CN110956140B

Movatterモバイル変換

Info

Publication number: CN110956140B
Application number: CN201911215794.2A
Authority: CN
Inventors: 韩冬; 雷继斌; 刘铭
Original assignee: Ant Shengxin Shanghai Information Technology Co ltd
Current assignee: Ant Shengxin Shanghai Information Technology Co ltd
Priority date: 2019-12-02
Filing date: 2019-12-02
Publication date: 2022-09-02
Anticipated expiration: 2039-12-02
Also published as: CN110956140A

Abstract

The method comprises the steps of receiving an image to be extracted of an auditor and extracting text information of the image to be extracted; matching the text information of the image to be extracted with a text information item specified by an auditor, and obtaining a matching result; determining missing positions of the missing text information items in the image to be extracted based on the matching result; and determining a preset image processing mode based on the requirements of the auditor, and extracting the missing text information of the missing position in the image to be extracted through the preset image processing mode.

Description

Image information extraction method, device, equipment and storage medium

Technical Field

The embodiment of the specification relates to the technical field of image recognition, in particular to an image information extraction method. One or more embodiments of the present specification also relate to an image information extraction apparatus, a computing device, and a computer-readable storage medium.

Background

The current OCR (Optical Character Recognition) technology cannot extract all information on an image at one hundred percent accurately, and especially when the image shooting quality is poor, situations such as blurring, shading, wrinkling and the like exist, and words in the image in a professional field exist, such as a large number of medical terms in medical diagnosis proofs, the extraction effect and accuracy are worse.

However, in some special scenarios, the extraction result of the OCR technology needs to be completely and accurately provided to a decision system as decision input, such as a small intelligent insurance claim scenario, and the claim settlement rule of the claim settlement system needs to automatically give a claim settlement result to reach the "second claim" effect after obtaining the OCR extraction result of the user certificate, so as to provide an ultimate claim settlement experience for the user, but the current OCR technology cannot achieve the "second claim" effect accurately and completely enough, and at this time, a more reliable image information extraction scheme needs to be provided.

Disclosure of Invention

In view of this, the present specification provides an image information extraction method. One or more embodiments of the present disclosure also relate to an image information extraction apparatus, a computing device, and a computer-readable storage medium to address technical deficiencies in the prior art.

According to a first aspect of embodiments herein, there is provided an image information extraction method including:

receiving an image to be extracted of an auditor, and extracting text information of the image to be extracted;

matching the text information of the image to be extracted with a text information item specified by an auditor, and obtaining a matching result;

determining the missing position of the missing text information item in the image to be extracted based on the matching result;

and determining a preset image processing mode based on the requirements of the auditor, and extracting the missing text information of the missing position in the image to be extracted through the preset image processing mode.

Optionally, before determining the preset image processing mode based on the requirement of the auditor, the method further includes:

judging whether the cost required for extracting the missing text information of the missing position in the image to be extracted is greater than or equal to a preset threshold value or not;

correspondingly, the determining a preset image processing mode based on the requirements of the auditor comprises the following steps:

and under the condition that the cost required for extracting the missing text information of the missing position in the image to be extracted is less than a preset threshold value, determining a preset image processing mode based on the requirement of the auditor.

Optionally, the requirement of the auditor includes a requirement of the auditor for extraction efficiency,

correspondingly, the determining a preset image processing mode based on the requirements of the auditor, and the extracting missing text information of the missing position in the image to be extracted by the preset image processing mode includes:

and selecting a preset first image processing mode based on the requirement of the auditor on the extraction efficiency, and extracting the missing text information of the missing position in the image to be extracted through the first image processing mode.

Optionally, after the missing text information of the missing position in the image to be extracted is extracted by the first image processing method, the method further includes:

combining the text information of the image to be extracted with the missing text information of the missing position in the image to be extracted to obtain combined text information;

matching the combined text information with a text information item specified by an auditor, and obtaining a matching result;

determining missing positions of the missing text information items in the image to be extracted based on the matching result;

and extracting the text information of the missing position in the image to be extracted based on a preset second image processing mode.

Optionally, the requirement of the auditor includes a requirement of the auditor for the extraction quality;

correspondingly, the determining a preset image processing mode based on the requirement of the auditor, and the extracting missing text information of the missing position in the image to be extracted by the preset image processing mode includes:

and selecting a preset second image processing mode based on the requirement of the auditor on the extraction efficiency, and extracting the missing text information of the missing position in the image to be extracted through the second image processing mode.

