Car insurance certificate image auditing method, device and equipment and storage medium

Abstract

The invention belongs to the field of artificial intelligence and insurance, and relates to a car insurance certificate image auditing method, which comprises the steps of obtaining a to-be-audited car insurance certificate image, and performing type identification on the car insurance certificate image by adopting a certificate type identification model to obtain a certificate type of the car insurance certificate image; according to the certificate type, carrying out content identification on the vehicle insurance certificate image by adopting a certificate content identification model, and extracting certificate information of the vehicle insurance certificate image; acquiring insurance information corresponding to the to-be-audited vehicle insurance certificate image, and performing consistency verification on the certificate information and the insurance information; and if the consistency verification is passed, judging that the vehicle insurance certificate image passes automatic auditing, and determining that a certificate corresponding to the vehicle insurance certificate image is a compliant certificate. The invention further provides a vehicle insurance certificate image auditing device and equipment and a storage medium. In addition, the invention also relates to a block chain technology, and certificate information and the like can be stored in a block chain. The auditing efficiency of the vehicle insurance certificate image can be improved.

Description

Technical Field

[0001] The present application relates to the fields of artificial intelligence and insurance, and in particular to a method for reviewing images of motor vehicle insurance certificates. Background Art

[0002] In the insurance field, verifying the documents uploaded by customers is one of the core links. The traditional way of reviewing auto insurance documents is done manually. The specific method is to verify the document type, the legality of the document, the consistency between the document and the insured information, and the validity of the document.

[0003] However, the traditional manual review method is time-consuming and error-prone, making it difficult to cope with a large number of insurance applications. In addition, the consistency verification between the certificate information and the insurance information is complicated and it is easy to miss important details, making the auto insurance application process cumbersome, time-consuming and labor-intensive, affecting the insurance company's operational efficiency and customer satisfaction.

[0004] Therefore, there is an urgent need for an automatic review method for vehicle insurance certificate images to improve the efficiency of vehicle insurance certificate review. Summary of the invention

[0005] The purpose of the embodiments of the present application is to propose a method, device, equipment and storage medium for reviewing vehicle insurance certificate images, the main purpose of which is to improve the efficiency of vehicle insurance certificate review.

[0006] In order to solve the above technical problems, the present application embodiment provides a method for reviewing a motor vehicle insurance certificate image, which adopts the following technical solution:

[0007] Acquire a motor vehicle insurance certificate image to be reviewed, and use a pre-trained certificate type recognition model to perform type recognition on the motor vehicle insurance certificate image to be reviewed to obtain the certificate type of the motor vehicle insurance certificate image;

[0008] According to the certificate type, a pre-trained certificate content recognition model is used to perform content recognition on the motor vehicle insurance certificate image to be reviewed, and the certificate information of the motor vehicle insurance certificate image to be reviewed is extracted;

[0009] Acquire the insurance information corresponding to the vehicle insurance certificate image to be reviewed, and perform consistency check between the certificate information and the insurance information;

[0010] If the consistency check passes, it is determined that the vehicle insurance certificate image to be reviewed has passed the automatic review, and it is determined that the certificate corresponding to the vehicle insurance certificate image to be reviewed is a compliant certificate.

[0011] Furthermore, the method of using a pre-trained certificate type recognition model to perform type recognition on the vehicle insurance certificate image to be reviewed to obtain the certificate type of the vehicle insurance certificate image includes:

[0012] Using the convolution layer in the pre-trained certificate type recognition model to perform a convolution operation on the image of the motor vehicle insurance certificate to be reviewed, and extracting target type features of the image of the motor vehicle insurance certificate to be reviewed;

[0013] Using the pooling layer in the pre-trained document type recognition model to perform a dimensionality reduction operation on the target type feature to obtain a target type feature after dimensionality reduction;

[0014] The output layer in the pre-trained certificate type recognition model is used to calculate the probability of the target type feature after dimensionality reduction belonging to each type and the type with the largest probability is selected to obtain the certificate type of the vehicle insurance certificate image.

[0015] Furthermore, according to the certificate type, a pre-trained certificate content recognition model is used to perform content recognition on the image of the motor vehicle insurance certificate to be reviewed, and certificate information of the image of the motor vehicle insurance certificate to be reviewed is extracted, including:

[0016] According to the document type, using the text detection module of the pre-trained document content recognition model, the key content position detection is performed on the image of the automobile insurance document to be reviewed to obtain the key content area;

[0017] Using the text recognition module of the pre-trained document content recognition model, performing OCR processing on the key content area to extract key text from the key content area;

[0018] The structured processing module using the pre-trained certificate content recognition model is used to convert the key text into a format according to a predefined data format to obtain the certificate information of the vehicle insurance certificate image to be reviewed.

[0019] Furthermore, after performing OCR processing on the key content area by using the text recognition module of the pre-trained document content recognition model to extract key text from the key content area, the method further includes:

[0020] Using the post-processing module of the pre-trained document content recognition model to perform field verification on the key text to query whether the predefined required fields have been filled in;

[0021] When all the predefined mandatory fields are filled in, all the predefined mandatory fields are verified one by one with the text corresponding to the image of the automobile insurance certificate to be reviewed;

[0022] The verified key text is input into the structured processing module using the pre-trained document content recognition model.

[0023] Furthermore, after extracting the certificate information of the vehicle insurance certificate image to be reviewed, the method further includes:

[0024] Acquire the current time, and verify the validity period of the vehicle insurance certificate image to be reviewed through the certificate information and the current time;

[0025] When the current time is not within the validity period of the vehicle insurance certificate image to be reviewed, it is determined that the vehicle insurance certificate image to be reviewed is not compliant, and the user is prompted to re-upload another type of certificate image.

[0026] Furthermore, the consistency check between the certificate information and the insurance information includes:

[0027] Structuring the insurance information according to a predefined structure template to obtain structured insurance information;

[0028] Perform a consistency check between the certificate information and the information with the same prefix in the structure insurance information.

[0029] Furthermore, after determining that the certificate corresponding to the to-be-examined vehicle insurance certificate image is a compliant certificate, the method further includes:

[0030] Storing the certificate information of the compliance certificate in a preset certificate information database;

[0031] When receiving a policy renewal request, obtaining the certificate information corresponding to the policy renewal request from the preset certificate information database;

[0032] Determine whether the certificate information corresponding to the renewal request is within the validity period;

[0033] When the certificate information corresponding to the policy renewal request is within the validity period, the certificate information corresponding to the policy renewal request is called to complete the policy renewal operation.

