JP2021157249A - Image acquisition system for OCR and image acquisition method for OCR - Google Patents

Abstract

To provide an image acquisition system for OCR that suppresses the transmission of document image data that cannot be subjected to character recognition processing to a server for OCR processing and reduces a load on the server due to the OCR processing.SOLUTION: The present invention relates to a system for acquiring a document image of a document that is a target of character recognition processing and transmitting the document image to a server for OCR processing, and includes a module group that is downloaded when performing acquisition processing for the document image to be subjected to character recognition and is included in an image acquisition application that operates with a web browser as a platform. The module group includes: a document image imaging module that causes an imaging device to image the document and generates the document image; an image form determination module that determines whether or not the document image is captured as an image form corresponding to a predetermined determination rule; and a document image selection module that performs a selection as the document image to be transmitted to the server that satisfies the form in the determination.SELECTED DRAWING: Figure 1

Description

The present invention relates to an OCR image acquisition system and an OCR image acquisition method.

In recent years, image data obtained by imaging identification cards such as licenses, My Number cards, and My Number notification cards using user terminals such as mobile phones and smartphones, and the image data are subjected to character recognition processing (OCR (Optical Character Recognition) processing). By using the text character information obtained by doing so, various application procedures such as opening a bank account have come to be performed electronically.
As a technique for performing OCR processing of document image data captured by an image pickup device of a user terminal, for example, there is a device described in Patent Document 1.

In the technique disclosed in Patent Document 1, for example, OCR processing is performed on each document image data of the identification card obtained by continuously imaging the identification card, and the identification card is included in the identification card. Acquires character information such as characters and numerical values of the item to be read required in the contract.
Then, based on the acquired character information of each item, the application form corresponding to the user is generated by filling in the entry field of the input form required at the time of the application procedure.

Japanese Patent No. 6364051

In the above-mentioned OCR processing, for example, each of a plurality of document image data obtained by capturing an image of a document to be recognized is subjected to OCR processing, the number of document image data for which the same recognition result is obtained is counted, and a predetermined set number is obtained. The text characters acquired from the above document image data are output as the recognition result of the character recognition process in the OCR process.
Therefore, in order to perform the OCR processing, it is necessary to transmit a plurality of document image data from the user terminal to the application reception server that performs the OCR processing.

However, depending on the form (imaging state) of each image of the transmitted document image data, character recognition may not be possible and the above recognition result may not be obtained. It will be sent back to the application reception server.
For example, when a user transmits a captured image of a quality that cannot be OCR processed, such as the entire identification card is not captured or the captured image is too small, the above recognition is performed in the OCR processing. I can't get the result.
As a result, the traffic in the communication network between the user terminal and the application reception server increases by repeating the transmission of a large amount of data a plurality of times.

Further, due to the above-mentioned problem of the quality of the document image data, the transmission of the document image data is repeated a plurality of times, so that the number of times of performing the OCR processing also increases, and the load of the OCR processing on the application reception server also increases.
Due to the increase in the traffic of the communication network and the increase in the load on the application reception server described above, the processing of the application procedure is delayed, and it takes time to complete the application procedure using the Internet. User satisfaction in processing is reduced.

The present invention has been made in view of such circumstances, and is an image for OCR that suppresses the transmission of document image data that cannot be subjected to character recognition processing to a server that performs OCR processing and reduces the load due to OCR processing on the server. An object of the present invention is to provide an acquisition system and an image acquisition method for OCR.

The present invention has been made to solve the above-mentioned problems, and the image acquisition system for OCR of the present invention acquires a document image of a document to be subjected to character recognition processing, and is used for a server that performs OCR processing. , An image acquisition system for OCR that transmits the acquired document image, and is a module included in an image acquisition application that is downloaded when the document image acquisition process for performing the character recognition process is performed and operates using a web browser as a platform. It is a group, and the module group is a document image imaging module that causes an imaging device to image a document to generate the document image, and whether or not the document image is imaged in the form of an image corresponding to a predetermined determination rule. It is characterized by including an image form determination module that performs the determination of the above, and a document image selection module that selects a document image satisfying the above form as the document image to be transmitted to the server.

