Recognition result detection method, device, equipment and medium based on text recognition

By performing text recognition and non-empty detection analysis on the form image, adding verification marks and displaying results, the problem of easy omission of manually detecting coordinate information in the prior art is solved, and high accuracy and high efficiency detection results are achieved.

CN114937272BActive Publication Date: 2024-11-22CHINA PING AN LIFE INSURANCE CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202210581573.2
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2022-05-26
Publication Date
2024-11-22
Estimated Expiration
2042-05-26

AI Technical Summary

Technical Problem

In the existing OCR technology, omissions are prone to occur when manually detecting coordinate information, resulting in inaccurate and inefficient detection of detection results, and failure to fully check the redundant position coordinates, resulting in inaccurate detection results.

Method used

By identifying the form image in text, obtaining field information and coordinate information, performing non-empty detection and analysis, adding verification marks, and displaying detection results on the image, ensuring the detection of all field values ​​and coordinate information.

Benefits of technology

Intelligent automatic detection is realized, the accuracy and efficiency of detection results are improved, the accuracy of coordinate information is ensured, omissions are reduced, and detection efficiency is improved.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN114937272B_ABST
    Figure CN114937272B_ABST
Patent Text Reader

Abstract

The present invention relates to the field of image processing, and in particular to a recognition result detection method, device, equipment and medium based on text recognition, wherein the method comprises: performing text recognition on a form image to obtain a recognition result; based on a number of field names, performing detection and analysis on the non-emptiness of a number of field values ​​and coordinate information to obtain a detection result; on the form image, adding a verification mark to the field value according to the detection result to obtain a verification image; setting a result display area in an adjacent area of ​​the verification image to obtain a detection result image; the result display area is used to display the detection result. The present invention can realize intelligent automatic detection and obtain a detection result by performing detection and analysis on the non-emptiness of a number of field values ​​and coordinate information, and the detection result takes into account the empty and non-empty conditions of all field values ​​and the empty and non-empty conditions of all coordinate information in the form image, and is not prone to omission, thereby improving the accuracy of the detection result.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of image processing, and in particular to a recognition result detection method, device, equipment and medium based on text recognition. Background Art

[0002] With the rapid development of OCR (Optical Character Recognition) technology, more and more industries use OCR technology to analyze and process text in image files to obtain field information.

[0003] In the existing OCR technology, after performing OCR text recognition and field extraction on an image file, the field information in the image file and the coordinate information corresponding to the field information are returned. The user needs to manually detect the coordinate information, that is, locate the position corresponding to the text information in the image file according to the field information and coordinate information, confirm the text information in the image file, and detect whether the coordinate information is correct.

[0004] In the existing process of detecting coordinate information, since the image file, text information and coordinate information are not displayed in the same interface, it is easy to miss the coordinate information during manual detection, resulting in inaccurate detection results and low detection efficiency. Secondly, in the existing process of detecting coordinate information, it is not considered to check the redundant position coordinates that are identified, resulting in incomplete detection and inaccurate detection results. Summary of the invention

[0005] Based on this, it is necessary to provide a recognition result detection method, device, equipment and medium based on text recognition to address the above technical problems, so as to solve the problem in the prior art that omissions are easily occurred in the process of detecting coordinate information, resulting in inaccurate detection results and low detection efficiency.

[0006] A recognition result detection method based on text recognition, comprising:

[0007] Performing text recognition on the form image to obtain a recognition result; the recognition result includes field information and coordinate information; the field information includes a number of field names and field values ​​corresponding to the field names; the coordinate information includes the position coordinates of the field values;

[0008] Based on the field names, the non-emptiness of the field values ​​and the coordinate information is detected and analyzed to obtain a detection result;

[0009] On the form image, adding a verification mark to the field value according to the detection result to obtain a verification image;

[0010] A result display area is set in an adjacent area of ​​the verification image, and the detection result is displayed in the result display area to obtain a detection result image.