Optionally, the receiving an image to be extracted of an auditor, and extracting text information of the image to be extracted includes:

and receiving an image to be extracted of an auditor, and extracting text information of the image to be extracted in an optical character recognition mode.

Optionally, after receiving the image to be extracted of the auditor and before extracting the text information of the image to be extracted in a preset image processing manner, the method further includes:

and determining the text information item in the image to be extracted and the position of the text information item in the image to be extracted in the optical character recognition mode.

Optionally, the determining, based on the matching result, a missing position of the missing text information item in the image to be extracted includes:

determining the missing text information item based on the matching result;

and determining the missing position of the missing text information item in the image to be extracted based on the corresponding relation between the text information item in the image to be extracted and the text information item specified by the auditor.

Optionally, after determining a preset image processing manner based on the requirement of the auditor and extracting the missing text information of the missing position in the image to be extracted by the preset image processing manner, the method further includes:

combining the text information of the image to be extracted with the missing text information to obtain combined text information;

and matching the combined text information with the text information item specified by the auditor, and obtaining a matching result.

According to a second aspect of embodiments of the present specification, there is provided an image information extraction apparatus including:

the system comprises a text information extraction module, a verification module and a display module, wherein the text information extraction module is configured to receive an image to be extracted of an auditor and extract text information of the image to be extracted;

the matching result obtaining module is configured to match the text information of the image to be extracted with a text information item specified by an auditor and obtain a matching result;

a missing position determination module configured to determine a missing position of the missing text information item in the image to be extracted based on the matching result;

and the missing text information extraction module is configured to determine a preset image processing mode based on the requirement of the auditor and extract the missing text information of the missing position in the image to be extracted through the preset image processing mode.

Optionally, the apparatus further includes:

the judging module is configured to judge whether the cost required for extracting the missing text information of the missing position in the image to be extracted is greater than or equal to a preset threshold value;

correspondingly, the missing text information extraction module is further configured to:

correspondingly, the missing text information extraction module is further configured to include:

Optionally, the apparatus further includes:

the first combination module is configured to combine the text information of the image to be extracted with the missing text information of the missing position in the image to be extracted, and obtain combined text information;

the first matching module is configured to match the combined text information with a text information item specified by an auditor and obtain a matching result;

a determination module configured to determine a missing position of the missing text information item in the image to be extracted based on the matching result;

and the extraction module is configured to extract the text information of the missing position in the image to be extracted based on a preset second image processing mode.

Optionally, the text information extracting module is further configured to:

Optionally, the apparatus further includes:

the position extraction module is configured to determine a text information item in the image to be extracted and the position of the text information item through the optical character recognition mode.

Optionally, the missing position determination module is further configured to:

determining the missing text information item based on the matching result;

Optionally, the apparatus further includes:

the second combination module is configured to combine the text information of the image to be extracted with the missing text information to obtain combined text information;

and the second matching module is configured to match the combined text information with the text information item specified by the auditor and obtain a matching result.

According to a third aspect of embodiments herein, there is provided a computing device comprising:

a memory and a processor;

the memory is to store computer-executable instructions, and the processor is to execute the computer-executable instructions to:

According to a fourth aspect of embodiments herein, there is provided a computer-readable storage medium storing computer-executable instructions that, when executed by a processor, implement the steps of any one of the image information extraction methods.

One or more embodiments of the present specification provide an image information extraction method, which includes receiving an image to be extracted of an auditor, and extracting text information of the image to be extracted; matching the text information of the image to be extracted with a text information item specified by an auditor, and obtaining a matching result; determining the missing position of the missing text information item in the image to be extracted based on the matching result; determining a preset image processing mode based on the requirements of the auditor, and extracting missing text information of the missing position in the image to be extracted through the preset image processing mode;

the image information extraction method can automatically introduce a new image information extraction mode to assist in strengthening the extraction result on the basis of the common identification result according to the image information extraction scene of the user, and the progressive image information extraction method can enable the accuracy of image information extraction to be higher and can greatly reduce the cost consumption cost of information extraction.

