Vehicle loss claim settlement policy generation method and device, electronic equipment and storage medium
By obtaining vehicle damage images and model information from the auto insurance accident report, damage identification and structure determination are carried out, and combined with vehicle structure information, accurate identification and maintenance solutions for vehicle component damage are achieved, and the accuracy of vehicle damage identification in traditional technology is solved. The problem of low increases the accuracy of vehicle damage claims insurance policies.
Patent Information
- Application Number
- CN202411368734.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-27
- Publication Date
- 2025-01-07
AI Technical Summary
Traditional image recognition technology is limited by the quality and integrity of on-site images in vehicle damage claims business, and cannot capture every damaged component in detail, resulting in low accuracy in vehicle damage recognition, affecting the accurate generation of vehicle damage claims insurance policies.
By obtaining vehicle damage images and model information from the auto insurance accident report, damage identification, structure determination and damage prediction are carried out, and combined with vehicle structure information, accurate identification of vehicle component damage and maintenance plan are achieved, and accurate vehicle damage claims insurance policy is generated .
It improves the identification ability of vehicle damage components, overcomes the problem that on-site images cannot capture damaged components in detail, and ensures the accuracy and accuracy of vehicle damage claims insurance policies.
Smart Images
Figure CN119273483A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of image processing technology, and in particular to a method and device for generating a vehicle damage claim policy, an electronic device, and a storage medium. Background Art
[0002] In the vehicle damage claims business, image processing technology is usually relied on to identify vehicle damage through on-site vehicle damage images and determine the repair plan accordingly. However, traditional image recognition technology is limited by the quality and integrity of on-site images. Since on-site images often cannot capture every damaged part in detail, the accuracy of vehicle damage image recognition is limited, which in turn affects the accurate generation of vehicle damage claims policies. Therefore, how to improve the ability to identify vehicle damage parts in order to obtain accurate vehicle damage claims policies has become an urgent problem to be solved. Summary of the invention
[0003] The main purpose of the embodiments of the present application is to propose a method and device for generating a vehicle damage claim policy, an electronic device and a storage medium, aiming to improve the ability to identify damaged vehicle parts in order to obtain an accurate vehicle damage claim policy.
[0004] To achieve the above-mentioned purpose, a first aspect of an embodiment of the present application proposes a method for generating a vehicle damage claim policy, the method comprising:
[0005] Obtaining vehicle damage images and vehicle model information of the damaged vehicle from the automobile insurance accident report;
[0006] Perform damage identification according to the vehicle damage image to obtain component damage information;
[0007] Determining the structure of the damaged vehicle according to the vehicle model information to obtain vehicle structure information;
[0008] Perform damage prediction based on the vehicle structure information and the component damage information to obtain vehicle damage information;
[0009] Determining a vehicle maintenance plan for the damaged vehicle based on the vehicle damage information;
[0010] The information is merged according to the vehicle insurance accident report and the vehicle repair plan to obtain a target vehicle damage claim policy.
[0011] In some embodiments, the step of performing damage identification according to the vehicle damage image to obtain component damage information includes:
[0012] Performing feature extraction based on the vehicle damage image to obtain image damage features;
[0013] Performing feature conversion according to the image damage feature to obtain a text damage feature;
[0014] Damage identification is performed according to the text damage features to obtain the component damage information.
[0015] In some embodiments, extracting features from the vehicle damage image to obtain image damage features includes:
[0016] Perform damage location according to the vehicle damage image to obtain damage location information;
[0017] Perform damage degree assessment according to the damage location information to obtain damage assessment classification information;
[0018] Feature aggregation is performed according to the damage location information and the damage assessment classification information to obtain the image damage feature.
[0019] In some embodiments, the performing damage location according to the vehicle damage image to obtain damage location information includes:
[0020] Performing image convolution on the vehicle damage image to obtain a damage convolution feature map;
[0021] Perform sliding window frame selection on the damage convolution feature map to obtain an initial damage positioning frame;
[0022] Perform positioning frame prediction according to the initial damage positioning frame to obtain positioning frame prediction data; wherein the positioning frame prediction data is used to indicate whether the image in the initial damage positioning frame is a damaged component image or a background image;
[0023] The initial damage positioning frame is screened according to the positioning frame prediction data to obtain the damage positioning information.
[0024] In some embodiments, performing feature conversion according to the image damage feature to obtain the text damage feature includes:
[0025] Performing feature encoding on the image damage feature by using a preset self-attention encoder to obtain encoded feature information;
[0026] Decoding the image damage feature information by a preset decoder to obtain decoded feature information;
[0027] The text damage feature is obtained by performing text generation on the decoded feature information through a preset text generator.
