Vehicle damage detection method and device, computer device and storage medium

Abstract

The embodiment of the application belongs to the field of artificial intelligence, and relates to a vehicle damage detection method, comprising the following steps: receiving a vehicle damage detection request triggered by a user; obtaining a vehicle damage picture and user information from the vehicle damage detection request; performing preset processing on the vehicle damage picture to obtain a target vehicle damage picture meeting a preset definition condition; obtaining a user level corresponding to the user information; obtaining a target recognition model corresponding to the user level from a plurality of preset recognition models; performing recognition processing on the vehicle damage picture based on the target recognition model to obtain a target damage part of a target vehicle corresponding to the target vehicle damage picture. The application also provides a vehicle damage detection device, a computer device and a storage medium. In addition, the application also relates to blockchain technology, and the target damage part can be stored in the blockchain. The application improves the recognition efficiency of the damage part of the vehicle and ensures the accuracy of the generated damage part result of the vehicle.

Description

Technical Field

[0001] This application relates to the field of artificial intelligence technology, and in particular to vehicle damage detection methods, devices, computer equipment, and storage media. Background Technology

[0002] With the increasing number of cars on the road, the density of vehicles in cities is also increasing, leading to a rise in traffic accidents. Determining the extent of damage to vehicles in traffic accidents is a crucial task for insurance companies. Currently, this is typically done manually by claims adjusters. This manual process is labor-intensive, inefficient, and prone to errors due to the personal biases of the assessors. Summary of the Invention

[0003] The purpose of this application is to provide a vehicle damage detection method, device, computer equipment, and storage medium to solve the technical problem that the existing method of determining the damaged parts of a damaged vehicle is usually done manually by damage assessors. This manual method is labor-intensive, inefficient, and prone to errors in judgment results due to the personal factors of the damage assessors.

[0004] To address the aforementioned technical problems, this application provides a vehicle damage detection method, employing the following technical solution:

[0005] Receive a vehicle damage detection request triggered by a user; wherein the vehicle damage detection request carries an image of the damaged vehicle and the user's user information;

[0006] Obtain the vehicle damage image and the user information from the vehicle damage detection request;

[0007] The vehicle damage image is subjected to preset processing to obtain a target vehicle damage image that meets preset clarity conditions.

[0008] Obtain the user level corresponding to the user information;

[0009] Obtain the target recognition model corresponding to the user level from a set of preset recognition models;

[0010] The target recognition model is used to identify and process the damaged vehicle image to obtain the target damage location of the target vehicle corresponding to the damaged vehicle image.

[0011] Furthermore, the step of identifying and processing the damaged vehicle image based on the target recognition model to obtain the target damage location of the target vehicle corresponding to the damaged vehicle image specifically includes:

[0012] Feature extraction is performed on the damaged vehicle images to obtain corresponding image features;

[0013] Generate a corresponding feature matrix based on the image features;

[0014] The feature matrix is ​​input into the target recognition model, the target recognition model performs recognition processing on the feature matrix, and outputs the recognition result corresponding to the damaged image of the target vehicle;

[0015] The identification result is used as the target damage location of the target vehicle.

[0016] Furthermore, after the step of identifying and processing the damaged vehicle image based on the target recognition model to obtain the target damage location of the target vehicle corresponding to the damaged vehicle image, the method further includes:

[0017] The damaged images of the target vehicle are processed based on a preset loss detection model to obtain the corresponding vehicle damage information.

[0018] Obtain the texture information of the target damaged area;

[0019] Based on the vehicle damage information and the texture information, the magnitude of the contact force on the target vehicle during the collision is determined;

[0020] The target spatial coordinates of the vehicle damage caused by the collision are determined based on the magnitude of the contact force.

[0021] The target damage spatial region of the target vehicle is determined based on the target spatial coordinates.

[0022] Furthermore, the step of determining the target spatial coordinates of the vehicle damage caused by the collision based on the magnitude of the contact force specifically includes:

[0023] Based on the magnitude of the contact force, the torque application time information corresponding to the magnitude of the contact force is determined;

[0024] Based on the torque application time information, a preset calculation formula is used to determine the force-bearing area of ​​the target vehicle caused by the collision.

[0025] The target spatial coordinates are determined based on the force-bearing region.