[0034] In order to solve the above technical problems, the embodiment of the present application further provides a device for reviewing an image of a motor vehicle insurance certificate, which adopts the following technical solution:

[0035] A type recognition module is used to obtain the image of the motor vehicle insurance certificate to be reviewed, and use a pre-trained certificate type recognition model to perform type recognition on the image of the motor vehicle insurance certificate to be reviewed to obtain the certificate type of the image of the motor vehicle insurance certificate;

[0036] A content extraction module, used to perform content recognition on the image of the motor vehicle insurance certificate to be reviewed using a pre-trained certificate content recognition model according to the certificate type, and extract certificate information of the image of the motor vehicle insurance certificate to be reviewed;

[0037] An information verification module, used to obtain the insurance information corresponding to the automobile insurance certificate image to be reviewed, and to perform consistency verification between the certificate information and the insurance information;

[0038] The compliance judgment module is used to judge whether the vehicle insurance certificate image to be reviewed has passed the automatic review if the consistency check passes, and determine that the certificate corresponding to the vehicle insurance certificate image to be reviewed is a compliant certificate.

[0039] In order to solve the above technical problems, the embodiment of the present application further provides a device, which adopts the following technical solution:

[0040] Acquire a motor vehicle insurance certificate image to be reviewed, and use a pre-trained certificate type recognition model to perform type recognition on the motor vehicle insurance certificate image to be reviewed to obtain the certificate type of the motor vehicle insurance certificate image;

[0041] According to the certificate type, a pre-trained certificate content recognition model is used to perform content recognition on the motor vehicle insurance certificate image to be reviewed, and the certificate information of the motor vehicle insurance certificate image to be reviewed is extracted;

[0042] Acquire the insurance information corresponding to the vehicle insurance certificate image to be reviewed, and perform consistency check between the certificate information and the insurance information;

[0043] If the consistency check passes, it is determined that the vehicle insurance certificate image to be reviewed has passed the automatic review, and it is determined that the certificate corresponding to the vehicle insurance certificate image to be reviewed is a compliant certificate.

[0044] In order to solve the above technical problems, the embodiment of the present application further provides a storage medium, which adopts the following technical solution:

[0045] Acquire a motor vehicle insurance certificate image to be reviewed, and use a pre-trained certificate type recognition model to perform type recognition on the motor vehicle insurance certificate image to be reviewed to obtain the certificate type of the motor vehicle insurance certificate image;

[0046] According to the certificate type, a pre-trained certificate content recognition model is used to perform content recognition on the motor vehicle insurance certificate image to be reviewed, and the certificate information of the motor vehicle insurance certificate image to be reviewed is extracted;

[0047] Acquire the insurance information corresponding to the vehicle insurance certificate image to be reviewed, and perform consistency check between the certificate information and the insurance information;

[0048] If the consistency check passes, it is determined that the vehicle insurance certificate image to be reviewed has passed the automatic review, and it is determined that the certificate corresponding to the vehicle insurance certificate image to be reviewed is a compliant certificate.

[0049] Compared with the prior art, the embodiments of the present application have the following beneficial effects:

[0050] The present application uses a pre-trained document type recognition model to identify the type of the motor vehicle insurance document image to be reviewed, and obtains the document type of the motor vehicle insurance document image, thereby avoiding the problem that the traditional manual review method is time-consuming, error-prone, and difficult to cope with a large number of insurance applications, thereby improving the review efficiency and accuracy in the insurance process, as well as the user experience.

[0051] According to the type of certificate, a pre-trained certificate content recognition model is used to perform content recognition on the motor vehicle insurance certificate image to be reviewed, and the certificate information of the motor vehicle insurance certificate image to be reviewed is extracted, thereby avoiding the problems of low efficiency and low precision caused by manual review, improving the degree of automation of the review, thereby improving the efficiency and accuracy of the review, and generating detailed records during the certificate review process through the model, making the review traceable and avoiding unnecessary disputes.

[0052] By performing consistency verification between the certificate information and the insurance information, the problems of low efficiency and low precision caused by manual one-time verification are avoided, the efficiency and accuracy of the consistency verification are improved, and the user experience is enhanced.

[0053] Through consistency verification, it is determined whether the image of the motor vehicle insurance certificate has passed the automatic review, and then it is determined whether the certificate corresponding to the image of the motor vehicle insurance certificate is a compliant certificate, which improves the review efficiency, reduces the review cost, and improves the user experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] In order to more clearly illustrate the scheme in the present application, a brief introduction is given below to the drawings required for use in the description of the embodiments of the present application. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0055] Figure 1 is an exemplary system architecture diagram to which the present application may be applied;

[0056] Figure 2 A flow chart of an embodiment of a method for reviewing an image of a motor vehicle insurance certificate according to the present application;

[0057] Figure 3 It is a structural schematic diagram of an embodiment of a vehicle insurance certificate image review device according to the present application;

[0058] Figure 4 It is a structural schematic diagram of an embodiment of a device according to the present application. DETAILED DESCRIPTION

[0059] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by technicians in the technical field of the present application; the terms used in the specification of the application herein are only for the purpose of describing specific embodiments and are not intended to limit the present application; the terms "including" and "having" and any variations thereof in the specification and claims of the present application and the above-mentioned drawings are intended to cover non-exclusive inclusions. The terms "first", "second", etc. in the specification and claims of the present application or the above-mentioned drawings are used to distinguish different objects, not to describe a specific order.

[0060] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase in various locations in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0061] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the accompanying drawings.

[0062] like Figure 1 As shown, the system architecture 100 may include a terminal device 101, a network 102 and a server 103. The terminal device 101 may be a laptop 1011, a tablet computer 1012 or a mobile phone 1013. The network 102 is used to provide a medium for a communication link between the terminal device 101 and the server 103. The network 102 may include various connection types, such as wired, wireless communication links or optical fiber cables.

[0063] The user can use the terminal device 101 to interact with the server 103 through the network 102 to receive or send messages, etc. Various communication client applications can be installed on the terminal device 101, such as web browser applications, shopping applications, search applications, instant messaging tools, email clients, social platform software, etc.