Drawings

Fig. 1 is a flowchart of an image information extraction method provided in an embodiment of the present specification;

fig. 2 is a flowchart illustrating an application of an image information extraction method in extracting text information in a medical diagnosis certificate image according to an embodiment of the present specification;

fig. 3 is a schematic structural diagram of an image information extraction apparatus provided in an embodiment of the present specification;

fig. 4 is a block diagram of a computing device according to an embodiment of the present disclosure.

Detailed Description

In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present description. This description may be embodied in many different forms and should not be construed as limited to the embodiments set forth herein, as those skilled in the art will be able to make and use the present disclosure without departing from the spirit and scope of the present disclosure.

The terminology used in the description of the one or more embodiments is for the purpose of describing the particular embodiments only and is not intended to be limiting of the description of the one or more embodiments. As used in one or more embodiments of the present specification and the appended claims, the singular forms "a," "an," and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. It should also be understood that the term "and/or" as used in one or more embodiments of the present specification refers to and encompasses any and all possible combinations of one or more of the associated listed items.

It will be understood that, although the terms first, second, etc. may be used herein in one or more embodiments to describe various information, these information should not be limited by these terms. These terms are only used to distinguish one type of information from another. For example, a first can also be referred to as a second and, similarly, a second can also be referred to as a first without departing from the scope of one or more embodiments of the present description. The word "if" as used herein may be interpreted as "at … …" or "when … …" or "in response to a determination", depending on the context.

First, the noun terms to which one or more embodiments of the present specification relate are explained.

OCR: optical Character Recognition (OCR), which refers to a process of analyzing and recognizing an image file of text data to obtain text and layout information, and here refers in particular to a technology of recognizing characters in an image through a machine learning algorithm.

Extracting image information: the process of obtaining structured information from unstructured image information generally refers to character information presented in an image.

Crowdsourcing: crowdsourcing refers to the practice of a company or organization outsourcing work tasks performed by employees in the past to unspecified (and often large) public volunteers in a free-voluntary manner. The crowdsourcing mentioned in the embodiments of the present specification refers to crowdsourcing the task of extracting image information.

Outsourcing: compared with crowdsourcing, the method is also a mode of work task collaboration, and the difference is that the outsourcing needs to employ full-time staff, so that high specialization is emphasized.

Progressive: the result of image information extraction will be different based on different technical means, if the result of a simpler technical means extraction cannot meet the requirements, a stronger technical means is required for extraction, and certainly there will be more cost overhead, and the process of obtaining a better result of image information extraction by continuously improving the means of image information extraction is called "progressive".

In the present specification, an image information extraction method is provided, and the present specification relates to an image information extraction apparatus, a computing device, and a computer-readable storage medium, which are described in detail one by one in the following embodiments.

Before each embodiment of the present invention is described, an application scenario of the embodiment of the present invention is described, and the image information extraction method of the embodiment of the present invention may be applied to any scenario that needs to extract characters in an image, for example, a claim settlement scenario of an insurance organization, where an audit system of the insurance organization receives a medical diagnosis certification image provided by a user participating in insurance, transmits the image to an image recognition system of the insurance organization, and extracts a claim settlement key field in the image by using the image information extraction method applied to the image recognition system so as to perform claim settlement calculation and the like.

Referring to fig. 1, fig. 1 shows a flowchart of an image information extraction method provided in accordance with an embodiment of the present specification, which includessteps 102 to 108.

Step 102: and receiving an image to be extracted of an auditor, and extracting text information of the image to be extracted.

The auditor is configured according to an actual application scenario of the image information extraction method, for example, when the image information extraction method is applied to an insurance claim settlement scenario, the auditor may be understood as an insurance claim settlement party, and the setting scenario of the auditor includes: after receiving a request of an item to be claimed of a user to be claimed, the authenticity verification is carried out according to claim materials uploaded by the user to be claimed, corresponding claim funds are given under the condition that the claim materials uploaded by the user to be claimed are real, and the claim materials can be images containing information to be claimed at the moment, such as images of diagnosis certificates, discharge knots and the like of hospitals.

The image information extraction method is applied to an insurance claim settlement scene, an image recognition system of an insurance agency firstly receives an image to be extracted input by an auditor of an audit system of the insurance agency, and then text information of the image to be extracted is extracted.

Specifically, the image to be extracted includes, but is not limited to, an image containing a text description of an event to be claimed, such as an image of a vehicle injury description corresponding to a vehicle insurance, a hospital diagnosis certificate corresponding to a medical insurance, an image of a discharge summary, and the like; the text information of the image to be extracted includes but is not limited to text description contents of the time to be claimed, such as text description of vehicle injury, hospital diagnosis certification, text description on discharge summary, and the like.