[0011] A recognition result detection device based on character recognition, comprising:

[0012] The recognition result module is used to perform text recognition on the form image to obtain a recognition result; the recognition result includes field information and coordinate information; the field information includes a number of field names and field values ​​corresponding to the field names; the coordinate information includes the position coordinates of the field values;

[0013] A detection result module, used for detecting and analyzing the non-emptiness of the field values ​​and coordinate information based on the field names to obtain a detection result;

[0014] A verification image module, used to add a verification mark to the field value on the form image according to the detection result to obtain a verification image;

[0015] The detection result image module is used to set a result display area in an adjacent area of ​​the verification image, display the detection result in the result display area, and obtain a detection result image.

[0016] A computer device comprises a memory, a processor and computer-readable instructions stored in the memory and executable on the processor, wherein the processor implements the above-mentioned recognition result detection method based on text recognition when executing the computer-readable instructions.

[0017] One or more readable storage media storing computer-readable instructions, wherein when the computer-readable instructions are executed by one or more processors, the one or more processors execute the recognition result detection method based on text recognition as described above.

[0018] The above-mentioned recognition result detection method, device, equipment and medium based on text recognition obtain recognition results by performing text recognition on the form image; the recognition results include field information and coordinate information; the field information includes several field names and field values ​​corresponding to the several field names; the coordinate information includes the position coordinates of several field values; based on the several field names, the non-emptiness of several field values ​​and coordinate information is detected and analyzed to obtain the detection result; on the form image, a verification mark is added to the field value according to the detection result to obtain a verification image; a result display area is set in the adjacent area of ​​the verification image, and the detection result is displayed in the result display area to obtain a detection result image. The present invention can realize intelligent automatic detection and obtain detection results by detecting and analyzing the non-emptiness of several field values ​​and coordinate information. The detection result takes into account the empty and non-empty conditions of all field values ​​in the form image and the empty and non-empty conditions of all coordinate information, and is not prone to omission, which can improve the accuracy of the detection result. Furthermore, based on the detection results, a verification mark is added to the field value on the original form image, and an annotation box is spliced ​​on the form image to display the detection results. When the position coordinates of the field value in the field information are further verified manually, the accuracy of the coordinate information in the form image can be quickly verified according to the verification mark and the detection results in the annotation box, thereby improving the detection efficiency, reducing omissions, and improving the accuracy of coordinate information detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments of the present invention will be briefly introduced below. Obviously, the accompanying drawings in the following description are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative labor.

[0020] Figure 1 is a schematic diagram of an application environment of a recognition result detection method based on text recognition in an embodiment of the present invention;

[0021] Figure 2 is a flow chart of a recognition result detection method based on text recognition in one embodiment of the present invention;

[0022] Figure 3 is a structural schematic diagram of a recognition result detection device based on character recognition in one embodiment of the present invention;

[0023] Figure 4 is a schematic diagram of a computer device in one embodiment of the present invention. DETAILED DESCRIPTION

[0024] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0025] The recognition result detection method based on text recognition provided in this embodiment can be applied to Figure 1 In an application environment, a client communicates with a server. The client includes but is not limited to various personal computers, laptops, smart phones, tablet computers, and portable wearable devices. The server can be implemented with an independent server or a server cluster consisting of multiple servers.

[0026] In one embodiment, if Figure 2 As shown, a recognition result detection method based on text recognition is provided, and the method is applied in Figure 1 The server in the example is used as an example to illustrate the following steps:

[0027] S10. Perform text recognition on the form image to obtain a recognition result; the recognition result includes field information and coordinate information; the field information includes a number of field names and field values ​​corresponding to the field names; the coordinate information includes the position coordinates of the field values.