[0028] In some embodiments, the vehicle construction information includes vehicle component information, component connection information, and connection strength information; and performing damage prediction based on the vehicle construction information and the component damage information to obtain vehicle damage information includes:
[0029] Perform vehicle modeling according to the vehicle component information, the component connection information and the connection strength information to obtain an initial vehicle model;
[0030] Performing damage alignment on the initial vehicle model according to the component damage information to obtain a damaged vehicle model;
[0031] Damage prediction is performed according to the damaged vehicle model to obtain the vehicle damage information.
[0032] In some embodiments, performing damage prediction according to the damaged vehicle model to obtain the vehicle damage information includes:
[0033] Performing component damage prediction according to the damaged vehicle model to obtain component damage prediction information;
[0034] Classify the maintenance type according to the component damage prediction information to obtain the component maintenance category;
[0035] The vehicle damage information is obtained by merging the component damage prediction information and the component maintenance category.
[0036] To achieve the above-mentioned purpose, a second aspect of an embodiment of the present application provides a vehicle damage claim policy generation device, the device comprising:
[0037] A data acquisition module is used to obtain a vehicle damage image and vehicle model information of a damaged vehicle from a motor vehicle insurance accident report;
[0038] A damage identification module, used to perform damage identification based on the vehicle damage image to obtain component damage information;
[0039] A structure determination module, used to determine the structure of the damaged vehicle according to the vehicle model information to obtain vehicle structure information;
[0040] A damage prediction module, used for performing damage prediction based on the vehicle structure information and the component damage information to obtain vehicle damage information;
[0041] A solution determination module, used to determine a vehicle maintenance solution for the damaged vehicle based on the vehicle damage information;
[0042] The information merging module is used to merge the information according to the vehicle insurance accident report and the vehicle repair plan to obtain a target vehicle damage claim policy.
[0043] To achieve the above objectives, a third aspect of an embodiment of the present application proposes an electronic device, which includes a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the method described in the first aspect is implemented.
[0044] To achieve the above-mentioned purpose, the fourth aspect of an embodiment of the present application proposes a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method described in the first aspect is implemented.
[0045] The present application proposes a method and device for generating a vehicle damage claim policy, an electronic device and a storage medium, which first obtains a vehicle damage image and vehicle model information of a damaged vehicle from a motor vehicle insurance accident report, then performs damage recognition based on the vehicle damage image to obtain component damage information, and then determines the structure of the damaged vehicle based on the vehicle model information to obtain vehicle structure information, and finally performs damage prediction based on the vehicle structure information and component damage information, so as to achieve further prediction of potential vehicle damage based on damage on vehicle components identified from the damage image combined with the specific structure of the vehicle, more accurately determine the damage to the vehicle, and overcome the problem that the accuracy of vehicle damage recognition is limited because the on-site image is often unable to capture every damaged component in detail; further, a vehicle repair plan for the damaged vehicle is determined based on the vehicle damage information, and then information is merged based on the motor vehicle insurance accident report and the vehicle repair plan, so as to improve the recognition capability of vehicle damaged components and obtain an accurate target vehicle damage claim policy. BRIEF DESCRIPTION OF THE DRAWINGS
[0046] Figure 1 is a flow chart of a method for generating a vehicle damage claim insurance policy provided by an embodiment of the present application;
[0047] Figure 2 yes Figure 1 Flow chart of step S102 in FIG.
[0048] Figure 3 yes Figure 2 Flow chart of step S201 in FIG.
[0049] Figure 4 yes Figure 3 Flow chart of step S301 in FIG.
[0050] Figure 5 yes Figure 2 Flow chart of step S202 in FIG.
[0051] Figure 6 yes Figure 1 Flow chart of step S104 in FIG.
[0052] Figure 7 yes Figure 6 Flow chart of step S601 in FIG.
[0053] Figure 8 It is a structural schematic diagram of a vehicle damage claim policy generating device provided in an embodiment of the present application;
[0054] Fig. 9 It is a schematic diagram of the hardware structure of the electronic device provided in the embodiment of the present application. DETAILED DESCRIPTION
[0055] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.
[0056] It should be noted that, although the functional modules are divided in the device schematic diagram and the logical order is shown in the flowchart, in some cases, the steps shown or described may be performed in a different order than the module division in the device or the order in the flowchart. The terms "first", "second", etc. in the specification, claims and the above drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence.
[0057] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art to which this application belongs. The terms used herein are only for the purpose of describing the embodiments of this application and are not intended to limit this application.
[0058] First, some nouns involved in this application are analyzed:
[0059] Vehicle construction information: It is the data and documents that study and record how the various components of a vehicle are assembled, connected and configured. It is a branch of mechanical engineering and automotive engineering. Vehicle construction information is used to indicate and understand the details of the design and function implementation of the vehicle, and is used to guide the manufacture, maintenance and modification of the vehicle, including the vehicle power system, chassis structure, body materials and electrical system.
