not mixed in the same group due to a processing error of the image processing apparatus 1, and the first feature amount with high accuracy can be extracted. In this case, at the time of registration, the worker inputs a group of document forms, but for a new form, the group is specified using the second feature amount as in step S1004.

<Fourth Embodiment>
Further, as another example of the processing of the image processing device 1, the image processing device 1 not only groups the document forms using the second feature amount, but also uses the first feature amount and the second feature amount. In addition, the document form may be grouped by using the first feature amount. The first feature is the feature of the recorded character string, but if it is the same type of document form, it is considered that the coordinates of the recorded character string and its character attributes are the same, and the form is created using the first feature. It is possible to divide into groups. The first grouping is performed by the operator as shown in the fourth embodiment, and the new document form is grouped using the first feature amount by the processing of step S1004, so that the new document form is recorded accurately in the OCR processing. It becomes possible to read a character string.
In this case, the acquisition unit 191 acquires a plurality of form image data and the recorded character string to be recorded among the character strings included in the form image data. Then, the group classification unit 194 divides the form image data into groups based on the first feature amount. Then, the feature amount extraction unit 192 extracts the recorded character string using the first feature amount corresponding to the form image data included in the group.

<Fifth Embodiment>
In the second embodiment, a new group of forms is specified in step S1004 based on the second feature amount. However, as another processing mode, the image processing apparatus 1 does not perform the process of specifying the group, but reads out the first feature amount by sequentially specifying each group for all the groups set by the operator. Count the number that matches the third feature quantity. In the case of the correct group, the first feature amount and the third feature amount should match most, so the image processing device 1 sets the character string included in each of the third feature amounts of the specific group when the number of matches is the largest. Record. As a result, the recorded character string can be recorded without specifying the group.
In this case, the acquisition unit 191 acquires a plurality of form image data and the recorded character string to be recorded among the character strings included in the form image data. Then, the feature amount extraction unit 192 indicates the first feature amount indicating the characteristics of the recorded character string or the second recognition information other than the recorded character string based on the result of the recognition processing of the form image data acquired by the acquisition unit 191. Extract features. The feature amount extraction unit 192 extracts a recorded character string using the first feature amount corresponding to the form image data included in a predetermined group set in advance.

As described above, the recording unit 193 is recorded in advance based on the results of learning using a plurality of document images, and among the feature quantities indicating the characteristics of the character string of the item for each type of document image and for each specific item. , The character string of a specific item in the first document image is detected based on the feature amount of the displayed first document image. The display unit 120 outputs correspondence information indicating the same specific item between the first document image and the second document image displayed corresponding to the first document image.
According to the image processing device 1, the position of the character string in the document image can be acquired by learning. Therefore, according to the image processing device 1, even if the format of the document to be recognized as a character is not known in advance, it is possible to assist the operator so that he / she can grasp which item is shown where.

Further, the recording unit 193 detects a character string of the same item as the specific item in the first document image from the second document image.
As a result, the image processing device 1 can present to the user the correspondence between the character string of the specific item in the first document image and the character string of the specific item in the second document image. The user can relatively easily confirm the correctness of the character string by referring to the presentation of the correspondence.

Further, the feature amount extraction unit 192 and the recording unit 193 machine-learn the correspondence between the first document image and the second document image of the same specific item.
As a result, the image processing device 1 can automatically learn the correspondence between the first document image and the second document image of the same specific item without requiring any special treatment by the user. The user may perform normal processing, and the burden on the user does not increase.

Further, the feature amount extraction unit 192 and the recording unit 193 use the images after the character string of the second document image is determined, and the same specific items of the first document image and the second document image are used. Machine learning the correspondence of.
When the character string of the second document image is confirmed, the confirmed character string is considered to be accurate. By performing machine learning using this accurate character string, the correspondence learning unit can learn the correspondence between the first document image and the second document image of the same specific item with relatively high accuracy. Is expected.