[0028] It is understandable that the form image refers to an image obtained by scanning the form through a camera or a scanning device. Among them, the form refers to a document with a specific form form. For example, an insurance policy, an invoice, a financial statement, etc. Generally, a form includes at least one field name and at least one field value, and one field name corresponds to one field value. Among them, the field name contains at least one character, and the field value contains at least one character. For example, the field name in form A is "user name", and the corresponding field value can be "Zhang San". The recognition result refers to the result obtained by performing text recognition on the text information in the form image through a text recognition algorithm. Among them, text recognition refers to the process of recognizing the text information in the form through a text recognition algorithm, obtaining several field names and several field values ​​corresponding to the field names, and obtaining the position coordinates of the field values. The position coordinates refer to the position information of the field value in the form image. Generally, the position coordinates can be the coordinates of four pixel points of the field value. For example, the position coordinates of the field value "Zhang San" are {(X1, Y1), (X2, Y2), (X3, Y3), (X4, Y4)}.

[0029] S20. Based on the field names, detect and analyze the non-emptiness of the field values ​​and coordinate information to obtain a detection result.

[0030] It is understandable that a field name corresponds to a field value, and a field value corresponds to a position coordinate. Field information includes several field names and field values ​​corresponding to the several field names. Coordinate information includes the position coordinates of several field values. Detecting and analyzing the non-emptiness of several field values ​​and coordinate information refers to judging the non-emptiness of each field value and each position coordinate corresponding to the field value, and classifying the field name, position coordinate and field value according to the judgment result, and finally obtaining the detection result. The detection result includes the field name, field value, classification type and sorting sequence number. Among them, the sorting sequence number refers to the arrangement sequence number of the field name. According to the arrangement sequence number of the field name, the position of the field name in the form image can be quickly located.

[0031] S30. Adding a verification mark to the field value on the form image according to the detection result to obtain a verification image.

[0032] It is understandable that adding a check mark to a field value means adding a check mark to the field value corresponding to the field name according to the classification type of the field name in the detection result. When the coordinate information is further manually proofread later, the position coordinates can be quickly located according to the check mark, which can improve the accuracy and efficiency of coordinate information detection. Among them, the check mark is used to identify the position coordinates of the field value. Preferably, different check marks are set for different classification types. The difference in the check mark can be reflected by the color, shape, size, etc. of the check mark. Here, the shape, color, size, etc. of the check mark are not limited and can be set according to the actual needs of the user. The check image refers to an image on the original form image with a check mark added to all field values ​​in the form image.

[0033] S40, setting a result display area in an adjacent area of ​​the verification image, and displaying the detection result in the result display area to obtain a detection result image.

[0034] It is understandable that the result display area refers to the area spliced ​​next to the verification image for displaying the test results. Preferably, the result display area is spliced ​​on the left or right side of the verification image, and the height of the result display area is equal to the height of the verification image; the width of the result display area is dynamically adjusted according to the length of the longest character in the test result. The test result image includes the verification mark and the test result.

[0035] In steps S10-S40, by performing text recognition on the form image, a recognition result is obtained; the recognition result includes field information and coordinate information; the field information includes several field names and field values ​​corresponding to the several field names; the coordinate information includes the position coordinates of the several field values; based on the several field names, the non-emptiness of the several field values ​​and coordinate information is detected and analyzed to obtain a detection result; on the form image, a verification mark is added to the field value according to the detection result to obtain a verification image; a result display area is set in an adjacent area of ​​the verification image, and the detection result is displayed in the result display area to obtain a detection result image. The present invention can realize intelligent automatic detection and obtain a detection result by detecting and analyzing the non-emptiness of several field values ​​and coordinate information. The detection result takes into account the empty and non-empty conditions of all field values ​​in the form image and the empty and non-empty conditions of all coordinate information, and is not prone to omission, which can improve the accuracy of the detection result. Furthermore, based on the detection results, a verification mark is added to the field value on the original form image, and an annotation box is spliced ​​on the form image to display the detection results. When the position coordinates of the field value in the field information are further verified manually, the accuracy of the coordinate information in the form image can be quickly verified according to the verification mark and the detection results in the annotation box, thereby improving the detection efficiency, reducing omissions, and improving the accuracy of coordinate information detection.