[0060] Sliding window: It is a technique used for local area analysis and object detection in image processing, and is a branch of computer vision and artificial intelligence. Sliding window technology gradually analyzes the local characteristics of the image by moving a fixed-size frame selection area across the entire image to identify and locate specific objects in the image; research in this field includes target tracking, feature extraction, behavior recognition, and real-time video analysis. Sliding window allows the system to perform detailed analysis of specific areas without processing the entire image, thereby achieving effective recognition and classification of multiple objects in complex scenes.
[0061] Bounding box: A rectangular box used to define the position of an object in an image in the field of computer vision. It is a branch of artificial intelligence and computer vision. Bounding box technology identifies and selects the position and range of a target object by drawing a rectangle in an image. Research in this field includes object detection, image segmentation, target tracking, and activity recognition. Bounding boxes not only help algorithms accurately identify and locate key objects in an image, but also provide a basis for further image analysis, such as object attribute analysis and behavior prediction.
[0062] In the vehicle damage claims business, image processing technology is usually relied on to identify vehicle damage through on-site vehicle damage images and determine the repair plan accordingly. However, traditional image recognition technology is limited by the quality and integrity of on-site images. Since on-site images often cannot capture every damaged part in detail, the accuracy of vehicle damage image recognition is limited, which in turn affects the accurate generation of vehicle damage claims policies. Therefore, how to improve the ability to identify vehicle damage parts in order to obtain accurate vehicle damage claims policies has become an urgent problem to be solved.
[0063] Based on this, the embodiments of the present application provide a method and device for generating a vehicle damage claim policy, an electronic device and a storage medium, aiming to improve the ability to identify damaged vehicle parts in order to obtain an accurate vehicle damage claim policy.
[0064] The embodiments of the present application provide a method and device for generating a vehicle damage claim policy, an electronic device, and a storage medium, which are specifically described through the following embodiments. First, the method for generating a vehicle damage claim policy in the embodiments of the present application is described.
[0065] The embodiments of the present application can acquire and process relevant data based on artificial intelligence technology. 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.
[0066] 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.
[0067] The method for generating a vehicle damage claim policy provided in the embodiment of the present application relates to the field of image processing technology. The method for generating a vehicle damage claim policy provided in the embodiment of the present application can be applied to a terminal, can be applied to a server side, or can be software running in a terminal or a server side. In some embodiments, the terminal can be a smart phone, a tablet computer, a laptop computer, a desktop computer, etc.; the server side can be configured as an independent physical server, or can be configured as a server cluster or a distributed system composed of multiple physical servers, or can be configured as a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms; the software can be an application that implements the method for generating a vehicle damage claim policy, etc., but is not limited to the above forms.
[0068] The present application can be used in many general or special computer system environments or configurations. For example: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, distributed computing environments including any of the above systems or devices, etc. The present application can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform specific tasks or implement specific abstract data types. The present application can also be practiced in distributed computing environments, in which tasks are performed by remote processing devices connected through a communication network. In a distributed computing environment, program modules can be located in local and remote computer storage media including storage devices.
[0069] It should be noted that in each specific implementation of the present application, when it comes to the need to perform relevant processing based on data related to user identity or characteristics such as user information, user behavior data, user historical data, and user location information, the user's permission or consent will be obtained first, and the collection, use, and processing of these data will comply with relevant laws, regulations, and standards. In addition, when the embodiment of the present application needs to obtain the user's sensitive personal information, the user's separate permission or consent will be obtained through a pop-up window or by jumping to a confirmation page. After clearly obtaining the user's separate permission or consent, the necessary user-related data for the normal operation of the embodiment of the present application will be obtained.
[0070] Figure 1 is an optional flow chart of the method for generating a vehicle damage claim policy provided in an embodiment of the present application. Figure 1 The method may include but is not limited to steps S101 to S106.
[0071] Step S101, obtaining a vehicle damage image and vehicle model information of a damaged vehicle from a vehicle insurance accident report;
[0072] Step S102, performing damage identification based on the vehicle damage image to obtain component damage information;
[0073] Step S103, determining the structure of the damaged vehicle according to the vehicle model information to obtain vehicle structure information;
[0074] Step S104, performing damage prediction based on the vehicle structure information and component damage information to obtain vehicle damage information;
[0075] Step S105, determining a vehicle maintenance plan for the damaged vehicle based on the vehicle damage information;
[0076] Step S106, merging information based on the vehicle insurance accident report and the vehicle repair plan to obtain a target vehicle damage claim policy.