In one or more embodiments of the present specification, the receiving an image to be extracted by an auditor, and extracting text information of the image to be extracted includes:

The Optical Character Recognition includes an OCR (Optical Character Recognition) technology, which can be understood as a machine learning image processing manner of a "positioning + Recognition" mode. In practical application, an image is input, and characters to be extracted and position coordinates of the texts to be extracted in the image can be output by adopting an OCR technology. For example, a medical invoice is entered, and the total amount (e.g., 33.10 dollars) of the invoice, the location coordinates of the total amount (e.g., the rectangular box where the total amount field is located) may be returned using OCR techniques.

Specifically, an image to be extracted of the auditor is received, and text information of the image to be extracted is extracted in an OCR recognition mode, for example, all recognizable text fields of the image to be extracted. The extraction cost of extracting the text information of the image to be extracted in an OCR recognition mode is low, and the execution efficiency is high.

In addition, in another embodiment of this specification, after receiving an image to be extracted by an auditor and before extracting text information of the image to be extracted by using a preset image processing manner, the method further includes:

The text information item comprises the attribute of the text information, for example, if the text information is 33 yuan, the text information item corresponding to the text information is the amount; and if the text information is reddish, the text information item corresponding to the text information is a name and the like.

Specifically, after an image to be extracted of an auditor is received, firstly, all text information items in the image to be extracted and the coordinate position of each text information item are extracted in an OCR (optical character recognition) mode, and then, the text information of the image to be extracted is extracted in the OCR mode.

Step 104: and matching the text information of the image to be extracted with a text information item specified by an auditor, and obtaining a matching result.

The text information items specified by the auditor include but are not limited to audit text information items predefined by the auditor, such as text information items of name, gender, year and month of birth, and the like.

Taking the image to be extracted as an image for medical diagnosis certification as an example, matching the text information of the image to be extracted with a text information item specified by an auditor, and obtaining a matching result to explain in detail.

If the text information extracted from the image of the medical diagnosis certificate comprises the following categories: internal medicine and name: small a, sex: female, age: 18. admission date, 1 month and 1 day in 2019, hospital number: 00000. the date of discharge from hospital is 2 months and 3 days in 2019, and the date of clinic visit: 26 months 12 and 2018, responsible physician: b is small;

the text information items specified by the auditor comprise the department, the name, the sex, the age, the admission date, the hospitalization number, the discharge date, the clinic visit date, the diagnosis opinions, the suggestions and the responsible physicians; then, after the extracted text information of the image of the medical diagnosis proof is in one-to-one correspondence with the text information item specified by the auditor, the diagnosis suggestion of the text information item specified by the auditor is found, and the two text information items are suggested to have no corresponding text information, which indicates that the extracted text information of the image of the medical diagnosis proof is not matched with the text information item specified by the auditor, and the extracted text information in the image of the medical diagnosis proof lacks the diagnosis suggestion and suggests the two text information, and the matching result at this time can include the text information extracted from the image of the medical diagnosis proof, which is not matched with the text information item specified by the auditor, and the text information items which are not specifically matched: a diagnostic opinion textual information item and a suggestion textual information item.

In another case, the extracted text information in the image of the medical diagnosis certificate includes text information of department, name, gender, age, date of admission, number of admission, date of discharge, date of clinic visit, diagnosis opinions, advice, responsible physicians, and the like, and corresponds to the text information items specified by the auditor one to one, in which case, the text information of the image to be extracted matches the text information items specified by the auditor, and the matching result may be the ending of the image extraction method, and the insurance institution may perform the settlement of the claim money based on the extracted text information of the medical diagnosis certificate.

Step 106: determining a missing position of the missing text information item in the image to be extracted based on the matching result.

Still taking the above as an example, based on the matching result, it may be determined that the text information of the image to be extracted does not match the text information specified by the reviewer, that is, the missing text information item.

Specifically, the determining, based on the matching result, a missing position of the missing text information item in the image to be extracted includes:

determining the missing text information item based on the matching result;

After the missing text information items are determined based on the text information items specified by the auditor, the missing positions of the missing text information items in the image to be extracted can be accurately and quickly determined based on the corresponding relation between the text information items in the image to be extracted and the text information items specified by the auditor.