[0026] Furthermore, the step of determining the target damage spatial region of the target vehicle based on the target spatial coordinates specifically includes:

[0027] Obtain multiple pre-segmented spatial regions;

[0028] Based on the target spatial coordinates, a first spatial region corresponding to the target spatial coordinates is selected from all the spatial regions;

[0029] Obtain the first priority value of the first spatial region;

[0030] The target damage spatial region is determined from all the first spatial regions based on the first priority value.

[0031] Furthermore, the step of determining the target damage spatial region from all the first spatial regions based on the first priority value specifically includes:

[0032] Select the second priority value with the largest value from all first priority values;

[0033] Obtain the second spatial region corresponding to the second priority value from all the first spatial regions;

[0034] The second spatial region is taken as the target damage spatial region.

[0035] Furthermore, before the step of obtaining the target recognition model corresponding to the user level from a plurality of preset recognition models, the method further includes:

[0036] Acquire pre-collected training data; wherein, the training data includes training images of damaged parts of several vehicles, and damage part labels corresponding one-to-one with each training image.

[0037] Obtain the preset initial model;

[0038] The training images are used as input to the initial model, and the damage site labels corresponding to the training images are used as output to train the initial model, thereby obtaining the trained initial model.

[0039] Obtain the loss function of the trained initial model and determine whether the loss function satisfies the preset convergence condition;

[0040] If so, the trained initial model is used as the designated recognition model; wherein, the designated recognition model is any one of all the recognition models.

[0041] To address the aforementioned technical problems, this application also provides a vehicle damage detection device, which employs the following technical solution:

[0042] The receiving module is used to receive a vehicle damage detection request triggered by a user; wherein the vehicle damage detection request carries a picture of the vehicle damage and the user's user information;

[0043] The first acquisition module is used to acquire the vehicle damage image and the user information from the vehicle damage detection request;

[0044] The first processing module is used to perform preset processing on the vehicle damage image to obtain a target vehicle damage image that meets the preset clarity conditions.

[0045] The second acquisition module is used to acquire the user level corresponding to the user information;

[0046] The third acquisition module is used to acquire the target recognition model corresponding to the user level from a plurality of preset recognition models;

[0047] The second processing module is used to identify and process the damaged vehicle image based on the target recognition model to obtain the target damage part of the target vehicle corresponding to the damaged vehicle image.

[0048] To address the aforementioned technical problems, this application also provides a computer device that employs the following technical solution:

[0049] Receive a vehicle damage detection request triggered by a user; wherein the vehicle damage detection request carries an image of the damaged vehicle and the user's user information;

[0050] Obtain the vehicle damage image and the user information from the vehicle damage detection request;

[0051] The vehicle damage image is subjected to preset processing to obtain a target vehicle damage image that meets preset clarity conditions.

[0052] Obtain the user level corresponding to the user information;

[0053] Obtain the target recognition model corresponding to the user level from a set of preset recognition models;

[0054] The target recognition model is used to identify and process the damaged vehicle image to obtain the target damage location of the target vehicle corresponding to the damaged vehicle image.

[0055] To address the aforementioned technical problems, this application also provides a computer-readable storage medium, employing the technical solution described below:

[0056] Receive a vehicle damage detection request triggered by a user; wherein the vehicle damage detection request carries an image of the damaged vehicle and the user's user information;

[0057] Obtain the vehicle damage image and the user information from the vehicle damage detection request;

[0058] The vehicle damage image is subjected to preset processing to obtain a target vehicle damage image that meets preset clarity conditions.

[0059] Obtain the user level corresponding to the user information;

[0060] Obtain the target recognition model corresponding to the user level from a set of preset recognition models;

[0061] The target recognition model is used to identify and process the damaged vehicle image to obtain the target damage location of the target vehicle corresponding to the damaged vehicle image.