In the image acquisition system for OCR of the present invention, the determination rule can determine whether or not character recognition processing of a character image in a document image is possible with higher accuracy than other image areas in the document image. It is characterized in that it is a regulation for the form of a predetermined image area.

The OCR image acquisition system of the present invention is characterized in that the image morphology determination module determines the type of the document of the document image and transmits the document image to the server together with the document image.

The OCR image acquisition system of the present invention is characterized in that the determination rule is set for each of the document types according to the characteristic form of the document for each type.

The OCR image acquisition method of the present invention is an OCR image acquisition method that acquires a document image of a document that is a target of character recognition processing and transmits the acquired document image to a server that performs OCR processing. , Each of the document image imaging module, the image morphology determination module, and the document image selection module, which are downloaded when the document image acquisition process for performing the character recognition process is performed and are included in the image acquisition application that operates using a web browser as a platform. A document image imaging process in which the document image imaging module causes an imaging device to image a document to generate the document image, and an image in which the document image corresponds to a predetermined determination rule by the image morphology determination module. The image form determination process for determining whether or not the image is captured as the form of the above, and the document image selection module selects a document image satisfying the form in the determination as the document image to be transmitted to the server. It is characterized by including a document image selection process.

According to the present invention, there is provided an OCR image acquisition system and an OCR image acquisition method that suppress the transmission of document image data that cannot be subjected to character recognition processing to a server that performs OCR processing and reduce the load of OCR processing on the server. can do.

It is a conceptual diagram which shows the structural example of the image acquisition system for OCR by embodiment of this invention. It is a figure explaining the generation of the input form of the application form based on the document image data obtained by the image acquisition system for OCR of this embodiment. It is a flowchart which shows the operation example of the process of the acquisition of the document image data of the identification card by the image acquisition system for OCR of this embodiment.

The image acquisition system for OCR of the present invention captures identification cards such as a driver's license, My Number card, and My Number notification card by user's operation when performing various application procedures such as opening a bank account using the Internet. ..
Then, the image acquisition system for OCR of the present invention selects the document image data in a form capable of OCR processing from the document image data which is the image data captured by this identification card, and the selected document image. The data is sent to the application reception server that accepts applications. As a result, the user terminal of the user does not send a document image in a form unsuitable for OCR processing to the application reception server, the traffic on the communication network is suppressed, and the application reception server performs the document image before the OCR processing. The load of preprocessing can be reduced.

Hereinafter, embodiments of the present invention will be described with reference to the drawings. FIG. 1 is a conceptual diagram showing a configuration example of an image acquisition system for OCR according to an embodiment of the present invention. In FIG. 1, the user terminal 100 and the application reception server 200 are connected by a communication network 300 including the Internet.
The user terminal 100 is a terminal such as a smartphone owned by the user. The application reception server 200 is a server that accepts applications for applications such as opening a bank account and joining a sports club.
The user terminal 100 includes an OCR image acquisition system 1, a web browser (web (World Wide Web) browser) 2, and a display unit 3.

The OCR image acquisition system 1 in the present embodiment is formed by an image acquisition application that operates using a web browser provided in a user terminal 100 owned by the user as a platform. That is, when the image acquisition application is started on the web browser, it is expanded as a plurality of modules (module groups) on the memory of the user terminal 100, and the function of the OCR image acquisition system 1 is realized. Further, the program of the image acquisition application is downloaded from the application reception server 200 to each user terminal 100 every time the user performs an application process by the user terminal 100 (it becomes necessary to perform a character recognition process), and the user terminal 100 has a program. It is written and stored in the storage unit, and is deleted from the storage unit after the application process is completed.
The OCR image acquisition system 1 includes a data input / output module 11, an imaging control module 12, an image form determination module 13, a document image selection module 14, and a determination rule table 15.