Step 108: and determining a preset image processing mode based on the requirements of the auditor, and extracting the missing text information of the missing position in the image to be extracted through the preset image processing mode.

Wherein, the requirements of the auditor include but are not limited to the requirements of the auditor on the extraction efficiency or the requirements of the auditor on the extraction quality; the requirement of the auditor on the extraction efficiency can be understood as a requirement of the auditor on the extraction efficiency of the text information in the image to be extracted, and the requirement of the auditor on the extraction efficiency can be understood as a requirement of the auditor on the extraction quality of the text information in the image to be extracted.

The preset image processing mode includes, but is not limited to, a preset first image processing mode and a preset second image processing mode.

In one or more embodiments of the present specification, the requirements of the auditor include requirements of the auditor for extraction efficiency,

Specifically, the preset first image processing mode includes a crowdsourcing extraction image processing mode, and the first image processing mode is to split the image to be extracted into a small image to be extracted including each text information item based on the coordinate position of each text information item of the image to be extracted by the OCR recognition mode on the basis of the image to be extracted by the OCR recognition mode, and then recognize the small image to be extracted including each text information item.

In the embodiment of the description, under the condition that text information in an image to be extracted is extracted in a missing manner by adopting an OCR recognition mode, the missing text information is extracted for the second time by adopting a first image processing mode, and the extraction result of the text information in the image to be extracted is enhanced by adopting the progressive recognition means, so that the accuracy and the integrity of the extraction of the text information in the image to be extracted are greatly improved.

In one or more embodiments of the present description, the requirements of the auditor include requirements of the auditor for extraction quality;

Specifically, the preset second image processing mode includes an outsourced extracted image processing mode, and the second image processing mode is to perform more professional and accurate examination on the missing text information of the missing position in the image to be extracted on the basis of the OCR recognition mode to-be-extracted image recognition.

In practical application, the outsourcing extraction image processing mode is to extract the missing text information at the missing position in the image to be extracted after reasonably analyzing the text information in the image to be extracted according to long-term work experience and the context information of a single image by an image text information extraction worker who carries out professional training and receives security inspection.

The quality of the missing text information of the missing position in the image to be extracted, which is extracted by the second image processing mode, is high, and the accuracy can basically meet the requirements of practical application scenes.

In one or more embodiments of the present specification, after the extracting missing text information of a missing position in the image to be extracted by the first image processing method, the method further includes:

In practical application, the missing text information at the missing position in the image to be extracted is extracted through the first image processing mode, then the text information of the image to be extracted and the missing text information at the missing position in the image to be extracted are combined to obtain combined text information, then the combined text information is matched with the text information item specified by the auditing party again, if the missing text information item still exists in the matching result, the missing position of the missing text information item in the image to be extracted is determined, and then the text information at the missing position in the image to be extracted is extracted through the second image processing mode.

Taking the example that the missing text information item includes a diagnosis opinion text information item and a suggested text information item, if the missing text information at the missing position in the image to be extracted is extracted again by the first image processing method, the extracted missing text information includes diagnosis opinion text information, and at this time, the diagnosis opinion text information is combined with the original text information in the image to be extracted, which is extracted by the OCR recognition method, to obtain combined text information: the method comprises the following steps: internal medicine and name: small a, sex: female, age: 18. admission date, 1 month and 1 day in 2019, hospital number: 00000. the date of discharge from hospital is 2 months and 3 days in 2019, and the date of clinic visit: 26 days 12 and 2018, diagnosis opinion: charge physician: and a sub-b, matching the combined text information with the text information item specified by the auditor to obtain a matching result: missing suggested text information items, and extracting the text information of the missing positions of the missing suggested text information items in the image to be extracted again based on a second image processing mode; by adopting the progressive image extraction mode, a new image information extraction mode can be automatically introduced on the basis of the extraction result of the common OCR recognition mode according to the individual scene of the user to assist in strengthening the recognition result, the text information extraction cost is consumed as low as possible, and the completeness and accuracy of extracting the text information of the image to be extracted can be improved.