[0064] The terminal device 101 can be any electronic device with a display screen and supporting web browsing. In addition to a laptop computer 1011, a tablet computer 1012 or a mobile phone 1013, the terminal device 101 can also be an e-book reader, an MP3 player (Moving Picture Experts Group Audio Layer III), an MP4 (Moving Picture Experts Group Audio Layer IV), a laptop computer, a desktop computer, etc.

[0065] The server 103 may be a server that provides various services, such as a background server that provides support for a web page displayed on the terminal device 101 .

[0066] It should be noted that the method for obtaining vehicle insurance investigation evidence provided in the embodiment of the present application is generally executed by a server / terminal device, and accordingly, the device for obtaining vehicle insurance investigation evidence is generally set in the server / terminal device.

[0067] It should be understood that Figure 1 The number of terminal devices, networks and servers in the embodiment is only for illustration. Any number of terminal devices, networks and servers may be provided according to implementation requirements.

[0068] Continue to refer Figure 2 , shows a flow chart of an embodiment of a method for reviewing an image of a motor vehicle insurance certificate according to the present application. The method for reviewing an image of a motor vehicle insurance certificate comprises the following steps:

[0069] S201. Obtain a motor vehicle insurance document image to be reviewed, and use a pre-trained document type recognition model to perform type recognition on the motor vehicle insurance document image to be reviewed to obtain the document type of the motor vehicle insurance document image.

[0070] In this embodiment, the motor vehicle insurance document image to be reviewed refers to the motor vehicle insurance document image uploaded by the user when applying for insurance, and may also be the document image entered during the motor vehicle insurance policy issuance process; the pre-trained document type recognition model refers to a pre-trained convolutional model, and the pre-trained convolutional model includes an output layer, a convolutional layer, a maximum pooling layer, a fully connected layer, and an output layer.

[0071] In this embodiment, a motor vehicle insurance document image to be reviewed is obtained, the motor vehicle insurance document image to be reviewed is preprocessed to obtain a standard motor vehicle insurance document image, the standard motor vehicle insurance document image is input into a pre-trained document type recognition model, and the document type of the standard motor vehicle insurance document image is identified by the pre-trained document type recognition model.

[0072] In this embodiment, the preprocessing step includes:

[0073] The vehicle insurance document image to be reviewed is subjected to image denoising, grayscale processing, image enhancement and normalization processing to obtain a standard vehicle insurance document image.

[0074] In this embodiment, image denoising is used to reduce the noise in the image of the motor vehicle insurance certificate to be reviewed and improve the image quality; the motor vehicle insurance certificate image is processed by grayscale processing, which can reduce the amount of calculation for subsequent processing and improve the efficiency and accuracy of subsequent recognition steps; the image contrast and clarity are improved by image enhancement, making the information in the image more prominent, which is convenient for subsequent certificate recognition steps and content extraction steps; the motor vehicle insurance certificate image can be converted into a numerical range that can be recognized by the model by normalization processing, avoiding the problem of subsequent unrecognition.

[0075] Specifically, the Gaussian filtering method is used to perform image denoising on the vehicle insurance certificate image to be reviewed, each pixel in the vehicle insurance certificate image is scanned, and the weighted average grayscale value of the pixels in the field determined by the template is used to replace the value of the central pixel of the template to obtain the denoised vehicle insurance certificate image; each pixel of the color vehicle insurance certificate image is converted from the RGB color space to the grayscale space to obtain the vehicle insurance certificate image after grayscale processing; the histogram equalization method is used to adjust the pixel distribution of the vehicle insurance certificate image using the histogram equalization function to obtain the vehicle insurance certificate image after image enhancement; the vehicle insurance certificate image is normalized by using the image normalization method to obtain a standard vehicle insurance certificate image, and the preprocessing steps are completed.

[0076] In one embodiment, before using the pre-trained document type recognition model to perform type recognition on the to-be-examined vehicle insurance document image, the following steps are included:

[0077] Obtain a document type data set, and divide the document type data set into a training set and a test set according to a preset ratio;

[0078] Using the training set to train the convolutional neural network model to obtain an initial document type recognition model;

[0079] Testing the initial document type recognition model according to the test set to obtain a loss value;

[0080] When the loss value is less than or equal to a preset loss threshold, a trained document type recognition model is obtained.

[0081] In this embodiment, a large number of document images with labels are obtained, and a large number of document images with type labels are aggregated into a document type data set. Each document image is classified according to the corresponding label to obtain multiple classified document type data sets. Each classified document type data set is split into multiple training subsets and multiple test subsets according to a preset ratio, and all training subsets and all test subsets are aggregated to obtain a training set and a test set; an initial document type recognition model is trained by the training set, and the initial document type recognition model is tested by using the test set. The testing method includes: using the initial document type recognition model to classify the document images corresponding to each type label A convolution operation is performed to obtain a test feature vector corresponding to each type of label, and the distance between the test feature vector and the type feature vector of the same type of label is calculated using the Euclidean distance method to obtain multiple loss distance values. The average of all loss distance values ​​is calculated to obtain the loss value, and it is determined whether the loss value is less than or equal to a preset loss threshold. When the loss value is less than or equal to the preset loss threshold, the training is completed to obtain a trained document type recognition model. When the loss value is greater than the preset loss threshold, the training set is used to continue training the convolutional neural network model until the loss value is less than or equal to the preset loss threshold, the training is terminated, and a trained document type recognition model is obtained.

[0082] In one embodiment, the method of using a pre-trained certificate type recognition model to perform type recognition on the to-be-examined vehicle insurance certificate image to obtain the certificate type of the vehicle insurance certificate image includes:

[0083] Using the convolution layer in the pre-trained certificate type recognition model to perform a convolution operation on the image of the motor vehicle insurance certificate to be reviewed, and extracting target type features of the image of the motor vehicle insurance certificate to be reviewed;

[0084] Using the pooling layer in the pre-trained document type recognition model to perform a dimensionality reduction operation on the target type feature to obtain a target type feature after dimensionality reduction;

[0085] The output layer in the pre-trained certificate type recognition model is used to calculate the probability of the target type feature after dimensionality reduction belonging to each type and the type with the largest probability is selected to obtain the certificate type of the vehicle insurance certificate image.