Next, the configuration of the embodiment of the present invention will be described with reference to FIG.
FIG. 12 is a diagram showing an example of the configuration of the image processing apparatus according to the embodiment. The image processing device 600 shown in FIG. 12 includes a character string detection unit 601 and an output unit 602.
With this configuration, the character string detection unit 601 is pre-recorded based on the results of learning using a plurality of document images, and is a feature amount that indicates the characteristics of the character string of the item for each type of document image and for each specific item. Among them, the character string of the specific item in the first document image is detected based on the feature amount of the displayed first document image.
The output unit 602 outputs correspondence information indicating the same specific item between the first document image and the second document image displayed corresponding to the first document image.
According to the image processing device 600, the position of the character string in the document image can be acquired by learning. Therefore, according to the image processing device 600, even if the format of the document to be recognized as a character is not known in advance, it is possible to assist the operator so that he / she can grasp which item is shown where.

Each of the above devices has a computer system inside. A program for causing each device to perform each of the above-described processes is stored in a computer-readable recording medium of those devices, and the computer of each device reads and executes this program to perform the above-mentioned processing. Will be done. Here, the computer-readable recording medium refers to a magnetic disk, a magneto-optical disk, a CD-ROM, a DVD-ROM, a semiconductor memory, or the like. Further, this computer program may be distributed to a computer via a communication line, and the computer receiving the distribution may execute the program.

Further, the above program may be for realizing a part of the functions of the above-mentioned processing units. Further, a so-called difference file (difference program) may be used, which can realize the above-mentioned functions in combination with a program already recorded in the computer system.

Although the embodiments of the present invention have been described in detail with reference to the drawings, the specific configuration is not limited to this embodiment, and includes designs and the like within a range that does not deviate from the gist of the present invention.

1 Image processing device 2 Image reader 3 Recording device 4 Database 110 Communication unit 120 Display unit 130 Operation input unit 180 Storage unit 190 Control unit 191 Acquisition unit 192 Feature extraction unit 193 Recording unit

Claims (8)

A character string recognition unit that recognizes the character string of a specific item in the image to be recognized based on the result of learning using a plurality of images including the character string.
Output that outputs the image to be recognized and the character string recognition result in a manner in which the correspondence relationship indicating the same specific item can be grasped between the image to be recognized and the character string recognition result by the character string recognition unit. Department and
An image processing device comprising. Further, an operation reception unit for receiving a correction operation of the character string included in the character string recognition result output by the output unit is provided.
The image processing device according to claim 1, wherein the character string recognition unit records a character string corrected by a correction operation received by the operation reception unit as a recognized character string. The character string recognition unit is recorded based on the result of the learning, and among the feature quantities showing the characteristics of the character string for each type of document and each specific item shown by the plurality of images including the character string, the recognition. The image processing apparatus according to claim 1 or 2, which recognizes the character string of the specific item based on the feature amount of the character string in the target image. The image processing apparatus according to claim 3, wherein the feature amount is generated based on the attributes of characters included in the character string and information indicating the position of the character string in the image to be recognized. The image processing apparatus according to claim 4, wherein the character attribute indicates at least one of a number, an alphabet, a hiragana, a kanji, a number of characters, a character height, and a font. The character string recognition unit detects a plurality of candidates for the character string of the specific item in the image to be recognized, and detects a plurality of candidates for the character string of the specific item.
The image processing apparatus according to any one of claims 1 to 5, wherein the output unit outputs a range of each of the detected character string candidates in a manner that can be grasped. A process of recognizing the character string of a specific item in the image to be recognized based on the result of learning using a plurality of images including a character string, and
A step of outputting the image of the recognition target and the character string recognition result in a manner in which the correspondence relationship indicating the same specific item in the image of the recognition target and the character string recognition result can be grasped.
Image processing method including. On the computer
A process of recognizing the character string of a specific item in the image to be recognized based on the result of learning using a plurality of images including a character string, and
A step of outputting the image of the recognition target and the character string recognition result in a manner in which the correspondence relationship indicating the same specific item in the image of the recognition target and the character string recognition result can be grasped.
A program to execute.

Images