In one or more embodiments of the present specification, before determining the preset image processing manner based on the requirement of the auditor, the method further includes:

correspondingly, the determining a preset image processing mode based on the requirement of the auditor includes:

In practical application, after extracting text information in an image to be extracted in an OCR (optical character recognition) mode, if the missing text information exists, judging whether the cost required for extracting the missing text information at the missing position in the image to be extracted is greater than or equal to a preset threshold value, if so, ending the extraction process of the image to be extracted, and if not, determining a preset image processing mode based on the requirements of an auditor; by adopting the method, the cost of extracting the text information in the image to be extracted is avoided being increased, and the system operation fund is saved.

In practical application, the image information extraction method is applied to an insurance claim settlement scene, and if the extraction cost of the text information in the image is high, the operation cost of an insurance agency is correspondingly increased, so that the cost of the image extraction is considered in practical application to realize throttling.

In one or more embodiments of the present specification, after determining a preset image processing manner based on a requirement of the auditor, and extracting missing text information of a missing position in the image to be extracted by using the preset image processing manner, the method further includes:

In actual use, if there are missing text information items in the matching result of matching the text information of the image to be extracted with the text information item specified by the auditor, the method may be used to extract and combine the texts in the missing text information items and match the texts with the text information item specified by the auditor again until the extracted text information of the image to be extracted is successfully matched with the text information item specified by the auditor, and the image extraction method is ended without missing text information items.

In the embodiment of the description, the image information extraction method can make an intelligent decision among extraction efficiency, cost and accuracy on the premise of meeting the extraction result of the text information in the image to be extracted, and give an optimal decision result, namely, a new image information extraction mode with efficiency or accuracy as a standard is automatically selected on the basis of a common OCR recognition result according to the image information extraction scene of a user to assist in strengthening the extraction result.

Referring to fig. 2, fig. 2 shows a flowchart for applying an image information extraction method provided in an embodiment of the present specification to extracting text information in a medical diagnosis certificate image, includingsteps 202 to 218.

Step 202: and receiving a medical diagnosis certification image input by an auditor, and extracting text information of the medical diagnosis certification image in an OCR (optical character recognition) mode.

Step 204: matching the text information of the medical diagnosis certification image with a text information item specified by an auditor, and obtaining a matching result;

step 206: and judging whether the matching result contains the missing text information item, if so, executingstep 208, and if not, executingstep 210.

Step 208: determining a missing position of the missing text information item in the medical diagnosis certification image based on the matching result.

Step 210: and (6) ending.

Step 212: judging whether the cost required for extracting the missing text information of the missing position in the medical diagnosis certification image is greater than or equal to a preset threshold value or not; if yes, go to step 210, if no, go to step 214.

Step 214: and judging whether the requirement of the auditor is the requirement of the auditor on the extraction efficiency, if so, executingstep 216, and if not, executingstep 218.

Step 216: and extracting missing text information of the missing position in the medical diagnosis certification image by the first image processing mode.

Specifically, afterstep 216 is executed, the extracted missing text information of the missing position in the medical diagnosis certification image may be combined with the text information of the medical diagnosis certification image extracted instep 202 by the OCR recognition method, and then step 204 and the subsequent steps are continuously executed until the extracted text information of the medical diagnosis certification image matches the text information item specified by the auditor and the process is finished when there is no missing text information item.

Step 218: and extracting text information of the missing position in the medical diagnosis certification image by the second image processing mode.

In practical applications, after the text information at the missing position in the medical diagnosis proof image is extracted by the second image processing method, the text information needs to be combined with the text information extracted by the OCR recognition method and the first image processing method, then the text information is re-matched with the text information item specified by the auditing party, and the following processes after matching are executed until the text information of the medical diagnosis proof image circularly extracted by the three methods is matched with the text information item specified by the auditing party, and the image information extraction method is ended under the condition that no missing text information item exists.

In the embodiment of the description, the image information extraction method can perform intelligent decision among extraction efficiency, cost and accuracy on the premise of meeting the extraction result of the text information in the medical diagnosis certification image, give an optimal decision result, and automatically select a first image processing mode or a second image processing mode which takes efficiency or accuracy as a standard on the basis of a common OCR recognition result according to the image information extraction scene of a user to assist in strengthening the extraction result.