[0062] Compared with the prior art, the embodiments of this application have the following main advantages:

[0063] Upon receiving a vehicle damage detection request triggered by a user, this application first obtains a vehicle damage image and user information from the request. Then, it performs pre-processing on the vehicle damage image to obtain a target vehicle damage image that meets pre-defined clarity conditions. Next, it obtains the user level corresponding to the user information. Subsequently, it selects a target recognition model corresponding to the user level from multiple pre-defined recognition models. Finally, it performs recognition processing on the vehicle damage image based on the target recognition model to obtain the target damage location of the target vehicle corresponding to the target vehicle damage image. This application embodiment can utilize a target recognition model to achieve fast and accurate recognition processing of the damage location of the target vehicle corresponding to the vehicle damage image, improving the efficiency of target vehicle damage location recognition and ensuring the accuracy of the generated target vehicle damage location results. Attached Figure Description

[0064] To more clearly illustrate the solutions in this application, the accompanying drawings used in the description of the embodiments of this application will be briefly introduced below. Obviously, the accompanying drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0065] Figure 1 This is an exemplary system architecture diagram to which this application can be applied;

[0066] Figure 2 A flowchart of an embodiment of the vehicle damage detection method according to this application;

[0067] Figure 3 This is a schematic diagram of one embodiment of the vehicle damage detection device according to this application;

[0068] Figure 4 This is a schematic diagram of the structure of one embodiment of the computer device according to this application. Detailed Implementation

[0069] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains; the terminology used herein in the specification of the application is for the purpose of describing particular embodiments only and is not intended to be limiting of the application; the terms "comprising" and "having," and any variations thereof, in the specification, claims, and foregoing drawings of this application, are intended to cover non-exclusive inclusion. The terms "first," "second," etc., in the specification, claims, or foregoing drawings of this application are used to distinguish different objects, not to describe a particular order.

[0070] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0071] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings.

[0072] like Figure 1 As shown, system architecture 100 may include terminal devices 101, 102, and 103, a network 104, and a server 105. Network 104 serves as the medium for providing communication links between terminal devices 101, 102, and 103 and server 105. Network 104 may include various connection types, such as wired or wireless communication links, or fiber optic cables, etc.

[0073] Users can use terminal devices 101, 102, and 103 to interact with server 105 via network 104 to receive or send messages, etc. Various communication client applications can be installed on terminal devices 101, 102, and 103, such as web browser applications, shopping applications, search applications, instant messaging tools, email clients, social media platform software, etc.

[0074] Terminal devices 101, 102, and 103 can be various electronic devices with displays and support web browsing, including but not limited to smartphones, tablets, e-book readers, MP3 players (Moving Picture Experts Group Audio Layer III), MP4 players (Moving Picture Experts Group Audio Layer IV), laptops, and desktop computers, etc.

[0075] Server 105 can be a server that provides various services, such as a backend server that supports the pages displayed on terminal devices 101, 102, and 103.

[0076] It should be noted that the vehicle damage detection method provided in this application embodiment is generally executed by a server / terminal device, and correspondingly, the vehicle damage detection device is generally set in the server / terminal device.

[0077] It should be understood that Figure 1 The number of terminal devices, networks, and servers shown is merely illustrative. Depending on implementation needs, any number of terminal devices, networks, and servers can be included.

[0078] Continue to refer to Figure 2 A flowchart of an embodiment of the vehicle damage detection method according to this application is shown. The vehicle damage detection method includes the following steps:

[0079] Step S201: Receive a vehicle damage detection request triggered by a user; wherein the vehicle damage detection request carries a picture of the damaged vehicle and the user's user information.

[0080] In this embodiment, the vehicle damage detection method operates on electronic equipment (e.g., Figure 1 The server / terminal device shown can acquire images of vehicle damage and user information via wired or wireless connections. It should be noted that the aforementioned wireless connection methods may include, but are not limited to, 3G / 4G / 5G connections, Wi-Fi connections, Bluetooth connections, Wi-Fi connections, Zigbee connections, UWB (ultra-Width band) connections, and other currently known or future wireless connection methods. The aforementioned vehicle damage images include images of the damaged parts of the target vehicle, and the aforementioned user information may include the user's name or ID information.

[0081] Step S202: Obtain the vehicle damage image and the user information from the vehicle damage detection request.

[0082] In this embodiment, the vehicle damage detection request can be parsed to extract the vehicle damage image and user information.

[0083] Step S203: Perform preset processing on the vehicle damage image to obtain a target vehicle damage image that meets preset clarity conditions.