The data input / output module 11 transmits / receives data to / from the application reception server 200 via the web browser 2.
The image pickup control module 12 displays the image pickup guide frame on the display unit 3, causes the user to perform an image pickup operation with the identification card included in the image pickup guide frame, and displays the document image data which is the image capture image of the document. get.
Here, the image pickup control module 12 drives an image pickup device (not shown) mounted on the user terminal 100 to perform continuous shooting of the captured images of the identification card at a predetermined cycle, and a plurality of images (for example, 15). Acquires the text image data (which may be a frame image of a moving image).
For example, if the identification card is a driver's license and My Number card, the document image data on the back and front is acquired, and if the identification card is My Number notification card, the document image data on the front is acquired.

The image form determination module 13 determines whether or not the form of the document object image, which is an image portion of the document in the document image data, corresponds to a predetermined determination rule set in advance. In the present embodiment, the determination of whether or not the character recognition processing of the character image in the document image data is possible is more accurate than the other image areas in the document object image of the document image data. It is set as a regulation for the form of a predetermined image area (outermost frame described later) of the document object image that can be performed in.
Here, the predetermined determination rule is included as the determination rule table 15 in the program of the downloaded image acquisition application, and at least the shape of the outermost frame of the document object image (the shape of the outline in the plan view of the document). Is a quadrangle (4 vertices), the outermost frame line of the document object image (for example, a line near the character image) is detected, or the inner angle of the quadrangle that is the outline of the document object image is Whether or not it is within a predetermined angle range (determination of the degree of shape deformation such as a trapezoidal shape), the line of the outermost frame of the document object image (for example, the line segment near the character image) is curved (distorted). Whether or not the area ratio shown in the document image data of the document object image is equal to or greater than a predetermined ratio (determination of the size of the document object image).
Further, in the predetermined determination rule, whether or not the hue in the outermost frame of the document object image is within the preset hue range, whether or not the saturation is within the preset saturation range, and whether or not the brightness is preset are set. It may also include whether or not it is in the lightness range.

That is, when the document object image is not a quadrangle, it is out of the imaging guide frame, indicating that the information for acquiring the document image when performing the OCR processing is missing. If the contour line is not detected, it means that the range of the document object image cannot be extracted, and therefore the keystone correction of the document image cannot be performed. The internal angle of the quadrangle is imaged diagonally and deformed into a trapezoidal shape, indicating that if the angle deviates from a predetermined angle, the quality of the character image will not be sufficient for OCR processing even if the trapezoidal correction is performed. .. If the contour line is curved, the character image is also distorted, indicating that the image is not captured with a quality capable of OCR processing. When the area ratio is less than a predetermined ratio, it indicates that the document image is not captured in a size capable of character recognition in the OCR processing.
In addition, when the hue in the outermost frame of the document object image is not in the preset hue range, when the saturation is not in the preset saturation range, and when the brightness is not in the preset brightness range, each of the documents It shows that the object image is not captured as a character image having sharpness capable of performing OCR processing.

The document image selection module 14 confirms the determination result of each sentence image data captured by the image form determination module 13, selects the document image data corresponding to the above determination rule, and deletes the other document image data. ..
Then, the document image selection module 14 transmits the selected document image data to the application reception server 200 via the data input / output module 11 and the web browser 2.

The application reception server 200 performs preprocessing such as keystone correction of the shape of the document object image, size adjustment, and rotation before performing OCR processing for each of the document image data supplied from the user terminal 100.
Then, the application reception server 200 determines whether the document object image for which the preprocessing has been completed is the image of the document of the driver's license, the My Number card, or the My Number notification card.
For example, when the application reception server 200 determines that the document object image in the document image data is the image of the license, the application reception server 200 compares the template image of the license with the document object image and character images of the items required for the application form. Is extracted, and OCR processing (character recognition processing) is performed to acquire a character image.