[0036] Optionally, in step S10, that is, performing text recognition on the form image to obtain a recognition result, includes:

[0037] S101, performing text recognition on the text information in the form image by using a text recognition algorithm to obtain the field information and the coordinate information;

[0038] S102, obtaining a template table corresponding to the plurality of field names;

[0039] S103, extracting the field information and the coordinate information through natural language technology, and filling the extracted field information and the coordinate information into the template table to obtain a recognition result.

[0040] It is understandable that the text information in the form image includes several field names and several field values ​​corresponding to the field names. The text recognition algorithm is used to perform text recognition on the text information in the form image, obtain the text information in the form image, and express it in the form of text to obtain field information; at the same time, obtain several position coordinates of several field values ​​in the field information in the form image to obtain coordinate information. According to the several field names recognized from the form image, the corresponding template form is obtained in the template form library. Natural language technology is used to extract the field names and field values ​​in the field information and the position coordinates in the coordinate information according to the context semantics, so that the extracted field names, field values ​​and position coordinates correspond to each other. After the field information and coordinate information are extracted, the extracted field information and coordinate information are filled into the template form according to the field name, and the field value and position coordinates corresponding to the field name are filled in to obtain the recognition result.

[0041] In steps S101-S103, the text information in the form image is recognized by a text recognition algorithm to obtain the field information and the coordinate information; a template table corresponding to the field names is obtained; the field information and the coordinate information are extracted by natural language technology, and the field information and the coordinate information are filled into the template table to obtain a recognition result. The present invention realizes text recognition of an image by combining a text recognition algorithm with natural language technology, which can improve the accuracy of text recognition, and the recognized field information and the coordinate information are filled into the template table, which can quickly test the accuracy of text recognition.

[0042] Optionally, in step S20, based on the field names, the non-emptiness of the field values ​​and the coordinate information is detected and analyzed to obtain the detection results, including:

[0043] S201, based on the field name, judging whether the field value corresponding to the field name and the coordinate information are non-empty, and obtaining a judgment result;

[0044] S202: According to the judgment results, the field names are matched to preset classification types respectively to obtain a detection result.

[0045] It is understandable that the detection and analysis of the non-emptiness of several field values ​​and coordinate information refers to judging the non-emptiness of each field value and each position coordinate corresponding to the field value, that is, judging the emptiness and non-emptiness of each field value and the emptiness and non-emptiness of each position coordinate corresponding to the field value. Among them, the emptiness of the field value means that the field name has no corresponding field value. The emptiness of the position coordinate means that the field name has no corresponding position coordinate. According to the judgment result, the field name is classified to obtain the classification type of the field name. According to the classification type, the field name and the field value, the detection result is generated. Preferably, the classification type includes redundant position coordinates, missing position coordinates, standard position coordinates and ignorable position coordinates. Based on the field name, when the field value is empty and the position coordinates are not empty, the position coordinates corresponding to the field name are determined to be redundant coordinates, and the field name is classified as redundant position coordinates; when the field value is not empty and the position coordinates are empty, the position coordinates corresponding to the field name are determined to be missing coordinates, and the field name is classified as missing position coordinates; when the field value is not empty and the position coordinates are not empty, the position coordinates corresponding to the field name are determined to be standard coordinates, and the field name is classified as standard position coordinates; when the field value is empty and the position coordinates are empty, the position coordinates corresponding to the field name are determined to be negligible coordinates, and the field name is classified as negligible position coordinates. When the classification type of the field name is standard position coordinates, it is necessary to manually further detect the accuracy of the field value and position coordinates corresponding to the field name.