[0084] In this embodiment, the aforementioned preset processing can be image enhancement processing. The specific sharpness value corresponding to the preset sharpness condition is not limited and can be set according to actual business needs. By performing preset processing on the vehicle damage image, unclear image data can be avoided when inputting it into the recognition model, enabling the recognition model to perform more accurate recognition processing based on the target vehicle damage image that meets the preset sharpness condition.

[0085] Step S204: Obtain the user level corresponding to the user information.

[0086] In this embodiment, the user level corresponding to the user information can be retrieved by querying a preset level data table. This level data table is a database that pre-stores customer information for each customer, along with the corresponding customer level. User levels include advanced users, intermediate users, and ordinary users.

[0087] Step S205: Obtain the target recognition model corresponding to the user level from a set of preset recognition models.

[0088] In this embodiment, multiple recognition models are pre-trained based on different training data. These models are then categorized based on their recognition efficiency: excellent, good, and average. A correlation is established between the recognition model's efficiency and the customer's level. Specifically, advanced users correspond to excellent recognition models, intermediate users to good models, and average users to models with average efficiency. By assigning corresponding recognition models to different customer levels, services can be provided that match the user's level, improving the user experience and service quality for different users.

[0089] Step S206: Based on the target recognition model, the damaged vehicle image is processed to obtain the target damage location of the target vehicle corresponding to the damaged vehicle image.

[0090] In this embodiment, the specific implementation process of identifying and processing the damaged vehicle image based on the target recognition model to obtain the target damaged part of the target vehicle corresponding to the damaged vehicle image will be further described in detail in subsequent specific embodiments of this application, and will not be elaborated on here.

[0091] Upon receiving a vehicle damage detection request triggered by a user, this application first obtains a vehicle damage image and user information from the request. Then, it performs pre-processing on the vehicle damage image to obtain a target vehicle damage image that meets pre-defined clarity conditions. Next, it obtains the user level corresponding to the user information. Subsequently, it selects a target recognition model corresponding to the user level from multiple pre-defined recognition models. Finally, it performs recognition processing on the vehicle damage image based on the target recognition model to obtain the target damage location of the target vehicle corresponding to the target vehicle damage image. This application can utilize a target recognition model to achieve fast and accurate recognition processing of the damage location of the target vehicle corresponding to the vehicle damage image, improving the efficiency of target vehicle damage location recognition and ensuring the accuracy of the generated target vehicle damage location results.

[0092] In some alternative implementations, step S206 includes the following steps:

[0093] Feature extraction is performed on the damaged vehicle image to obtain the corresponding image features.

[0094] A corresponding feature matrix is ​​generated based on the image features.

[0095] The feature matrix is ​​input into the target recognition model, which then performs recognition processing on the feature matrix and outputs the recognition result corresponding to the damaged image of the target vehicle.

[0096] In this embodiment, the target recognition model described above can be obtained by training a deep learning model based on pre-collected training data.

[0097] The identification result is used as the target damage location of the target vehicle.

[0098] This application extracts corresponding image features from vehicle damage images, generates a corresponding feature matrix based on the image features, and then inputs the feature matrix into a target recognition model. The target recognition model then processes the feature matrix to identify the target damage location of the target vehicle, thereby achieving accurate identification of the damage location of the target vehicle and improving the efficiency of damage location identification.

[0099] In some optional implementations of this embodiment, after step S201, the electronic device may further perform the following steps:

[0100] The damaged images of the target vehicle are processed based on a preset loss detection model to obtain the corresponding vehicle damage information.

[0101] In this embodiment, the vehicle damage information may include the damage state and degree corresponding to the target damaged part inside the target vehicle. For example, damage information includes, but is not limited to: scratches, dents, tears, etc.; damage degree includes, but is not limited to: minor damage, moderate damage, severe damage, etc. The loss detection model can be generated by training a preset deep learning model based on pre-collected image training data and the corresponding damage information labels and damage degree labels. Furthermore, the training and generation of this loss detection model can refer to the training and generation process of the subsequent recognition model, which will not be elaborated upon here.

[0102] Obtain the texture information of the target damaged area.

[0103] In this embodiment, the texture information corresponding to the target damaged area can be obtained by querying the vehicle information of the target vehicle.

[0104] Based on the vehicle damage information and the texture information, the magnitude of the contact force of the target vehicle during the collision is determined.