Here, the character image is information obtained by performing OCR processing on the document object image, and includes an image indicating an item and an image of characters indicating item data. The image showing the item is an image of a character string of the name printed in the margin of the item such as a name and an address. The image showing the data of the item is an image of personal data printed inside the column, for example, an image of a character string (including numbers, symbols, etc.).

Then, the application reception server 200 performs OCR processing for each of the document object images whose preprocessing has been completed, and when the document image of each item is converted into the same text character in three or more sentence image data, the text is displayed. Characters are used as the result of converting character images to text.
Here, in the present embodiment, only the document image data determined to be capable of OCR processing is transmitted, so that when the text character is determined as the conversion result, it is converted to the same text character as in the conventional case. It is possible to reduce the number of document images, for example, if the number is 3 or more in the past, the number is reduced to 2 or more to reduce the load in the OCR processing of the application reception server 200. be able to.

The data input / output module 11 of the user terminal 100 is an item necessary for application, in which a character image of document image data is converted into text characters by OCR processing in the application reception server 200 for a predetermined field of an application input form. Data is written (display area 501 and display area 502 in FIG. 2C described later).
Further, at this time, the application reception server 200 performs a compositing process of pasting the document object image 510 on the input form, and at the time of this compositing process, the sensitive information portion 510B in the document object image 510 (image of the identification card). Is invisible (for example, blackened or whitened, or erased a character image, etc., hereinafter referred to as masking).

FIG. 2 is a diagram illustrating the generation of an application form input form based on the document image data obtained by the OCR image acquisition system of the present embodiment.
FIG. 2A shows an identification card displayed on the display screen 3S of the display unit 3 by the image pickup control module 12 when the image acquisition application is downloaded from the application reception server 200 and started on the web browser. This is an example of an image that encourages imaging.
In FIG. 2A, the display area 501 is an area in which the document image data of the identification card is OCR-processed and the character string of the user's name is displayed. The display area 502 is an area in which the document image data of the identification card is OCR-processed and the character string of the user's address is displayed.
Further, the button 503 is a button displayed so that the user can press (touch) the image of the identification card.

FIG. 2B is an example of an image in which the imaging control module 12 performs imaging processing of an identification card, which is displayed on the display screen 3S of the display unit 3 when the user presses the button 503 in FIG. 2A. Is. The image pickup control module 12 displays the image pickup guide frame 504 on the display screen 3S of the display unit 3. The user holds the user terminal 100 at a position where the document object image 600 of the identification card is included in the image pickup guide frame 504 to take an image of the identification card.
The button 507 is a shutter button for causing the image pickup device of the user terminal 100 to take an image of the identification card. The image pickup control module 12 acquires a plurality of predetermined text image data by pressing (touching) the button 507 (shutter button).

FIG. 2C is an example of an image of the input form generated by the application reception server 200 after the document image data is subjected to OCR processing. When the input form is supplied from the application reception server 200, the image pickup control module 12 displays the input form of FIG. 2C on the display screen 3S of the display unit 3.
In FIG. 2B, the display area 501 is an area in which the document image data of the identification card is OCR-processed and the character string of the user's name is displayed. In the display area 501, a character string of the user's name converted from a character image indicating the user's name into text characters by OCR processing of the document image data is displayed.

The display area 502 is an area in which the document image data of the identification card is OCR-processed and the character string of the user's address is displayed. In the display area 502, a character string of the user's address converted from a character image indicating the user's address into text characters by OCR processing of the document image data is displayed.
Further, the document object image is an image of a document object having a shape extracted from the document image and corrected by preprocessing, and is pasted (combined) in an input format and displayed. The portion 510B in which the sensitive information of the document object image is described is, for example, painted black or masked as white.