[0086] In this embodiment, the type of the vehicle insurance certificate image to be reviewed is identified by a pre-trained certificate recognition model, which can avoid the problems of poor accuracy and low efficiency caused by manual review, thereby improving the overall efficiency and accuracy of vehicle insurance verification and reducing labor costs. In addition, by using the model for recognition, the model can be further trained and optimized, so that the model can also recognize more types of certificates, thereby improving the degree of automation of recognition.

[0087] In this embodiment, the motor vehicle insurance certificate image to be reviewed is preprocessed to obtain a standard motor vehicle insurance certificate image, the standard motor vehicle insurance certificate image is input into a pre-trained certificate type recognition model, the convolution layer of the pre-trained certificate type recognition model is used to perform a convolution operation on the standard motor vehicle insurance certificate image, and the target type features of the standard motor vehicle insurance certificate image are extracted; the pooling layer in the pre-trained certificate type recognition model is used to perform a dimensionality reduction operation on the target type features to obtain the target type features after dimensionality reduction; the output layer refers to a softmax classifier, and the output layer in the pre-trained certificate type recognition model is used to calculate the probability that the target type features after dimensionality reduction belong to each type to obtain the probability distribution of the certificate type, and the type with the largest probability is selected from the probability distribution of the certificate type to obtain the certificate type of the motor vehicle insurance certificate image.

[0088] In another embodiment, if the identified document type is inconsistent with the document type pre-selected by the user, or any probability in the probability distribution is less than a preset probability threshold, it is determined that the document type uploaded by the customer is not a regular type of document, and the user is prompted to re-upload or transfer it for manual review.

[0089] In an implementable example A, when a user applies for insurance, the user is prompted to upload a motor vehicle insurance document image. The user uploads a photo of an ID card according to the prompt. The motor vehicle insurance document image uploaded by the user is obtained, and firstly a pre-processing operation is performed on the motor vehicle insurance document image uploaded by the user. The motor vehicle insurance document image uploaded by the user is subjected to image denoising, grayscale processing, image enhancement and normalization to obtain a standard motor vehicle insurance document image. The standard motor vehicle insurance document image is input into a pre-trained document type recognition model to extract target features of the standard motor vehicle insurance document image. According to the output layer of the pre-trained document type recognition model, the target features of the standard motor vehicle insurance document image are calculated to belong to an ID card with a probability of 0.95 and to belong to a driver's license with a probability of 0.05. Then, it is determined that the motor vehicle insurance document image uploaded by the user is an ID card.

[0090] In this embodiment, a pre-trained certificate type recognition model is used to perform type recognition on the motor vehicle insurance certificate image to be reviewed to obtain the certificate type of the motor vehicle insurance certificate image, thereby avoiding the problem that the traditional manual review method is time-consuming, error-prone, and difficult to cope with a large number of insurance applications, thereby improving the review efficiency and accuracy in the insurance process, as well as the user experience.

[0091] S202: According to the certificate type, a pre-trained certificate content recognition model is used to perform content recognition on the motor vehicle insurance certificate image to be reviewed, and certificate information of the motor vehicle insurance certificate image to be reviewed is extracted.

[0092] In this embodiment, the pre-trained document content recognition model includes a text detection module, a text recognition module, a post-processing module and a structured processing module.

[0093] In one embodiment, according to the certificate type, using a pre-trained certificate content recognition model to perform content recognition on the image of the motor vehicle insurance certificate to be reviewed, and extracting the certificate information of the image of the motor vehicle insurance certificate to be reviewed, includes:

[0094] According to the document type, using the text detection module of the pre-trained document content recognition model, the key content position detection is performed on the image of the automobile insurance document to be reviewed to obtain the key content area;

[0095] Using the text recognition module of the pre-trained document content recognition model, performing OCR processing on the key content area to extract key text from the key content area;

[0096] The structured processing module using the pre-trained certificate content recognition model is used to convert the key text into a format according to a predefined data format to obtain the certificate information of the vehicle insurance certificate image to be reviewed.

[0097] Since the position of the key content corresponding to each certificate type is different, in this embodiment, it is necessary to locate the key position of the content according to the certificate type using the text detection module of the pre-trained certificate content recognition model, locate the key content area, such as the name area, gender area, certificate number area, validity period area, etc., based on the located key content area; use the text recognition module of the pre-trained certificate content recognition model to extract the content contained in the key content area using OCR (optical character recognition) technology and convert the text in the key content area in the image into a preset text format to obtain the key text; use the structured processing module of the pre-trained certificate content recognition model to convert the format of the key text according to a predefined structure template (such as name: XXX, gender: XXX, certificate number: XXX, validity period: XXX, etc.) to obtain structured data, and finally obtain the certificate information of the motor vehicle insurance certificate image to be reviewed.

[0098] In this embodiment, according to the type of certificate, a pre-trained certificate content recognition model is used to perform content recognition on the motor vehicle insurance certificate image to be reviewed, and the certificate information of the motor vehicle insurance certificate image to be reviewed is extracted, thereby avoiding manual extraction of the content of the identity card, improving the efficiency and accuracy of certificate information extraction, and structuring the output of the certificate information to facilitate processing in subsequent steps.

[0099] Because the content contained in the key area of ​​the content extracted by OCR (Optical Character Recognition) technology may not contain required fields or extraction errors;

[0100] Therefore, in another embodiment, after performing OCR processing on the key content area using the text recognition module of the pre-trained document content recognition model to extract the key text of the key content area, the method further includes:

[0101] Using the post-processing module of the pre-trained document content recognition model to perform field verification on the key text to query whether the predefined required fields have been filled in;

[0102] When all the required fields are filled in, all the required fields are verified one by one with the text corresponding to the image of the automobile insurance certificate to be reviewed;

[0103] The verified key text is input into the structured processing module using the pre-trained document content recognition model.

[0104] In this embodiment, the required boxes of the fields are predefined, and the post-processing module of the pre-trained certificate content recognition model is used to perform field verification on the key text, and all the required boxes are queried whether there are required fields. When any required box is missing a value, the OCR processing step is returned to reprocess until all the required boxes have corresponding required fields, and all the required fields are verified one by one with the text fields corresponding to the motor vehicle insurance certificate image to be reviewed. When the verification fails, the error fields are deleted and filled until the verification is passed, and the key text that passes the verification is input into the structured processing module using the pre-trained certificate content recognition model.