[0105] In this embodiment, the magnitude of the contact force of the target vehicle during a collision can be deduced using the aforementioned vehicle damage information and texture information, based on the original test data provided by different vehicle models. The aforementioned original test data includes the correlation between vehicle damage information, component texture information, and the magnitude of the contact force for different vehicle models.

[0106] The target spatial coordinates of the vehicle damage caused by the collision are determined based on the magnitude of the contact force.

[0107] In this embodiment, the specific implementation process of determining the target spatial coordinates of the vehicle damage caused by the collision based on the magnitude of the contact force will be further described in detail in subsequent specific embodiments of this application, and will not be elaborated on here.

[0108] The target damage spatial region of the target vehicle is determined based on the target spatial coordinates.

[0109] In this embodiment, the specific implementation process of determining the target damage space area of ​​the target vehicle based on the target spatial coordinates will be further described in detail in subsequent specific embodiments of this application, and will not be elaborated on here.

[0110] After identifying the target damage area of ​​the target vehicle based on the target recognition model, this application intelligently processes the damaged image of the target vehicle based on a preset loss detection model to obtain the corresponding vehicle damage information and the texture information of the target damage area. Then, based on the vehicle damage information and texture information, it determines the magnitude of the contact force of the target vehicle during the collision, and determines the target spatial coordinates of the vehicle damage caused by the collision based on the magnitude of the contact force. Furthermore, it determines the target damage spatial area of ​​the target vehicle based on the target spatial coordinates, so as to realize the range of the vehicle's hidden damage based on the loss detection model and the collision mechanics related to the magnitude of the contact force, thereby accurately determining the final damage spatial area of ​​the target vehicle. This further improves the accuracy of identifying the external damage of the target vehicle, thus facilitating subsequent matching and repair processing of the target vehicle based on the target damage spatial area.

[0111] In some optional implementations, the step of determining the target spatial coordinates of the vehicle damage caused by the collision based on the magnitude of the contact force includes the following steps:

[0112] Based on the magnitude of the contact force, the torque application time information corresponding to the magnitude of the contact force is determined.

[0113] In this embodiment, the torque application time information includes the cumulative time t1->t2 of the torque application corresponding to the magnitude of the contact force.

[0114] Based on the torque application time information, a preset calculation formula is used to determine the force-bearing area of ​​the target vehicle caused by the collision.

[0115] In this embodiment, the aforementioned preset calculation formula is the calculation formula corresponding to the impulse moment principle. For a particle, the impulse moment equals the product of the torque and the time of torque action, i.e., impulse moment ΔL = M·Δt. For a system of particles, since internal torques can cancel each other out, we can obtain ΔL = (M_external + M_internal)·Δt = (M_external + 0)·Δt = M_external·Δt. During the time interval t1->t2, the impulse moment experienced by a particle or system of particles is the accumulation of impulse moments during this time, i.e., ΔL_total = ∑ΔL = ∑M_external·Δt, where ΔL is a vector, M_external is the external torque, M_internal is the internal torque, and the direction is the same as M_external, with units of N·m·s, and Δt = t2 - t1. After obtaining the total impulse moment ΔL, the force area caused by the collision of the target vehicle can be calculated based on collision mechanics.

[0116] The target spatial coordinates are determined based on the force-bearing region.

[0117] In this embodiment, after obtaining the stress area, the stress area is converted into the corresponding target spatial coordinates through the conversion relationship between the area and spatial coordinates.

[0118] This application determines the torque application time information corresponding to the contact force based on the magnitude of the contact force, and then uses a preset calculation formula based on the torque application time information to determine the force area of ​​the target vehicle caused by the collision. Subsequently, the target spatial coordinates are determined based on the force area. This enables the rapid and accurate determination of the target spatial coordinates of the vehicle damage caused by the collision. This facilitates the subsequent determination of the target damage spatial area of ​​the target vehicle based on the obtained target spatial coordinates, thereby further improving the accuracy of external damage identification of the target vehicle. This also makes it easier to perform matching repair processing on the target vehicle based on the target damage spatial area.

[0119] In some optional implementations, the step of determining the target damage spatial region of the target vehicle based on the target spatial coordinates includes the following steps:

[0120] Obtain multiple pre-divided spatial regions.