[0046] In steps S201 and S202, based on the field name, the non-emptiness of the field value and the coordinate information corresponding to the field name is judged to obtain a judgment result; according to the judgment results, the field names are matched to the preset classification types to obtain a detection result. The field names are classified according to the judgment results to obtain a detection result. The detection result takes into account the non-emptiness of the field value and the position coordinates, as well as the classification of the field name, so that the detection result is more accurate, concise and clear, which is conducive to further manual verification.

[0047] Optionally, in step S201, based on the field name, the field value corresponding to the field name and the non-emptiness of the coordinate information are judged to obtain a judgment result, including:

[0048] S2011, obtaining the field value corresponding to the field name and the position coordinates corresponding to the field name;

[0049] S2012, judging whether the field value and the location coordinates are non-empty;

[0050] S2013. When the field value is empty and the position coordinates are not empty, determine that the position coordinates corresponding to the field name are redundant coordinates.

[0051] It can be understood that based on the field name, when the field value is empty and the position coordinates are not empty, the position coordinates corresponding to the field name are determined to be redundant coordinates, and the field name is classified as redundant position coordinates.

[0052] In steps S2011-S2013, the judgment result takes into account the situation when the field value is empty and the location coordinates are not empty. This can avoid the situation where, during manual verification, the location coordinates and field name corresponding to the field name are ignored due to the lack of field value of the field name in the recognition result, thereby improving the accuracy of detection.

[0053] Optionally, in step S30, that is, the detection result includes several classification types; adding a verification mark to the field value on the form image according to the detection result to obtain a verification image includes:

[0054] S301, obtaining a classification type corresponding to the field name;

[0055] S302: On the form image, add the verification mark to the field value corresponding to the field name based on the classification type.

[0056] It can be understood that the classification type is the type of field name obtained by classifying the field name according to the judgment result. The classification types include redundant position coordinates, missing position coordinates, standard position coordinates, and negligible position coordinates. The check mark is used to identify the position coordinates of the field value. Different check marks are set for different classification types. The difference in the check mark can be reflected by the color, shape, size, etc. of the check mark. Here, the shape, color, size, etc. of the check mark are not limited and can be set according to the actual needs of the user. For example, when the classification type of the field name is redundant position coordinates, a blue check mark can be added to the field value corresponding to the field name; when the classification type of the field name is missing position coordinates, a red check mark can be added to the field value corresponding to the field name.

[0057] In steps S301 and S302, by obtaining the classification type corresponding to the field name; on the form image, the verification mark is added to the field value corresponding to the field name based on the classification type. The present invention can add different verification marks to the field value corresponding to the field name according to the classification type of different field names, which can improve the detection efficiency and accuracy. Furthermore, the verification mark is added to the form image, so that when the position coordinates are further verified manually, the position coordinates can be quickly and accurately located, thereby improving the detection efficiency.

[0058] Optionally, in step S302, the adding of the verification mark to the field value corresponding to the field name based on the classification type includes:

[0059] S3021, obtaining the verification mark corresponding to the classification type;

[0060] S3022. Add the verification identifier to the field value corresponding to the field name, and generate a label number for the verification identifier according to the sorting sequence number of the verification identifier.

[0061] It is understandable that the form contains several field names, several field values ​​corresponding to the field names, and several position coordinates corresponding to the field values; preferably, the check mark can be sorted according to the arrangement order of the field names to obtain the sorting sequence number of the check mark. The sorting sequence number of the field name corresponds to the sorting sequence number of the check mark one by one. The marked number refers to the arrangement sequence number of the check mark. For example, the check mark can be sorted according to (1, 2, 3, ..., N). Preferably, the marked number can be marked below the check mark.

[0062] In steps S3021 and S3022, the verification mark corresponding to the classification type is obtained; the verification mark is added to the field value corresponding to the field name, and a label number is generated for the verification mark according to the sorting sequence number of the verification mark. By adding the verification mark to the field value and generating a label number for the verification mark, the present invention can quickly and accurately locate the verification mark according to the label number, and quickly perform verification according to the verification mark, thereby improving detection efficiency and accuracy.