Corresponding to the above method embodiment, the present specification further provides an image information extraction apparatus embodiment, and fig. 3 shows a schematic structural diagram of an image information extraction apparatus provided in an embodiment of the present specification. As shown in fig. 3, the apparatus includes:

a text information extraction module 302 configured to receive an image to be extracted of an auditor and extract text information of the image to be extracted;

a matching result obtaining module 304 configured to match the text information of the image to be extracted with a text information item specified by an auditor and obtain a matching result;

a missing position determination module 306 configured to determine a missing position of the missing text information item in the image to be extracted based on the matching result;

the missing text information extraction module 308 is configured to determine a preset image processing mode based on the requirement of the auditor, and extract the missing text information of the missing position in the image to be extracted by the preset image processing mode.

Optionally, the apparatus further includes:

accordingly, the missing text information extraction module 308 is further configured to:

accordingly, the missing text information extraction module 308 is further configured to include:

Optionally, the apparatus further includes:

the first matching module is configured to match the combined text information with text information items specified by an auditor and obtain a matching result;

Optionally, the text information extracting module 302 is further configured to:

Optionally, the apparatus further includes:

Optionally, the missing position determining module 306 is further configured to:

determining the missing text information item based on the matching result;

Optionally, the apparatus further includes:

The image information extraction device provided by the embodiment of the description can automatically introduce a new image information extraction mode to assist in strengthening the extraction result on the basis of the common identification result according to the image information extraction scene of the user, and the progressive image information extraction method can enable the accuracy of image information extraction to be higher and can greatly reduce the cost consumption cost of information extraction.

The above is a schematic configuration of an image information extraction apparatus of the present embodiment. It should be noted that the technical solution of the image information extraction device and the technical solution of the image information extraction method belong to the same concept, and details that are not described in detail in the technical solution of the image information extraction device can be referred to the description of the technical solution of the image information extraction method.

FIG. 4 illustrates a block diagram of a computing device 400 provided in accordance with one embodiment of the present description. The components of the computing device 400 include, but are not limited to, a memory 410 and a processor 420. Processor 420 is coupled to memory 410 via bus 430 and database 450 is used to store data.

Computing device 400 also includes access device 440, access device 440 enabling computing device 400 to communicate via one or more networks 460. Examples of such networks include the Public Switched Telephone Network (PSTN), a Local Area Network (LAN), a Wide Area Network (WAN), a Personal Area Network (PAN), or a combination of communication networks such as the internet. The access device 440 may include one or more of any type of network interface (e.g., a Network Interface Card (NIC)) whether wired or wireless, such as an IEEE802.11 Wireless Local Area Network (WLAN) wireless interface, a worldwide interoperability for microwave access (Wi-MAX) interface, an ethernet interface, a Universal Serial Bus (USB) interface, a cellular network interface, a bluetooth interface, a Near Field Communication (NFC) interface, and so forth.

In one embodiment of the present description, the above-described components of computing device 400, as well as other components not shown in FIG. 4, may also be connected to each other, such as by a bus. It should be understood that the block diagram of the computing device structure shown in FIG. 4 is for purposes of example only and is not limiting as to the scope of the description. Those skilled in the art may add or replace other components as desired.

Computing device 400 may be any type of stationary or mobile computing device, including a mobile computer or mobile computing device (e.g., tablet computer, personal digital assistant, laptop computer, notebook computer, netbook, etc.), mobile phone (e.g., smartphone), wearable computing device (e.g., smartwatch, smart glasses, etc.), or other type of mobile device, or a stationary computing device such as a desktop computer or PC. Computing device 400 may also be a mobile or stationary server.

Wherein processor 420 is configured to execute the following computer-executable instructions:

determining the missing text information item based on the matching result;

Optionally, after determining a preset image processing mode based on the requirement of the auditor and extracting the missing text information of the missing position in the image to be extracted by the preset image processing mode, the method further includes:

The above is an illustrative scheme of a computing device of the present embodiment. It should be noted that the technical solution of the computing device and the technical solution of the image information extraction method belong to the same concept, and details that are not described in detail in the technical solution of the computing device can be referred to the description of the technical solution of the image information extraction method.

An embodiment of the present specification also provides a computer readable storage medium storing computer instructions that, when executed by a processor, are operable to:

matching the combined text information with text information items specified by an auditor, and obtaining a matching result;

and extracting text information of the missing position in the image to be extracted based on a preset second image processing mode.

determining the missing text information item based on the matching result;

The above is an illustrative scheme of a computer-readable storage medium of the embodiment. It should be noted that the technical solution of the storage medium belongs to the same concept as the technical solution of the image information extraction method, and details that are not described in detail in the technical solution of the storage medium can be referred to the description of the technical solution of the image information extraction method.