[0121] In this embodiment, the current vehicle can be divided into 33 spatial regions, specifically including front left, center left, rear left, front right, center right, rear right, center front, center center, center rear, center left, center left, center left, center right, center right, center right, center front, center rear, front left, rear left, front left, rear left, front right, rear right, left side, right side, center transverse, front, rear, center longitudinal, front left, rear left, front right, rear right, left side, right side, front, and rear.

[0122] Based on the target spatial coordinates, a first spatial region corresponding to the target spatial coordinates is selected from all the spatial regions.

[0123] In this embodiment, there is a correspondence between spatial coordinates and spatial regions. Based on the correspondence between spatial coordinates and spatial regions, a first spatial region corresponding to the target spatial coordinates can be selected from all the spatial regions.

[0124] Obtain the first priority value of the first spatial region.

[0125] In this embodiment, historical accident cases can be analyzed from multiple dimensions, such as vehicle type, region, and major accidents, based on big data, to prioritize and weight all spatial regions. For example, priority weighting can be based on the damage results of each spatial region in each accident. If spatial region A has been damaged x times in all historical accident cases, then the priority value of spatial region A is x.

[0126] The target damage spatial region is determined from all the first spatial regions based on the first priority value.

[0127] In this embodiment, the specific implementation process of determining the target damage spatial region from all the first spatial regions based on the first priority value will be further described in detail in subsequent specific embodiments of this application, and will not be elaborated on here.

[0128] This application obtains multiple pre-segmented spatial regions, then filters out the first spatial region corresponding to the target spatial coordinates from all spatial regions based on the target spatial coordinates, and subsequently obtains the first priority value of the first spatial region. Based on the first priority value, the target damage spatial region can be accurately determined from all the first spatial regions, thereby further improving the accuracy of identifying external damage to the target vehicle. This is beneficial for subsequent matching and repair processing of the target vehicle based on the target damage spatial region.

[0129] In some optional implementations of this embodiment, the step of determining the target damage spatial region from all the first spatial regions based on the first priority value includes the following steps:

[0130] Select the second priority value with the largest value from all first priority values.

[0131] Obtain the second spatial region corresponding to the second priority value from all the first spatial regions.

[0132] In this embodiment, the highest first priority value, i.e. the second priority value mentioned above, refers to the spatial region with the highest priority, i.e., the second spatial region mentioned above.

[0133] The second spatial region is taken as the target damage spatial region.

[0134] This application improves the accuracy of identifying external damage to the target vehicle by selecting the second priority value with the largest value from all first priority values, then obtaining the second spatial region corresponding to the second priority value from all first spatial regions, and using the second spatial region as the target damage spatial region. In other words, the spatial region with the highest priority in the first spatial region is used as the final target damage spatial region of the target vehicle. This is beneficial for subsequent matching and repair processing of the target vehicle based on the target damage spatial region.

[0135] In some optional implementations of this embodiment, before step S205, the electronic device may further perform the following steps:

[0136] Acquire pre-collected training data; wherein, the training data includes training images of damaged parts of several vehicles, and damage part labels corresponding one-to-one with each training image.

[0137] Obtain the preset initial model.

[0138] In this embodiment, the initial model can be any deep learning model, such as decision tree, random forest, GBDT, XGB, etc.

[0139] The training images are used as input to the initial model, and the damage site labels corresponding to the training images are used as output to train the initial model, thereby obtaining the trained initial model.

[0140] Obtain the loss function of the initial model after training, and determine whether the loss function satisfies the preset convergence condition.

[0141] In this embodiment, the above convergence conditions are not specifically limited and can be set according to actual business usage needs.

[0142] If so, the trained initial model is used as the designated recognition model; wherein, the designated recognition model is any one of all the recognition models.

[0143] This application trains an initial model based on pre-collected training data, which can quickly generate the required recognition model. This facilitates the subsequent rapid and accurate identification of the damaged parts of the target vehicle corresponding to the damaged vehicle image, thereby improving the efficiency of identifying the damaged parts of the target vehicle and ensuring the accuracy of the generated damage parts results.

[0144] It should be emphasized that, in order to further ensure the privacy and security of the aforementioned target damage areas, these target damage areas can also be stored in a blockchain node.