[0105] In one embodiment, after extracting the certificate information of the vehicle insurance certificate image to be reviewed, the method further includes:

[0106] Acquire the current time, and verify the validity period of the vehicle insurance certificate image to be reviewed through the certificate information and the current time;

[0107] When the current time is not within the validity period of the vehicle insurance certificate image to be reviewed, it is determined that the vehicle insurance certificate image to be reviewed is not compliant, and the user is prompted to re-upload another type of certificate image.

[0108] In this embodiment, the current time is obtained, and the current time is the real-time time when the motor vehicle insurance certificate image to be reviewed is processed; and the current time is compared with the validity period time on the certificate information of the motor vehicle insurance certificate image to be reviewed to determine whether the current time is within the validity period time on the certificate information of the motor vehicle insurance certificate image to be reviewed. When the current time is not within the validity period time on the certificate information of the motor vehicle insurance certificate image to be reviewed, it is determined that the motor vehicle insurance certificate image to be reviewed uploaded by the user is not compliant, and the user is prompted to re-upload other types of motor vehicle insurance certificate images within the validity period. When the current time is within the validity period time on the certificate information of the motor vehicle insurance certificate image to be reviewed, the validity period authentication is passed.

[0109] In this embodiment, by comparing the certificate information with the current time, the validity of the certificate is ensured, the cost of manual review is reduced, the risk rate of missing the validity period of the certificate is reduced, and the compliance of the certificate and the efficiency of the certificate review are improved.

[0110] Continuing with the above-mentioned feasible example A, after recognizing that the type of certificate uploaded by the user is an ID card type, the ID card image is input into the trained content recognition model, and according to the ID card type, the text detection module of the content recognition model is used to determine the key content areas of the ID card image, and the name area, gender area, ID number area, validity period area, and address area are determined; according to the determined and identified key areas, the text recognition module of the pre-trained content recognition model is used to perform OCR processing on each key area, and the key text corresponding to each key area, name xxx, gender x, ID number xxx, validity period xxx, address xxx, is extracted, and these key texts are verified with the corresponding texts on the ID card image. When the verification passes, the key text is structured according to the format of name: xxx, gender: x, ID number: xxx, validity period: xxx, address: xxx, to obtain the ID card information, and the validity of the ID card information is verified with the current time to obtain the verification result.

[0111] In this embodiment, according to the type of certificate, a pre-trained certificate content recognition model is used to perform content recognition on the motor vehicle insurance certificate image to be reviewed, and the certificate information of the motor vehicle insurance certificate image to be reviewed is extracted, thereby avoiding the problems of low efficiency and low precision caused by manual review, improving the degree of automation of the review, and thus improving the efficiency and accuracy of the review. In addition, the model will generate detailed records during the certificate review process, making the review traceable and avoiding unnecessary disputes.

[0112] S203: Obtain insurance information corresponding to the vehicle insurance certificate image to be reviewed, and perform consistency verification between the certificate information and the insurance information.

[0113] In this embodiment, the insurance information refers to the information filled in by the user in the insurance application process, including the personal information of the insured, such as name, date of birth, ID number, gender, etc.; the insurance information corresponding to the vehicle insurance certificate image is obtained, and the certificate information is checked for consistency with the insurance information.

[0114] In one embodiment, performing consistency check between the certificate information and the insurance information includes:

[0115] Structuring the insurance information according to a predefined structure template to obtain structured insurance information;

[0116] Perform a consistency check between the certificate information and the information with the same prefix in the structure insurance information.

[0117] In this embodiment, the acquired insurance information is also structured according to the same structured model as the certificate information to obtain the structured insurance information. Since the same structured template is used, the prefix corresponding to the certificate information and the information contained in the prefix corresponding to the structured insurance information can be verified, such as the name in the certificate information: XXX and the name in the structured insurance information: XXX, until all consistency verifications of the certificate information are completed.

[0118] In another embodiment, the step of performing consistency check between the certificate information and the information with the same prefix in the structure insurance information includes:

[0119] If the consistency check fails, the vehicle insurance document image to be reviewed is marked as a suspicious document image, and the suspicious document image is pushed to manual processing.

[0120] In this embodiment, if the consistency check fails, there is a problem with the certificate or the user's insurance information is filled in incorrectly, so the user is not prompted to re-upload but is first marked as a suspicious certificate image, and the certificate image and insurance information are pushed to manual review.

[0121] In this embodiment, by reviewing suspicious ID images, the shortcomings of machine verification are compensated, the accuracy of verification is improved, and the misjudgment rate is reduced. The cause of illegality can be accurately found, and the user is prompted to modify the insurance information or re-upload the ID image based on the cause, thereby improving the user experience.

[0122] Continuing with the above-mentioned feasible example A, the insurance information corresponding to the ID card information is extracted from the database storing the user's insurance information, such as the insured person XXX, ID number XXX, age XXX, gender XXX, etc. The insurance information is structured according to the preset structure template to obtain name: XXX, ID number: XXX, age: XXX, gender: XXX, etc. The name: XXX of the ID information is checked for consistency with the name: XXX of the insured information, the ID number: XXX of the ID information is checked for consistency with the ID number: XXX of the insured information, the gender: XXX of the ID information is checked for consistency with the gender: XXX of the insured information, and the age: XXX of the ID information is checked for consistency with the age: XXX of the insured information.

[0123] In this embodiment, by performing consistency verification between the certificate information and the insurance information, the problems of low efficiency and low precision caused by manual one-time verification are avoided, the efficiency and accuracy of the consistency verification are improved, and the user experience is enhanced.

[0124] S204: If the consistency check passes, it is determined that the vehicle insurance certificate image to be reviewed has passed the automatic review, and it is determined that the certificate corresponding to the vehicle insurance certificate image to be reviewed is a compliant certificate.

[0125] In this embodiment, if the consistency check passes, it is determined that the vehicle insurance certificate image to be reviewed has passed the automatic review, and further it is determined that the certificate corresponding to the certificate image uploaded by the user is a compliant certificate.