[0077] Steps S101 to S106 shown in the embodiment of the present application first obtain a vehicle damage image and vehicle model information of the damaged vehicle from a motor vehicle insurance accident report, then perform damage identification based on the vehicle damage image to obtain component damage information, and then determine the structure of the damaged vehicle based on the vehicle model information to obtain vehicle structure information, and finally perform damage prediction based on the vehicle structure information and component damage information, so as to achieve further prediction of potential vehicle damage based on damage on vehicle components identified from the damage image combined with the specific structure of the vehicle, and more accurately determine the damage to the vehicle, thereby overcoming the problem that the accuracy of vehicle damage identification is limited because the on-site images are often unable to capture every damaged component in detail; further, determine a vehicle repair plan for the damaged vehicle based on the vehicle damage information, and then merge information based on the motor vehicle insurance accident report and the vehicle repair plan, so as to improve the recognition capability of vehicle damaged components and obtain an accurate target vehicle damage claim policy.
[0078] In step S101 of some embodiments, the vehicle insurance accident report is a document used to record relevant details of the vehicle accident, including but not limited to the time and location of the accident, the situation of the vehicles involved, the preliminary cause of the accident, the vehicle damage image and the vehicle model information. The vehicle damage image is a photo taken at the scene of the accident reflecting the damage to the vehicle, which is used to show the damaged parts of the vehicle, such as scratches, dents or cracks. The vehicle model information is the specific model of the vehicle, which is determined by the manufacturer when leaving the factory.
[0079] See also Figure 2 In some embodiments, step S102 may include but is not limited to steps S201 to S203:
[0080] Step S201, extracting features from the vehicle damage image to obtain image damage features;
[0081] Step S202, performing feature conversion according to the image damage feature to obtain a text damage feature;
[0082] Step S203: perform damage identification based on the text damage features to obtain component damage information.
[0083] In steps S201 to S203 shown in the embodiment of the present application, feature extraction is performed based on the vehicle damage image to obtain image damage features, and then feature conversion is performed based on the image damage features to obtain text damage features, thereby converting a large number of details contained in the image into text-level features, thereby avoiding blur and misrecognition in image recognition, and at the same time avoiding the situation where vehicle texture detail features are ignored. Finally, damage recognition is performed based on the text damage features, thereby further refining and analyzing the damage on the vehicle through natural language processing, and obtaining more accurate component damage information.
[0084] See also Figure 3 In some embodiments, step S201 may include but is not limited to steps S301 to S303:
[0085] Step S301, performing damage location according to the vehicle damage image to obtain damage location information;
[0086] Step S302, performing damage degree assessment according to the damage location information to obtain damage assessment classification information;
[0087] Step S303: performing feature aggregation based on the damage location information and the damage assessment classification information to obtain image damage features.
[0088] In steps S301 to S303 shown in the embodiment of the present application, damage is located based on the vehicle damage image, and the specific damaged parts on the vehicle are marked to obtain damage location information. Then, the damage degree is assessed based on the damage location information to obtain damage assessment classification information, thereby improving the level of detail of damage identification. Finally, feature aggregation is performed based on the damage location information and the damage assessment classification information to generate comprehensive image damage features, thereby providing comprehensive feature information for subsequent damage prediction and improving the accuracy of damage prediction.
[0089] See also Figure 4 In some embodiments, step S301 may include but is not limited to steps S401 to S404:
[0090] Step S401, performing image convolution on the vehicle damage image to obtain a damage convolution feature map;
[0091] Step S402, performing sliding window frame selection on the damage convolution feature map to obtain an initial damage positioning frame;
[0092] Step S403, performing positioning frame prediction according to the initial damage positioning frame to obtain positioning frame prediction data; wherein the positioning frame prediction data is used to indicate whether the image in the initial damage positioning frame is a damaged component image or a background image;
[0093] Step S404: Screen the initial damage location frame according to the location frame prediction data to obtain damage location information.
[0094] In steps S401 to S404 shown in the embodiment of the present application, image convolution is performed on the vehicle damage image to obtain a damage convolution feature map, and then a sliding window frame selection is performed on the damage convolution feature map to obtain an initial damage positioning frame, and then a positioning frame prediction is performed based on the initial damage positioning frame to obtain positioning frame prediction data; wherein the positioning frame prediction data is used to indicate whether the image in the initial damage positioning frame is a damaged component image or a background image, and finally the initial damage positioning frame is screened based on the positioning frame prediction data to obtain damage positioning information, thereby accurately extracting the damage positioning information from the vehicle damage image.
[0095] In step S401 of some embodiments, image convolution is to convolve the vehicle damage image through a preset convolutional neural network to obtain a damage convolution feature map.
[0096] In step S402 of some embodiments, a pre-set sliding window is used to select the damage convolution feature map, and multiple damage positioning frames, i.e., initial damage positioning frames, are obtained. It should be noted that there are multiple preset sliding windows, and the sliding window size can be set as needed, such as a 2×2 or 1×3 sliding window. In one embodiment, the damage convolution feature map is selected by sliding the window through a Q-former model to obtain an initial damage positioning frame.