According to the present embodiment, according to the above-described configuration, the document image data is imaged by the image acquisition device mounted on the user terminal 100, and each of the document object images in the plurality of imaged document image data is captured. , It is included in the image pickup guide frame 504 because it is determined whether or not the image is captured in a form capable of character recognition and only the document image data captured in the possible form is transmitted to the application reception server 200. Conventionally, document image data that cannot be recognized by OCR processing, such as erroneous imaging even though it is not present, the imaging angle when imaging the identification card is inappropriate, or the size of the document object image is small, is conventionally obtained. In the case of transmission such as This makes it possible to reduce the communication load on the communication network 300.

Further, according to the present embodiment, by supplying the document image data captured in a form capable of character recognition, the document image data for the quality that the application reception server 200 cannot perform character recognition as in the conventional case can be obtained. On the other hand, since it is not necessary to perform the pre-processing and the OCR processing, the load on the OCR processing on the application reception server 200 can be reduced.

Further, according to the present embodiment, the image acquisition application is deployed and operated in each module in the OCR image acquisition system 1 on the web browser, and a part of the preprocessing in the OCR processing is performed on behalf of the image. The form determination process of the document object image performed by the form determination module 13 can be executed without sending and receiving data necessary for determination between the application reception server 200 and the web browser 2. It is no longer necessary to send and receive data to and from the document image selection module 14, and the time required to select the document image data of the document image selection module 14 and send it to the application reception server 200 is shortened, so that the time from the acquisition of the document image data to the OCR is shortened. The process can be performed in a short time, and can be executed in real time from the user taking the image of the identification card to the acquisition of the image of the input form shown in FIG. 2 (c).

Further, according to the present embodiment, an application for acquiring document image data capable of character recognition by OCR processing is downloaded from the application reception server 200 for each application processing and executed on the web browser of the user terminal 100. Since it is configured as a form to be used, the application reception server 200 can easily obtain the latest version of the document acquisition application, as compared with the case where the application operating on the OS as a platform requires time and effort for the user such as version upgrade. Can be provided from.

FIG. 3 is a flowchart showing an operation example of the process of acquiring the document image data of the identification card by the image acquisition system for OCR of the present embodiment. In the following description, acquisition of document image data of a driver's license when opening a bank account will be described as an example.
Step S101: The user uses the smartphone as the user terminal 100 to access the bank application reception server 200 in order to open a bank account via the Internet in the communication network 300.

Step S102: The user presses (touches) the download button of the image acquisition application on the account opening explanation page displayed by the application reception server 200.
As a result, the web browser 2 downloads the image acquisition application, and writes and stores the program of the image acquisition application in the storage unit (not shown) of the user terminal 100.

Step S103: The user starts an image acquisition application in order to acquire the document image data to be transmitted to the application reception server 200.
At this time, the web browser 2 expands the image acquisition application on the memory and generates each of the data input / output module 11, the imaging control module 12, the image form determination module 13, the document image selection module 14, and the determination rule table.
Then, the image pickup control module 12 displays the screen for prompting the image capture of the driver's license shown in FIG. 2A on the display screen 3S of the display unit 3 on the user terminal 100.

Step S104: The user presses the button 503 in FIG. 2A to capture an image of the driver's license.
As a result, when the image pickup control module 12 detects that the button 503 is pressed, the image pickup control module 12 displays the image of the identification card shown in FIG. 2B on the display screen 3S of the display unit 3.

The user can include the document object image 600 of the driver's license in the image pickup guide frame 504 on the display screen 3S of the display unit 3, that is, so that the entire outer circumference of the driver's license fits in the image pickup guide frame 504. , Adjust the imaging position of the driver's license.
Then, the user confirms that the document object image is included in the imaging guide frame 504, and presses the button 507.
When the image pickup control module 12 detects that the button 507 is pressed, the image pickup control module 12 takes an image of a plurality of documents, for example, 15 documents, by an image pickup device at a predetermined cycle.