[0063] Optionally, in step S40, the detection result includes a plurality of field names and a plurality of field values;

[0064] The step of setting a result display area in an adjacent area of ​​the verification image comprises:

[0065] S401, obtaining the first character lengths of several field names and the second character lengths of several field values;

[0066] S402, determining a maximum character length according to a number of the first character lengths and a number of the second character lengths;

[0067] S403: Dynamically adjust the width of the result display area according to the maximum character length.

[0068] It is understandable that the first character length refers to the length of all characters contained in a field name. The second character length refers to the length of all characters contained in a field value. The maximum character length refers to the character length corresponding to the character row with the longest character length among all character rows in the generated detection results. Generally, the character row with the longest character length consists of the field name and the field value and related characters. Therefore, the maximum character length can be determined based on the first character length of the field name and the second character length of the field value. Dynamically adjusting the width of the result display area according to the maximum character length can improve the adjustability of the result display area and at the same time improve the user experience.

[0069] It should be understood that the order of execution of the steps in the above embodiment does not necessarily mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiment of the present invention.

[0070] In one embodiment, a recognition result detection device based on text recognition is provided, and the recognition result detection device based on text recognition corresponds to the recognition result detection method based on text recognition in the above embodiment. Figure 3 As shown, the recognition result detection device based on text recognition includes a recognition result module 10, a detection result module 20, a verification image module 30 and a detection result image module 40. The detailed description of each functional module is as follows:

[0071] The recognition result module 10 is used to perform text recognition on the form image to obtain a recognition result; the recognition result includes field information and coordinate information; the field information includes a number of field names and field values ​​corresponding to the field names; the coordinate information includes the position coordinates of the field values;

[0072] A detection result module 20, configured to detect and analyze the non-emptiness of the field values ​​and the coordinate information based on the field names to obtain a detection result;

[0073] A verification image module 30, used to add a verification mark to the field value on the form image according to the detection result to obtain a verification image;

[0074] The detection result image module 40 is used to set a result display area in an adjacent area of ​​the verification image, display the detection result in the result display area, and obtain a detection result image.

[0075] Optionally, the recognition result module 10 includes:

[0076] A text recognition unit, used to perform text recognition on the text information in the form image by using a text recognition algorithm to obtain the field information and the coordinate information;

[0077] A template table acquisition unit is used to acquire a template table corresponding to the field names;

[0078] The information filling unit is used to extract the field information and the coordinate information through natural language technology, and fill the extracted field information and the coordinate information into the template table to obtain the recognition result.

[0079] Optionally, the detection result module 20 includes:

[0080] A judgment result unit, used for judging the non-emptiness of the field value and the coordinate information corresponding to the field name based on the field name, to obtain a judgment result;

[0081] The classification unit is used to match the plurality of field names to preset classification types according to the plurality of judgment results to obtain a detection result.

[0082] Optionally, the judgment result unit includes:

[0083] A judgment data acquisition unit, used to acquire a field value corresponding to the field name and a position coordinate corresponding to the field name;

[0084] A judging unit, used for judging whether the field value is non-empty and whether the position coordinates are non-empty;

[0085] The redundant coordinate determination unit is used to determine that the position coordinates corresponding to the field name are redundant coordinates when the field value is empty and the position coordinates are not empty.

[0086] Optionally, the verification image module 30 includes:

[0087] A classification type unit, used to obtain the classification type corresponding to the field name;

[0088] A verification mark adding unit is used to add the verification mark to the field value corresponding to the field name on the form image based on the classification type.

[0089] Optionally, the verification identification unit includes:

[0090] A verification mark acquisition unit, used to acquire the verification mark corresponding to the classification type;

[0091] The marking numbering unit is used to add the verification mark to the field value corresponding to the field name, and generate a marking number for the verification mark according to the sorting sequence number of the verification mark.