[0097] In step S403 of some embodiments, the positioning box prediction is used to predict whether the image in the initial damage positioning box belongs to an image of a damaged component, or a background image. The background image includes a non-component image and a normal component image, and the damaged component image is an image of a damaged component. The positioning box prediction is performed by inputting the positioning box into a preset fully connected layer and an activation layer. First, the positioning box prediction data is input into the fully connected layer, and then the output of the fully connected layer is input into the activation layer. The activation layer outputs data of [0,1]. When the data output by the activation layer is greater than or equal to 0.5, it is confirmed that the initial damage positioning box belongs to the damaged component image. When the output data of the activation layer is less than 0.5, it is confirmed that the initial damage positioning box belongs to the background image. It should be noted that the threshold for the output layer to judge whether it belongs to a damaged component image or a background image can be set as needed.
[0098] In step S404 of some embodiments, the screening is to screen the initial damage positioning frame through the positioning frame prediction data, obtain the damaged component image therein, and then perform non-maximum suppression processing on the damaged component image to reduce or eliminate frames with a lot of overlap, thereby ensuring that each damaged component has only one image frame, that is, the damage positioning information.
[0099] In step S302 of some embodiments, the damage location information is location information in the vehicle damage image, which is expressed as a tuple formed by multiple groups of four coordinate points. In one tuple, the four coordinate points form a box, and the box selects the image of the damaged part in the vehicle damage image. The image of the damaged part is input into a preset damage assessment classification model to obtain damage assessment classification information. The damage assessment classification information is divided into multiple levels, for example, the damage assessment classification information includes severe damage, moderate damage, slight damage and wear.
[0100] In step S303 of some embodiments, feature aggregation is performed on the damage location information and the damage assessment classification information, and the aggregation method is matrix splicing.
[0101] See also Figure 5 In some embodiments, step S202 includes but is not limited to steps S501 to S503:
[0102] Step S501, feature encoding the image damage features through a preset self-attention encoder to obtain encoded feature information;
[0103] Step S502, feature decoding the image damage feature information by a preset decoder to obtain decoded feature information;
[0104] Step S503, generating text from the decoded feature information by using a preset text generator to obtain text damage features.
[0105] In steps S501 to S503 shown in the embodiment of the present application, a preset self-attention encoder is used to perform feature encoding on the image damage feature to obtain encoded feature information, and then a preset decoder is used to perform feature decoding on the image damage feature information to obtain decoded feature information, and then a preset text generator is used to generate text for the decoded feature information to obtain text damage features, thereby converting the feature information of the image into text information that can contain a large amount of image details, avoiding ignoring the detailed information of the image when directly processing the image, thereby improving the recognition accuracy of damage recognition.
[0106] In step S501 of some embodiments, feature encoding is to input the image damage feature into a preset self-attention encoder to obtain encoded feature information. In one embodiment, the preset self-attention encoder is composed of the self-attention layer and encoding layer of the Transformer model, and the self-attention layer and the encoding layer are connected in sequence to form the preset self-attention encoder.
[0107] In step S502 of some embodiments, feature decoding is to input the encoded feature information into the image damage feature information to complete feature decoding and obtain decoded feature information. In one embodiment, the preset decoder is composed of a decoding layer of a Transformer model.
[0108] In step S503 of some embodiments, the preset text generator is composed of a pre-trained Bert model, which is used to convert the decoded feature information generated by the decoder into text damage features that focus on the description of the damaged parts and the degree of damage of the damaged parts. For example: the decoded feature information "obvious dent damage was found on the front bumper, about 10 cm wide and 2 cm deep, located to the left of the center of the bumper. The damage may be caused by a low-speed front-end collision" is generated by the preset text generator for the decoded feature information. The resulting text damage feature is: "Dedent damage was detected in the front bumper area, with a size of approximately 10x2 cm and a position to the left of the center." The decoded feature information is refined, thereby ignoring unnecessary information and providing more accurate text damage features for the follow-up.
[0109] In step S203 of some embodiments, the text damage feature is further refined to complete the damage recognition and obtain the component damage information. The damage recognition method is pattern matching, that is, the table information, that is, the component damage information, is extracted from the text according to the preset extraction mode. For example, the text damage feature is "a concave damage is detected in the front bumper area, the size is about 10x2 cm, and the position is on the left side of the center", and the component damage information is ("left side of the center of the front bumper", "concave damage", "10x2cm").
[0110] In step S103 of some embodiments, the vehicle construction information refers to the assembly method, connection method and component configuration data of each component of the vehicle. This includes specific information of the vehicle's power system, chassis, body structure, electrical system and other components, as well as connection information and connection strength information between various components. The structure of the damaged vehicle can be determined by querying a preset vehicle structure database, and the vehicle construction information can be queried based on the vehicle model information.