Step S105: The image form determination module 13 determines whether or not the form of the document object image of the document image data satisfies each determination rule in the determination rule table 15.
Then, when the form of the document object image of the document image data does not deviate from the definition of the determination rule, the image form determination module 13 adds a flag indicating appropriateness to the document image data.
On the other hand, when the form of the document object image of the document image data is out of the definition of the determination rule, the image form determination module 13 adds a flag indicating improperness to the document image data.

Step S106: The document image selection module 14 confirms the flag given to the document image data, and determines whether or not it indicates suitability (corresponds to the determination rule).
At this time, the document image selection module 14 advances the process to step S107 when the flag is appropriate, and proceeds to step S110 when the flag is inappropriate.

Step S107: The document image selection module S109 transmits the document image data selected by the document image selection module 14 as appropriate for the flag to the application reception server 200 via the data input / output module 11.

Step S108: The document image selection module 14 has performed selection processing for all document image data captured by the imaging control module 12, that is, document object images in all 15 document image data (selection processing has been completed). Judge whether or not.
At this time, when the document image selection module 14 completes the selection process for all the document image data captured by the image pickup control module 12, the process proceeds to step S109.
On the other hand, when the document image selection module 14 has not completed the selection process for all the document image data captured by the image pickup control module 12, the document image selection module 14 advances the process to step S105 in order to perform the next sentence image data selection process. ..

Step S109: The application reception server 200 performs preprocessing on each of the document images supplied from the document image selection module 14 of the user terminal 100, and then performs OCR processing to convert character images of necessary items into text. Convert.
Then, the application reception server 200 describes a character string in the required item field, synthesizes the preprocessed document object image, generates the input form shown in FIG. 2C, and uses the generated input form as the user. It is transmitted to the terminal 100.

Step S110: The data input / output module 11 receives the input form supplied from the application reception server 200 and displays it on the display screen 3S of the display unit 3 on the user terminal 100.
As a result, the user confirms the input form displayed on the display screen 3S of the display unit 3, and uses this input form to apply to the application reception server 200 for opening a bank account.
Then, the data input / output module 11 erases the program of the image acquisition application in the storage unit of the user terminal 100 after the above-mentioned application process is completed.

Step S111: The document image selection module S109 does not perform a process in which the document image selection module 14 deletes the document image data whose flag is not appropriate and transmits it to the application reception server 200.

In the above-described embodiment, the application reception server 200 is configured to perform preprocessing of OCR processing on the document image data in the form corresponding to the determination rule supplied from the user terminal 100. The document image selection module may also perform preprocessing for OCR processing, and the application reception server 200 may perform only OCR processing.

Further, in the present embodiment, the template (reference image) of the format for each type of the ID card or the type discrimination information such as the judgment standard of the reading process of the ID card type is included in the image acquisition application and downloaded. As a result, the type of the identification card of the document object image in the document image data is determined based on the above type identification information, and the document image data to which the information of the determined identification card type is added is transmitted to the application reception server. May be.

Further, in the present embodiment, a judgment rule table included in the image acquisition application to be downloaded is provided for each identification card, and the judgment rule table corresponding to the identification card determined from the above-mentioned document image data is referred to. , The configuration may be such that it is determined whether or not the document image data is in a form corresponding to the determination rule. In the case of this configuration, the judgment rule is that the judgment of whether or not the character recognition of the character image in the document object image is possible is higher than that of other parts, or the highest accuracy in the document, for each type of identification card. Image morphology such as line state (blurring, distortion, etc.) in the part (area) that can be done, angle of the internal angle of the quadrangle in the column or symbolic mark state (blurring, distortion, etc.), hue, re-brightness, etc. Includes the provisions of.
Further, in the present embodiment, the outermost frame, the ruled line forming the frame of the column, or the symbolic mark is used as an example, but if the image region has a known form, the determination is made. Any image region may be used because the accuracy can be easily improved.