[0126] In one embodiment, after determining that the certificate corresponding to the to-be-reviewed vehicle insurance certificate image is a compliant certificate, the method further includes:

[0127] Storing the certificate information of the compliance certificate in a preset certificate information database;

[0128] When receiving a policy renewal request, obtaining the certificate information corresponding to the policy renewal request from a preset certificate information database;

[0129] Determine whether the certificate information corresponding to the renewal request is within the validity period;

[0130] When the certificate information corresponding to the policy renewal request is within the validity period, the certificate information corresponding to the policy renewal request is called to complete the policy renewal operation.

[0131] In this embodiment, after obtaining the user's consent, the certificate information of the compliant certificate is stored in a preset certificate information database. In the next policy renewal cycle, the user initiates a policy renewal request, and the certificate information corresponding to the policy renewal request is extracted from the preset certificate information database to determine whether the certificate information corresponding to the policy renewal request is within the validity period. If the certificate information of the policy renewal is within the validity period, the certificate information of the compliant certificate is directly used to complete the policy renewal request. If the certificate information of the policy renewal request is not within the validity period, the user is prompted to re-upload the compliant certificate image for verification.

[0132] In this embodiment, by storing the certificate information of the compliance certificate and reusing it in the policy renewal process, the efficiency of handling the policy renewal is improved and the user experience is enhanced.

[0133] Continuing with the above-mentioned feasible example A, when the ID card image uploaded by the user passes the consistency check, an option will appear on the computer interface: "Do you allow the storage of your ID card information?" When the user selects the "Yes" option, the ID card information will be stored in the preset ID card information database. In the next renewal cycle, the user initiates a renewal request, obtains the ID card information corresponding to the renewal request, and determines whether the ID card information is within the validity period. If the ID card information is within the validity period, the process is completed directly and the user can go to the payment page.

[0134] In this embodiment, consistency check is performed to determine whether the vehicle insurance certificate image has passed automatic review, and then to determine whether the certificate corresponding to the vehicle insurance certificate image is a compliant certificate, thereby improving review efficiency, reducing review costs, and improving user experience.

[0135] It should be emphasized that in order to further ensure the privacy and security of the above-mentioned certificate information and user information, the above-mentioned certificate information and user information can also be stored in a node of a blockchain.

[0136] The blockchain referred to in this application is a new application model of computer technologies such as distributed data storage, peer-to-peer transmission, consensus mechanism, encryption algorithm, etc. Blockchain is essentially a decentralized database, a string of data blocks generated by cryptographic methods. Each data block contains a batch of network transaction information, which is used to verify the validity of its information (anti-counterfeiting) and generate the next block. Blockchain can include the underlying blockchain platform, platform product service layer, and application service layer.

[0137] The embodiments of the present application can acquire and process relevant data based on artificial intelligence technology. Among them, artificial intelligence (AI) is the theory, method, technology and application system that uses digital computers or machines controlled by digital computers to simulate, extend and expand human intelligence, perceive the environment, acquire knowledge and use knowledge to obtain the best results.

[0138] AI basic technologies generally include sensors, dedicated AI chips, cloud computing, distributed storage, big data processing technology, operation / interaction systems, mechatronics, etc. AI software technologies mainly include computer vision technology, robotics technology, biometrics technology, speech processing technology, natural language processing technology, and machine learning / deep learning.

[0139] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through computer-readable instructions, and the computer-readable instructions can be stored in a computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, the aforementioned storage medium can be a non-volatile storage medium such as a disk, an optical disk, a read-only memory (ROM), or a random access memory (RAM).

[0140] It should be understood that, although the steps in the flowchart of the accompanying drawings are displayed in sequence as indicated by the arrows, these steps are not necessarily executed in sequence in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least a part of the steps in the flowchart of the accompanying drawings may include multiple sub-steps or multiple stages, and these sub-steps or stages are not necessarily executed at the same time, but can be executed at different times, and their execution order is not necessarily sequential, but can be executed in turn or alternately with other steps or at least a part of the sub-steps or stages of other steps.

[0141] Further references Figure 3 , as a response to the above Figure 2 The present application provides an embodiment of a device for reviewing an image of a motor vehicle insurance certificate. Figure 2 Corresponding to the method embodiment shown, the apparatus can be specifically applied to various computer devices.

[0142] like Figure 3As shown, the vehicle insurance certificate image review device 300 described in this embodiment includes: a type identification module 301, a content extraction module 302, an information verification module 303, and a compliance judgment module 304. Among them:

[0143] The type recognition module 301 is used to obtain the image of the motor vehicle insurance certificate to be reviewed, and use a pre-trained certificate type recognition model to perform type recognition on the image of the motor vehicle insurance certificate to be reviewed to obtain the certificate type of the image of the motor vehicle insurance certificate;

[0144] The content extraction module 302 is used to perform content recognition on the image of the motor vehicle insurance certificate to be reviewed using a pre-trained certificate content recognition model according to the certificate type, and extract certificate information of the image of the motor vehicle insurance certificate to be reviewed;

[0145] The information verification module 303 is used to obtain the insurance information corresponding to the automobile insurance certificate image to be reviewed, and to perform consistency verification between the certificate information and the insurance information;

[0146] The compliance judgment module 304 is used to judge whether the vehicle insurance certificate image to be reviewed has passed the automatic review if the consistency check passes, and determine that the certificate corresponding to the vehicle insurance certificate image to be reviewed is a compliant certificate.

[0147] In this embodiment, the present application uses a pre-trained certificate type recognition model to perform type recognition on the vehicle insurance certificate image to be reviewed, and obtains the certificate type of the vehicle insurance certificate image, thereby avoiding the problem that the traditional manual review method is time-consuming, error-prone, and difficult to handle a large number of insurance applications, thereby improving the review efficiency and accuracy in the insurance process, as well as the user experience;

[0148] According to the document type, a pre-trained document content recognition model is used to perform content recognition on the image of the automobile insurance document to be reviewed, and the document information of the image of the automobile insurance document to be reviewed is extracted, thereby avoiding the problems of low efficiency and low precision caused by manual review, improving the degree of automation of the review, thereby improving the efficiency and accuracy of the review, and generating detailed records during the document review process through the model, so that the review is traceable and unnecessary disputes are avoided;

[0149] By performing consistency check between the certificate information and the insurance information, the problems of low efficiency and low precision caused by manual one-time verification are avoided, the efficiency and accuracy of consistency check are improved, and thus the user experience is improved;

[0150] Through consistency verification, it is determined whether the image of the motor vehicle insurance certificate has passed the automatic review, and then it is determined whether the certificate corresponding to the image of the motor vehicle insurance certificate is a compliant certificate, which improves the review efficiency, reduces the review cost, and improves the user experience.