[0111] See also Figure 6In some embodiments, the vehicle construction information includes vehicle component information, component connection information, and connection strength information. Step S104 includes but is not limited to steps S601 to S603:
[0112] Step S601, performing vehicle modeling according to vehicle component information, component connection information and connection strength information to obtain an initial vehicle model;
[0113] Step S602, performing damage alignment on the initial vehicle model according to the component damage information to obtain a damaged vehicle model;
[0114] Step S603: perform damage prediction based on the damaged vehicle model to obtain vehicle damage information.
[0115] In steps S601 to S603 shown in the embodiment of the present application, vehicle modeling is performed based on vehicle component information, component connection information and connection strength information to obtain an initial vehicle model, and then damage alignment is performed on the initial vehicle model based on the component damage information to obtain a damaged vehicle model, and then damage prediction is performed based on the damaged vehicle model, thereby predicting the damage to the entire vehicle and obtaining vehicle damage information more accurately.
[0116] In step S601 of some embodiments, vehicle modeling is performed based on vehicle component information, component connection information and connection strength information. The modeling method is to establish a graph data structure, with the vehicle component information as the nodes in the graph data structure, the connection information between components as the edges in the nodes, and the connection strength information as the weights between the nodes and edges, to obtain an initial vehicle model.
[0117] In step S602 of some embodiments, damage alignment is performed on the initial vehicle model based on the component damage information. The components and damage degrees corresponding to the component damage information are obtained by damage alignment. The components are aligned according to the nodes in the initial vehicle model, i.e., the graph data structure. The nodes in the graph data structure are marked as damaged, and the damage degrees are assigned to the nodes.
[0118] See also Figure 7 In some embodiments, step S603 may include but is not limited to steps S701 to S703:
[0119] Step S701, performing component damage prediction according to a damaged vehicle model to obtain component damage prediction information;
[0120] Step S702, classifying the maintenance type according to the component damage prediction information to obtain the component maintenance category;
[0121] Step S703: Merge the component damage prediction information and the component repair category to obtain vehicle damage information.
[0122] In steps S701 to S703 shown in the embodiment of the present application, component damage prediction is performed based on the damaged vehicle model to obtain component damage prediction information, and then the maintenance type is classified based on the component damage prediction information to obtain the component maintenance category. Finally, the component damage prediction information and the component maintenance category are merged to obtain the final vehicle damage information, thereby completing the damage identification and damage prediction of the vehicle.
[0123] In step S701 of some embodiments, component damage prediction refers to predicting potential damage to the vehicle through a damaged vehicle model. In one embodiment, damage prediction is performed through graph neural networks (GNNs), and the damaged vehicle model is input into a pre-trained graph neural network for prediction to predict component damage prediction information. The component damage prediction information includes damaged component information obtained by damage identification and damaged components predicted based on the damaged component information, connection strength information and component connection information, as well as damage classification of each damaged component, such as severe damage, moderate damage, minor damage and wear.
[0124] In step S702 of some embodiments, the components are classified into maintenance types according to the damage classification of each damaged component, specifically by querying through a preset maintenance plan table, which includes ("carbon fiber cover", "wear", "replacement"), ("fender", "wear", "repair"). The component damage prediction information and the preset maintenance plan table are classified for maintenance to determine the maintenance category of each component.
[0125] In step S703 of some embodiments, the vehicle damage information can be obtained by merging the component damage prediction information and component maintenance categories of each component. It should be noted that the information merging method can be set as needed, and this application does not impose any specific restrictions.
[0126] In step S105 of some embodiments, the vehicle damage information includes component repair categories, and the repair locations of each component are determined based on the component repair categories and a preset repair location query table to obtain a vehicle repair plan.
[0127] In step S106 of some embodiments, the motor vehicle insurance accident report and the vehicle repair plan are merged to obtain the target motor vehicle insurance claim policy.
[0128] See also Figure 8 The embodiment of the present application also provides a vehicle damage claim policy generation device, which can implement the above-mentioned vehicle damage claim policy generation method, and the device includes:
[0129] The data acquisition module 801 is used to acquire the vehicle damage image and vehicle model information of the damaged vehicle from the vehicle insurance accident report;
[0130] The damage identification module 802 is used to perform damage identification based on the vehicle damage image to obtain component damage information;
[0131] A structure determination module 803 is used to determine the structure of the damaged vehicle according to the vehicle model information to obtain vehicle structure information;
[0132] A damage prediction module 804 is used to perform damage prediction based on vehicle structure information and component damage information to obtain vehicle damage information;
[0133] A solution determination module 805 is used to determine a vehicle maintenance solution for the damaged vehicle based on the vehicle damage information;
[0134] The information merging module 806 is used to merge the information according to the vehicle insurance accident report and the vehicle repair plan to obtain the target vehicle damage claim policy.