[0092] Optionally, the detection result image module 40 includes:

[0093] A character length obtaining unit, used for obtaining a first character length of a plurality of the field names and a second character length of a plurality of the field values;

[0094] a maximum character length determining unit, configured to determine a maximum character length according to a plurality of the first character lengths and a plurality of the second character lengths;

[0095] A dynamic adjustment unit is used to dynamically adjust the width of the result display area according to the maximum character length.

[0096] For the specific limitations of the recognition result detection device based on text recognition, please refer to the limitations of the recognition result detection method based on text recognition above, which will not be repeated here. Each module in the above-mentioned recognition result detection device based on text recognition can be implemented in whole or in part by software, hardware and a combination thereof. The above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory in the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

[0097] In one embodiment, a computer device is provided. The computer device may be a terminal, and its internal structure diagram may be as follows: Figure 4 As shown. The computer device includes a processor, a memory, a network interface, a display screen and an input device connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a readable storage medium and an internal memory. The non-volatile storage medium stores an operating system and computer-readable instructions. The internal memory provides an environment for the operation of the operating system and computer-readable instructions in the readable storage medium. The network interface of the computer device is used to communicate with an external server through a network connection. When the computer-readable instructions are executed by the processor, a recognition result detection method based on text recognition is implemented. The readable storage medium provided in this embodiment includes a non-volatile readable storage medium and a volatile readable storage medium.

[0098] In one embodiment, a computer device is provided, including a memory, a processor, and computer-readable instructions stored in the memory and executable on the processor, wherein when the processor executes the computer-readable instructions, the following steps are implemented:

[0099] Performing text recognition on the form image to obtain a recognition result; the recognition result includes field information and coordinate information; the field information includes a number of field names and field values ​​corresponding to the field names; the coordinate information includes the position coordinates of the field values;

[0100] Based on the field names, the non-emptiness of the field values ​​and the coordinate information is detected and analyzed to obtain a detection result;

[0101] On the form image, adding a verification mark to the field value according to the detection result to obtain a verification image;

[0102] A result display area is set in an adjacent area of ​​the verification image, and the detection result is displayed in the result display area to obtain a detection result image.

[0103] In one embodiment, one or more computer-readable storage media storing computer-readable instructions are provided. The readable storage media provided in this embodiment include non-volatile readable storage media and volatile readable storage media. The readable storage media store computer-readable instructions, and when the computer-readable instructions are executed by one or more processors, the following steps are implemented:

[0104] Performing text recognition on the form image to obtain a recognition result; the recognition result includes field information and coordinate information; the field information includes a number of field names and field values ​​corresponding to the field names; the coordinate information includes the position coordinates of the field values;

[0105] Based on the field names, the non-emptiness of the field values ​​and the coordinate information is detected and analyzed to obtain a detection result;

[0106] On the form image, adding a verification mark to the field value according to the detection result to obtain a verification image;

[0107] A result display area is set in an adjacent area of ​​the verification image, and the detection result is displayed in the result display area to obtain a detection result image.

[0108] Those of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be implemented by instructing related hardware through computer-readable instructions, and the computer-readable instructions can be stored in a non-volatile readable storage medium or a volatile readable storage medium. When the computer-readable instructions are executed, they may include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0109] Those skilled in the art can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example. In actual applications, the above-mentioned functions can be distributed and completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above.

[0110] The embodiments described above are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included in the protection scope of the present invention.