Further, in the present embodiment, there are many document image data determined to be unsuitable by the image form determination module 13, and the number of document image data determined to be appropriate is used to determine the result of character recognition in the OCR process of the application reception server 200. If the number of document image data to be used is less than the number, a notification prompting the addition of the imaging process may be displayed on the display screen 3S of the display unit 3.

Further, the program of the image acquisition application for realizing the function of the image acquisition system 1 for OCR in FIG. 1 is recorded on a computer-readable recording medium, and the program recorded on the recording medium is read into the computer system. By executing the program, it may be determined whether or not the form of the document object image in the text image data corresponds to the determination rule. The term "computer system" as used herein includes hardware such as an OS and peripheral devices.

Further, the "computer system" includes a homepage providing environment (or a display environment) if a WWW system is used.
Further, the "computer-readable recording medium" refers to a portable medium such as a flexible disk, a magneto-optical disk, a ROM, or a CD-ROM, or a storage device such as a hard disk built in a computer system. Further, a "computer-readable recording medium" is a communication line for transmitting a program via a network such as the Internet or a communication line such as a telephone line, and dynamically holds the program for a short period of time. In that case, it also includes the one that holds the program for a certain period of time, such as the volatile memory inside the computer system that is the server or client. Further, the above-mentioned program may be a program for realizing a part of the above-mentioned functions, and may be a program for realizing the above-mentioned functions in combination with a program already recorded in the computer system.

Although the embodiment of the present invention has been described in detail with reference to the drawings, the specific configuration is not limited to this embodiment, and the design and the like within a range not deviating from the gist of the present invention are also included.

1 ... Image acquisition system for OCR 2 ... Web browser 3 ... Display unit 3S ... Display screen 11 ... Data input / output module 12 ... Imaging control module 13 ... Image form judgment module 14 ... Document image selection module 15 ... Judgment rule table 100 ... User Terminal 200 ... Application reception server 300 ... Communication network

Claims (5)

This is an OCR image acquisition system that acquires a document image of a document that is the target of character recognition processing and transmits the acquired document image to a server that performs OCR processing.
It is a module group included in an image acquisition application that is downloaded when performing the document image acquisition process that performs the character recognition process and operates using a web browser as a platform.
A document image imaging module that causes an image pickup device to image a document and generates the document image,
An image form determination module that determines whether or not the document image is captured as an image form corresponding to a predetermined determination rule, and an image form determination module.
An image acquisition system for OCR, comprising: a document image selection module that selects a document image satisfying the above-mentioned form in the determination as the document image to be transmitted to the server. The determination rule is a regulation for a predetermined image region form in which it is possible to determine whether or not character recognition processing of a character image in a document image is possible with higher accuracy than other image regions in the document image. The image acquisition system for OCR according to claim 1. The image morphology determination module
The image acquisition system for OCR according to claim 1 or 2, wherein the type of the document of the document image is determined and transmitted to the server together with the document image. Any one of claims 1 to 3, wherein the determination rule is set for each of the document types according to the characteristic form of the document for each type. Image acquisition system for OCR described in. This is an OCR image acquisition method that acquires a document image of a document that is the target of character recognition processing and transmits the acquired document image to a server that performs OCR processing.
Each of the document image imaging module, the image morphology determination module, and the document image selection module, which are downloaded when the document image acquisition process for performing the character recognition process and are included in the image acquisition application operating using a web browser as a platform, are included. make use of,
A document image capturing process in which the document image capturing module causes an imaging device to image a document to generate the document image.
An image form determination process in which the image form determination module determines whether or not the document image is captured as an image form corresponding to a predetermined determination rule.
An image acquisition method for OCR, wherein the document image selection module includes a document image selection process of selecting a document image satisfying the above-mentioned form in the determination as the document image to be transmitted to the server.

Images