[0135] The specific implementation of the vehicle damage claim policy generation device is basically the same as the specific implementation of the above-mentioned vehicle damage claim policy generation method, and will not be repeated here.
[0136] The embodiment of the present application also provides an electronic device, the electronic device includes a memory and a processor, the memory stores a computer program, and the processor implements the above-mentioned vehicle damage claim insurance policy generation method when executing the computer program. The electronic device can be any smart terminal including a tablet computer, a car computer, etc.
[0137] See also Fig. 9 , Fig. 9 The hardware structure of an electronic device of another embodiment is illustrated, and the electronic device includes:
[0138] The processor 901 may be implemented by a general-purpose CPU (Central Processing Unit), a microprocessor, an application-specific integrated circuit (Application Specific Integrated Circuit, ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of the present application;
[0139] The memory 902 can be implemented in the form of a read-only memory (ROM), a static storage device, a dynamic storage device, or a random access memory (RAM). The memory 902 can store an operating system and other application programs. When the technical solution provided in the embodiment of this specification is implemented by software or firmware, the relevant program code is stored in the memory 902, and the processor 901 calls and executes the vehicle damage claim policy generation method of the embodiment of this application;
[0140] Input / output interface 903, used to implement information input and output;
[0141] Communication interface 904, used to realize communication interaction between the device and other devices, which can be realized through wired mode (such as USB, network cable, etc.) or wireless mode (such as mobile network, WI FI, Bluetooth, etc.);
[0142] A bus 905 that transmits information between various components of the device (e.g., the processor 901, the memory 902, the input / output interface 903, and the communication interface 904);
[0143] The processor 901 , the memory 902 , the input / output interface 903 and the communication interface 904 are connected to each other in communication within the device via a bus 905 .
[0144] An embodiment of the present application also provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, the above-mentioned vehicle damage claim policy generation method is implemented.
[0145] The memory, as a non-transient computer-readable storage medium, can be used to store non-transient software programs and non-transient computer executable programs. In addition, the memory may include a high-speed random access memory, and may also include a non-transient memory, such as at least one disk storage device, a flash memory device, or other non-transient solid-state storage device. In some embodiments, the memory may optionally include a memory remotely disposed relative to the processor, and these remote memories may be connected to the processor via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.
[0146] The vehicle damage claim policy generation method, vehicle damage claim policy generation device, electronic device and storage medium provided in the embodiments of the present application first obtain the vehicle damage image and vehicle model information of the damaged vehicle from the vehicle insurance accident report, then perform damage identification based on the vehicle damage image to obtain component damage information, and then determine the structure of the damaged vehicle based on the vehicle model information to obtain vehicle structure information, and finally perform damage prediction based on the vehicle structure information and component damage information, so as to achieve further prediction of potential vehicle damage based on the damage on vehicle components identified from the damage image combined with the specific structure of the vehicle, and more accurately determine the damage to the vehicle, thereby overcoming the problem that the accuracy of vehicle damage identification is limited because the on-site image is often unable to capture every damaged component in detail; further, determine the vehicle repair plan of the damaged vehicle based on the vehicle damage information, and then merge the information based on the vehicle insurance accident report and the vehicle repair plan, so as to improve the recognition ability of vehicle damaged components and obtain an accurate target vehicle damage claim policy.
[0147] The embodiments described in the embodiments of the present application are intended to more clearly illustrate the technical solutions of the embodiments of the present application and do not constitute a limitation on the technical solutions provided in the embodiments of the present application. Those skilled in the art will appreciate that with the evolution of technology and the emergence of new application scenarios, the technical solutions provided in the embodiments of the present application are also applicable to similar technical problems.
[0148] Those skilled in the art will appreciate that the technical solutions shown in the figures do not constitute a limitation on the embodiments of the present application, and may include more or fewer steps than shown in the figures, or a combination of certain steps, or different steps.
[0149] The device embodiments described above are merely illustrative, and the units described as separate components may or may not be physically separated, that is, they may be located in one place or distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
[0150] Those skilled in the art will appreciate that all or some of the steps in the methods disclosed above, and the functional modules / units in the systems and devices may be implemented as software, firmware, hardware, or a suitable combination thereof.
[0151] The terms "first", "second", "third", "fourth", etc. (if any) in the specification of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.
[0152] It should be understood that in the present application, "at least one (item)" means one or more, and "plurality" means two or more. "And / or" is used to describe the association relationship of associated objects, indicating that three relationships may exist. For example, "A and / or B" can mean: only A exists, only B exists, and A and B exist at the same time, where A and B can be singular or plural. The character " / " generally indicates that the objects associated before and after are in an "or" relationship. "At least one of the following" or similar expressions refers to any combination of these items, including any combination of single or plural items. For example, at least one of a, b or c can mean: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, c can be single or multiple.