[0151] In order to solve the above technical problems, the present application embodiment also provides a device (computer device). Figure 4 , Figure 4 This is a basic structural block diagram of the computer device in this embodiment.

[0152] The computer device 4 includes a memory 41, a processor 42, and a network interface 43 that are interconnected and communicated through a system bus. It should be noted that the figure only shows a computer device 4 with a memory 41, a processor 42, and a network interface 43, but it should be understood that it is not required to implement all the components shown, and more or fewer components can be implemented instead. Among them, those skilled in the art can understand that the computer device here is a device that can automatically perform numerical calculations and / or information processing according to pre-set or stored instructions, and its hardware includes but is not limited to microprocessors, application specific integrated circuits (Application Specific Integrated Circuit, ASIC), programmable gate arrays (Fieid-Programmable Gate Array, FPGA), digital processors (Digital Signal Processor, DSP), embedded devices, etc.

[0153] The computer device may be a computing device such as a desktop computer, a notebook, a PDA, a cloud server, etc. The computer device may interact with a user through a keyboard, a mouse, a remote controller, a touch pad, or a voice control device.

[0154] The memory 41 includes at least one type of readable storage medium, and the readable storage medium includes flash memory, hard disk, multimedia card, card-type memory (for example, SD or DX memory, etc.), random access memory (RAM), static random access memory (SRAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), programmable read-only memory (PROM), magnetic memory, disk, optical disk, etc. In some embodiments, the memory 41 can be an internal storage unit of the computer device 4, such as a hard disk or memory of the computer device 4. In other embodiments, the memory 41 can also be an external storage device of the computer device 4, such as a plug-in hard disk equipped on the computer device 4, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (FlashCard), etc. Of course, the memory 41 can also include both the internal storage unit of the computer device 4 and its external storage device. In this embodiment, the memory 41 is generally used to store the operating system and various application software installed on the computer device 4, such as computer-readable instructions of the adjustment method of pet insurance, etc. In addition, the memory 41 can also be used to temporarily store various types of data that have been output or are to be output.

[0155] The processor 42 may be a central processing unit (CPU), a controller, a microcontroller, a microprocessor, or other data processing chips in some embodiments. The processor 42 is generally used to control the overall operation of the computer device 4. In this embodiment, the processor 42 is used to run the computer-readable instructions stored in the memory 41 or process data, such as computer-readable instructions for running the method for reviewing the vehicle insurance document image.

[0156] The network interface 43 may include a wireless network interface or a wired network interface. The network interface 43 is generally used to establish a communication connection between the computer device 4 and other electronic devices.

[0157] During the implementation of the electronic device of the present application, a pre-trained document type recognition model is used to identify the type of the motor vehicle insurance document image to be reviewed, so as to obtain the document type of the motor vehicle insurance document image, thereby avoiding the problem that the traditional manual review method is time-consuming, error-prone, and difficult to cope with a large number of insurance applications, thereby improving the review efficiency and accuracy in the insurance process, as well as the user experience.

[0158] According to the type of certificate, a pre-trained certificate content recognition model is used to perform content recognition on the motor vehicle insurance certificate image to be reviewed, and the certificate information of the motor vehicle insurance certificate image to be reviewed is extracted, thereby avoiding the problems of low efficiency and low precision caused by manual review, improving the degree of automation of the review, thereby improving the efficiency and accuracy of the review, and generating detailed records during the certificate review process through the model, making the review traceable and avoiding unnecessary disputes.

[0159] By performing consistency verification between the certificate information and the insurance information, the problems of low efficiency and low precision caused by manual one-time verification are avoided, the efficiency and accuracy of the consistency verification are improved, and the user experience is enhanced.

[0160] Through consistency verification, it is determined whether the image of the motor vehicle insurance certificate has passed the automatic review, and then it is determined whether the certificate corresponding to the image of the motor vehicle insurance certificate is a compliant certificate, which improves the review efficiency, reduces the review cost, and improves the user experience.

[0161] The present application also provides another embodiment, namely, providing a storage medium (computer-readable storage medium), wherein the computer-readable storage medium stores computer-readable instructions, and the computer-readable instructions can be executed by at least one processor to enable the at least one processor to perform the steps of the above-mentioned method for reviewing the vehicle insurance document image.

[0162] During the implementation of the computer-readable storage medium of the present application, a pre-trained document type recognition model is used to perform type recognition on the motor vehicle insurance document image to be reviewed, thereby obtaining the document type of the motor vehicle insurance document image. This avoids the problem that the traditional manual review method is time-consuming, error-prone, and difficult to cope with a large number of insurance applications, thereby improving the review efficiency and accuracy in the insurance process, as well as the user experience.

[0163] According to the type of certificate, a pre-trained certificate content recognition model is used to perform content recognition on the motor vehicle insurance certificate image to be reviewed, and the certificate information of the motor vehicle insurance certificate image to be reviewed is extracted, thereby avoiding the problems of low efficiency and low precision caused by manual review, improving the degree of automation of the review, thereby improving the efficiency and accuracy of the review, and generating detailed records during the certificate review process through the model, making the review traceable and avoiding unnecessary disputes.

[0164] By performing consistency verification between the certificate information and the insurance information, the problems of low efficiency and low precision caused by manual one-time verification are avoided, the efficiency and accuracy of the consistency verification are improved, and the user experience is enhanced.

[0165] Through consistency verification, it is determined whether the image of the motor vehicle insurance certificate has passed the automatic review, and then it is determined whether the certificate corresponding to the image of the motor vehicle insurance certificate is a compliant certificate, which improves the review efficiency, reduces the review cost, and improves the user experience.

[0166] The non-Company software tools or components appearing in the embodiments of this application are merely examples and do not represent actual use.

[0167] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present application, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), and includes a number of instructions for a terminal device (which can be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in each embodiment of the present application.