The foregoing description has been directed to specific embodiments of this disclosure. Other embodiments are within the scope of the following claims. In some cases, the actions or steps recited in the claims may be performed in a different order than in the embodiments and still achieve desirable results. In addition, the processes depicted in the accompanying figures do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In some embodiments, multitasking and parallel processing may also be possible or may be advantageous.

The computer instructions comprise computer program code which may be in the form of source code, object code, an executable file or some intermediate form, or the like. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, usb disk, removable hard disk, magnetic disk, optical disk, computer Memory, Read-Only Memory (ROM), Random Access Memory (RAM), electrical carrier wave signals, telecommunications signals, software distribution medium, and the like. It should be noted that the computer readable medium may contain content that is subject to appropriate increase or decrease as required by legislation and patent practice in jurisdictions, for example, in some jurisdictions, computer readable media does not include electrical carrier signals and telecommunications signals as is required by legislation and patent practice.

It should be noted that, for the sake of simplicity, the foregoing method embodiments are described as a series of combinations of acts, but it should be understood by those skilled in the art that the embodiments are not limited by the described order of acts, as some steps may be performed in other orders or simultaneously according to the embodiments. Further, those skilled in the art should also appreciate that the embodiments described in this specification are preferred embodiments and that acts and modules referred to are not necessarily required for an embodiment of the specification.

In the above embodiments, the descriptions of the respective embodiments have respective emphasis, and for parts that are not described in detail in a certain embodiment, reference may be made to related descriptions of other embodiments.

The preferred embodiments of the present specification disclosed above are intended only to aid in the description of the specification. Alternative embodiments are not exhaustive and do not limit the invention to the precise embodiments described. Obviously, many modifications and variations are possible in light of the above teaching. The embodiments were chosen and described in order to best explain the principles of the embodiments and the practical application, to thereby enable others skilled in the art to best understand and utilize the embodiments. The specification is limited only by the claims and their full scope and equivalents.

Claims

1. An image information extraction method, comprising:

determining missing text information items and missing positions of the missing text information items in the image to be extracted based on the matching result;

selecting a preset first image processing mode based on the requirement of the auditor on the extraction efficiency, and extracting the missing text information of the missing position in the image to be extracted through the first image processing mode;

after the missing text information of the actual position in the image to be extracted is extracted through the first image processing mode, the method further comprises the steps of combining the text information of the image to be extracted with the missing text information of the missing position in the image to be extracted to obtain combined text information, matching the combined text information with a text information item specified by an auditor to obtain a matching result, determining the missing position of the missing text information item in the image to be extracted based on the matching result, and extracting the text information of the missing position in the image to be extracted based on a preset second image processing mode.

2. The image information extraction method according to claim 1, before determining the preset image processing mode based on the requirement of the auditor, further comprising:

3. The image information extraction method according to claim 1, wherein the auditor's requirements include the auditor's requirements for extraction quality;

4. The image information extraction method according to claim 1, wherein the receiving of the image to be extracted of the auditor and the extracting of the text information of the image to be extracted include:

5. The image information extraction method according to claim 4, further comprising, after receiving the image to be extracted of the auditor and before extracting the text information of the image to be extracted by a preset image processing manner:

6. The image information extraction method according to claim 5, said determining, based on the matching result, a missing position of the missing text information item in the image to be extracted including:

determining the missing text information item based on the matching result;

7. The image information extraction method according to claim 1, wherein after determining a preset image processing mode based on a requirement of the auditor and extracting missing text information of a missing position in the image to be extracted by the preset image processing mode, the method further comprises:

and matching the combined text information with text information items specified by an auditor, and obtaining a matching result.

8. An image information extraction apparatus comprising:

a missing position determination module configured to determine missing text information items and missing positions of the missing text information items in the image to be extracted based on the matching result;

the missing text information extraction module is configured to select a preset first image processing mode based on the requirement of the auditor on the extraction efficiency, and extract the missing text information of the missing position in the image to be extracted through the first image processing mode;

9. The image information extraction apparatus according to claim 8, further comprising:

10. A computing device, comprising:

a memory and a processor;

11. A computer-readable storage medium storing computer instructions which, when executed by a processor, implement the steps of the image information extraction method of any one of claims 1 to 7.