[0153] In the several embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the above units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.
[0154] The units described above as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
[0155] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.
[0156] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including multiple instructions to enable a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of various embodiments of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (Read-Only Memory, referred to as ROM), random access memory (Random Access Memory, referred to as RAM), disk or optical disk and other media that can store programs.
[0157] The preferred embodiments of the present invention are described above with reference to the accompanying drawings, but the scope of the rights of the present invention is not limited thereto. Any modification, equivalent substitution and improvement made by a person skilled in the art without departing from the scope and essence of the present invention should be within the scope of the rights of the present invention.
Claims
1. A method for generating a vehicle damage claim policy, characterized in that: The method comprises: Obtaining vehicle damage images and vehicle model information of the damaged vehicle from the automobile insurance accident report; Perform damage identification according to the vehicle damage image to obtain component damage information; Determining the structure of the damaged vehicle according to the vehicle model information to obtain vehicle structure information; Perform damage prediction based on the vehicle structure information and the component damage information to obtain vehicle damage information; Determining a vehicle maintenance plan for the damaged vehicle based on the vehicle damage information; The information is merged according to the vehicle insurance accident report and the vehicle repair plan to obtain a target vehicle damage claim policy.
2. The method according to claim 1, characterized in that The step of performing damage identification according to the vehicle damage image to obtain component damage information includes: Performing feature extraction based on the vehicle damage image to obtain image damage features; Performing feature conversion according to the image damage feature to obtain a text damage feature; Damage identification is performed according to the text damage features to obtain the component damage information.
3. The method according to claim 2, characterized in that The extracting features according to the vehicle damage image to obtain image damage features includes: Perform damage location according to the vehicle damage image to obtain damage location information; Perform damage degree assessment according to the damage location information to obtain damage assessment classification information; Feature aggregation is performed according to the damage location information and the damage assessment classification information to obtain the image damage feature.
4. The method according to claim 3, characterized in that The performing damage location according to the vehicle damage image to obtain damage location information includes: Performing image convolution on the vehicle damage image to obtain a damage convolution feature map; Perform sliding window frame selection on the damage convolution feature map to obtain an initial damage positioning frame; Perform positioning frame prediction according to the initial damage positioning frame to obtain positioning frame prediction data; wherein the positioning frame prediction data is used to indicate whether the image in the initial damage positioning frame is a damaged component image or a background image; The initial damage positioning frame is screened according to the positioning frame prediction data to obtain the damage positioning information.
5. The method according to claim 2, characterized in that: The step of performing feature conversion according to the image damage feature to obtain the text damage feature includes: Performing feature encoding on the image damage feature by using a preset self-attention encoder to obtain encoded feature information; Decoding the image damage feature information by a preset decoder to obtain decoded feature information; The text damage feature is obtained by performing text generation on the decoded feature information through a preset text generator.
6. The method according to any one of claims 1 to 5, characterized in that: The vehicle construction information includes vehicle component information, component connection information, and connection strength information; the damage prediction is performed based on the vehicle construction information and the component damage information to obtain the vehicle damage information, including: Perform vehicle modeling according to the vehicle component information, the component connection information and the connection strength information to obtain an initial vehicle model; Performing damage alignment on the initial vehicle model according to the component damage information to obtain a damaged vehicle model; Damage prediction is performed according to the damaged vehicle model to obtain the vehicle damage information.
7. The method according to claim 6, characterized in that The performing damage prediction according to the damaged vehicle model to obtain the vehicle damage information includes: Performing component damage prediction according to the damaged vehicle model to obtain component damage prediction information; Classify the maintenance type according to the component damage prediction information to obtain the component maintenance category; The vehicle damage information is obtained by merging the component damage prediction information and the component maintenance category.
8. A vehicle damage claim insurance policy generation device, characterized in that: The device comprises: A data acquisition module is used to obtain a vehicle damage image and vehicle model information of a damaged vehicle from a motor vehicle insurance accident report; A damage identification module, used to perform damage identification based on the vehicle damage image to obtain component damage information; A structure determination module, used to determine the structure of the damaged vehicle according to the vehicle model information to obtain vehicle structure information; A damage prediction module, used for performing damage prediction based on the vehicle structure information and the component damage information to obtain vehicle damage information; A solution determination module, used to determine a vehicle maintenance solution for the damaged vehicle based on the vehicle damage information; The information merging module is used to merge the information according to the vehicle insurance accident report and the vehicle repair plan to obtain a target vehicle damage claim policy.
9. An electronic device, characterized in that: The electronic device includes a memory and a processor, the memory stores a computer program, and the processor implements the method for generating a vehicle damage claim policy according to any one of claims 1 to 7 when executing the computer program.
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 generating a vehicle damage claim policy according to any one of claims 1 to 7 is implemented.