[0168] Obviously, the embodiments described above are only some embodiments of the present application, rather than all embodiments. The preferred embodiments of the present application are given in the accompanying drawings, but they do not limit the patent scope of the present application. The present application can be implemented in many different forms. On the contrary, the purpose of providing these embodiments is to make the understanding of the disclosure of the present application more thorough and comprehensive. Although the present application is described in detail with reference to the aforementioned embodiments, for those skilled in the art, it is still possible to modify the technical solutions recorded in the aforementioned specific implementation methods, or to perform equivalent replacement of some of the technical features therein. Any equivalent structure made using the contents of the specification and drawings of this application, directly or indirectly used in other related technical fields, is similarly within the scope of patent protection of this application.

Claims

1. A method for reviewing an image of a motor vehicle insurance certificate, characterized in that: The steps include: Acquire a motor vehicle insurance certificate image to be reviewed, and use a pre-trained certificate type recognition model to perform type recognition on the motor vehicle insurance certificate image to be reviewed to obtain the certificate type of the motor vehicle insurance certificate image; According to the certificate type, a pre-trained certificate content recognition model is used to perform content recognition on the motor vehicle insurance certificate image to be reviewed, and the certificate information of the motor vehicle insurance certificate image to be reviewed is extracted; Acquire the insurance information corresponding to the vehicle insurance certificate image to be reviewed, and perform consistency check between the certificate information and the insurance information; If the consistency check passes, it is determined that the vehicle insurance certificate image to be reviewed has passed the automatic review, and it is determined that the certificate corresponding to the vehicle insurance certificate image to be reviewed is a compliant certificate.

2. The method for reviewing an automobile insurance certificate image according to claim 1, characterized in that: The method of using a pre-trained certificate type recognition model to perform type recognition on the to-be-examined motor vehicle insurance certificate image to obtain the certificate type of the motor vehicle insurance certificate image includes: Using the convolution layer in the pre-trained certificate type recognition model to perform a convolution operation on the image of the motor vehicle insurance certificate to be reviewed, and extracting target type features of the image of the motor vehicle insurance certificate to be reviewed; Using the pooling layer in the pre-trained document type recognition model to perform a dimensionality reduction operation on the target type feature to obtain a target type feature after dimensionality reduction; The output layer in the pre-trained certificate type recognition model is used to calculate the probability of the target type feature after dimensionality reduction belonging to each type and the type with the largest probability is selected to obtain the certificate type of the vehicle insurance certificate image.

3. The method for reviewing the image of the motor vehicle insurance certificate according to claim 1, characterized in that: The method of using a pre-trained document content recognition model to perform content recognition on the image of the automobile insurance document to be reviewed according to the document type, and extracting document information of the image of the automobile insurance document to be reviewed, includes: According to the document type, using the text detection module of the pre-trained document content recognition model, the key content position detection is performed on the image of the automobile insurance document to be reviewed to obtain the key content area; Using the text recognition module of the pre-trained document content recognition model, performing OCR processing on the key content area to extract key text from the key content area; The structured processing module using the pre-trained certificate content recognition model is used to convert the key text into a format according to a predefined data format to obtain the certificate information of the vehicle insurance certificate image to be reviewed.

4. The method for reviewing an automobile insurance certificate image as claimed in claim 3, characterized in that: After performing OCR processing on the key content area using the text recognition module of the pre-trained document content recognition model to extract key text from the key content area, the method further includes: Using the post-processing module of the pre-trained document content recognition model to perform field verification on the key text to query whether the predefined required fields have been filled in; When all the predefined mandatory fields are filled in, all the predefined mandatory fields are verified one by one with the text corresponding to the image of the automobile insurance certificate to be reviewed; The verified key text is input into the structured processing module using the pre-trained document content recognition model.

5. The method for reviewing an automobile insurance certificate image according to claim 1, characterized in that: After extracting the certificate information of the vehicle insurance certificate image to be reviewed, the method further includes: Acquire the current time, and verify the validity period of the vehicle insurance certificate image to be reviewed through the certificate information and the current time; When the current time is not within the validity period of the vehicle insurance certificate image to be reviewed, it is determined that the vehicle insurance certificate image to be reviewed is not compliant, and the user is prompted to re-upload another type of certificate image.

6. The method for reviewing an automobile insurance certificate image according to claim 1, characterized in that: The consistency check between the certificate information and the insurance information includes: Structuring the insurance information according to a predefined structure template to obtain structured insurance information; Perform a consistency check between the certificate information and the information with the same prefix in the structure insurance information.

7. The method for reviewing an automobile insurance certificate image according to claim 1, characterized in that: After determining that the certificate corresponding to the to-be-examined vehicle insurance certificate image is a compliant certificate, the method further includes: Storing the certificate information of the compliance certificate in a preset certificate information database; When receiving a policy renewal request, obtaining the certificate information corresponding to the policy renewal request from the preset certificate information database; Determine whether the certificate information corresponding to the renewal request is within the validity period; When the certificate information corresponding to the policy renewal request is within the validity period, the certificate information corresponding to the policy renewal request is called to complete the policy renewal operation.

8. A device for reviewing images of motor vehicle insurance certificates, characterized in that: The device comprises: A type recognition module is used to obtain the image of the motor vehicle insurance certificate to be reviewed, and use a pre-trained certificate type recognition model to perform type recognition on the image of the motor vehicle insurance certificate to be reviewed to obtain the certificate type of the image of the motor vehicle insurance certificate; A content extraction module, used to perform content recognition on the image of the motor vehicle insurance certificate to be reviewed using a pre-trained certificate content recognition model according to the certificate type, and extract certificate information of the image of the motor vehicle insurance certificate to be reviewed; An information verification module, used to obtain the insurance information corresponding to the automobile insurance certificate image to be reviewed, and to perform consistency verification between the certificate information and the insurance information; The compliance judgment module is used to judge whether the vehicle insurance certificate image to be reviewed has passed the automatic review if the consistency check passes, and determine that the certificate corresponding to the vehicle insurance certificate image to be reviewed is a compliant certificate.

9. A computer device, characterized in that: The computer device comprises: at least one processor; and, a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the method for reviewing the vehicle insurance document image as described in any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the method for reviewing the vehicle insurance document image as described in any one of claims 1 to 7 is implemented.

Images