Claims

1. A recognition result detection method based on text recognition, characterized in that: include: Performing text recognition on the form image to obtain a recognition result; the recognition result includes field information and coordinate information; The field information includes a number of field names and field values ​​corresponding to the field names; the coordinate information includes the position coordinates of the field values; Based on the field names, performing detection and analysis on the non-emptiness of the field values ​​and the coordinate information to obtain a detection result; On the form image, adding a verification mark to the field value according to the detection result to obtain a verification image; Setting a result display area in an adjacent area of ​​the verification image, and displaying the detection result in the result display area to obtain a detection result image; Based on the field names, the non-emptiness of the field values ​​and the coordinate information is detected and analyzed to obtain the detection results, including: Based on the field name, judging whether the field value corresponding to the field name and the coordinate information are non-empty, and obtaining a judgment result; According to the judgment results, the field names are matched to the preset classification types to obtain the detection results; Based on the field name, the non-emptiness of the field value and the coordinate information corresponding to the field name is judged to obtain a judgment result, including: Get the field value corresponding to the field name and the position coordinates corresponding to the field name; Determine whether the field value is non-empty and whether the location coordinates are non-empty; When the field value is empty and the position coordinates are not empty, it is determined that the position coordinates corresponding to the field name are redundant coordinates.

2. The recognition result detection method based on character recognition according to claim 1, characterized in that: The step of performing text recognition on the form image to obtain a recognition result includes: Performing text recognition on the text information in the form image by using a text recognition algorithm to obtain the field information and the coordinate information; Obtaining a template table corresponding to the field names; The field information and the coordinate information are extracted by natural language technology, and the extracted field information and the coordinate information are filled into the template table to obtain a recognition result.

3. The recognition result detection method based on character recognition according to claim 1, characterized in that: The test results include several classification types; The step of adding a verification mark to the field value on the form image according to the detection result to obtain a verification image includes: Get the classification type corresponding to the field name; On the form image, the verification mark is added to the field value corresponding to the field name based on the classification type.

4. The recognition result detection method based on character recognition as claimed in claim 3, characterized in that: Adding the verification mark to the field value corresponding to the field name based on the classification type includes: Obtaining the verification identifier corresponding to the classification type; The verification identifier is added to the field value corresponding to the field name, and a label number is generated for the verification identifier according to the sorting sequence number of the verification identifier.

5. The recognition result detection method based on character recognition according to claim 1, characterized in that: The detection result includes several field names and several field values; the setting of the result display area in the adjacent area of ​​the verification image includes: Obtain the first character lengths of several of the field names and the second character lengths of several of the field values; Determine a maximum character length according to a plurality of the first character lengths and a plurality of the second character lengths; The width of the result display area is dynamically adjusted according to the maximum character length.

6. A recognition result detection device based on character recognition, characterized in that: include: The recognition result module is used to perform text recognition on the form image to obtain a recognition result; the recognition result includes field information and coordinate information; The field information includes a number of field names and field values ​​corresponding to the field names; the coordinate information includes the position coordinates of the field values; A detection result module, used for detecting and analyzing the non-emptiness of the field values ​​and coordinate information based on the field names to obtain a detection result; A verification image module, used to add a verification mark to the field value on the form image according to the detection result to obtain a verification image; A detection result image module, used to set a result display area in an adjacent area of ​​the verification image, display the detection result in the result display area, and obtain a detection result image; The detection result module comprises: Based on the field name, judging whether the field value corresponding to the field name and the coordinate information are non-empty, and obtaining a judgment result; According to the judgment results, the field names are matched to the preset classification types to obtain the detection results; Based on the field name, the non-emptiness of the field value and the coordinate information corresponding to the field name is judged to obtain a judgment result, including: Get the field value corresponding to the field name and the position coordinates corresponding to the field name; Determine whether the field value is non-empty and whether the location coordinates are non-empty; When the field value is empty and the position coordinates are not empty, it is determined that the position coordinates corresponding to the field name are redundant coordinates.

7. A computer device comprising a memory, a processor, and computer-readable instructions stored in the memory and executable on the processor, characterized in that: When the processor executes the computer-readable instructions, the recognition result detection method based on text recognition as described in any one of claims 1 to 5 is implemented.

8. One or more readable storage media storing computer-readable instructions, wherein when the computer-readable instructions are executed by one or more processors, the one or more processors execute the recognition result detection method based on text recognition as described in any one of claims 1 to 5.