[0145] The blockchain referred to in this application is a novel application model of computer technologies such as distributed data storage, peer-to-peer transmission, consensus mechanisms, and encryption algorithms. Essentially, a blockchain is a decentralized database, a chain of data blocks linked together using cryptographic methods. Each data block contains information about a batch of network transactions, used to verify the validity of the information (anti-counterfeiting) and generate the next block. A blockchain can include an underlying blockchain platform, a platform product service layer, and an application service layer.

[0146] The embodiments of this 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 that knowledge to obtain optimal results.

[0147] Foundational technologies for artificial intelligence generally include sensors, dedicated AI chips, cloud computing, distributed storage, big data processing, operating / interactive systems, and mechatronics. AI software technologies mainly encompass computer vision, robotics, biometrics, speech processing, natural language processing, and machine learning / deep learning.

[0148] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by instructing related hardware through computer-readable instructions. These 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 methods described above. The aforementioned storage medium can be a non-volatile storage medium such as a magnetic disk, optical disk, or read-only memory (ROM), or random access memory (RAM).

[0149] It should be understood that although the steps in the flowcharts of the accompanying figures are shown sequentially as indicated by the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the accompanying figures may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily completed at the same time, but can be executed at different times, and their execution order is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the sub-steps or stages of other steps.

[0150] Further reference Figure 3 As a response to the above Figure 2 The present application provides an embodiment of a vehicle damage detection device, which is similar to the method shown. Figure 2 Corresponding to the method embodiments shown, this device can be specifically applied to various electronic devices.

[0151] like Figure 3As shown, the vehicle damage detection device 300 described in this embodiment includes: a receiving module 301, a first acquisition module 302, a first processing module 303, a second acquisition module 304, a third acquisition module 305, and a second processing module 306. Wherein:

[0152] The receiving module 301 is used to receive a vehicle damage detection request triggered by a user; wherein the vehicle damage detection request carries a picture of the vehicle damage and the user's user information;

[0153] The first acquisition module 302 is used to acquire the vehicle damage image and the user information from the vehicle damage detection request;

[0154] The first processing module 303 is used to perform preset processing on the vehicle damage image to obtain a target vehicle damage image that meets the preset clarity conditions.

[0155] The second acquisition module 304 is used to acquire the user level corresponding to the user information;

[0156] The third acquisition module 305 is used to acquire the target recognition model corresponding to the user level from a plurality of preset recognition models;

[0157] The second processing module 306 is used to perform recognition processing on the vehicle damage image based on the target recognition model to obtain the target damage part of the target vehicle corresponding to the target vehicle damage image.

[0158] In this embodiment, the operations performed by the above-mentioned modules or units correspond one-to-one with the steps of the vehicle damage detection method of the aforementioned embodiments, and will not be repeated here.

[0159] In some optional implementations of this embodiment, the second processing module 306 includes:

[0160] The extraction submodule is used to extract features from the damaged vehicle image to obtain the corresponding image features;

[0161] A generation submodule is used to generate a corresponding feature matrix based on the image features;

[0162] The processing submodule is used to input the feature matrix into the target recognition model, perform recognition processing on the feature matrix through the target recognition model, and output the recognition result corresponding to the damaged image of the target vehicle;

[0163] The first determining submodule is used to use the identification result as the target damage location of the target vehicle.

[0164] In this embodiment, the operations performed by the above-mentioned modules or units correspond one-to-one with the steps of the vehicle damage detection method of the aforementioned embodiments, and will not be repeated here.

[0165] In some optional implementations of this embodiment, the vehicle damage detection device further includes:

[0166] The third processing module is used to process the damaged image of the target vehicle based on a preset loss detection model to obtain the corresponding vehicle damage information.

[0167] The fourth acquisition module is used to acquire the texture information of the target damaged area;

[0168] The first determining module is used to determine the magnitude of the contact force of the target vehicle during the collision process based on the vehicle damage information and the texture information.

[0169] The second determining module is used to determine the target spatial coordinates of the vehicle damage caused by the collision based on the magnitude of the contact force.

[0170] The third determining module is used to determine the target damage spatial region of the target vehicle based on the target spatial coordinates.

[0171] In this embodiment, the operations performed by the above modules or units correspond one-to-one with the steps of the vehicle damage detection method of the aforementioned implementation method, and will not be repeated here.

[0172] In some optional implementations of this embodiment, the second determining module includes:

[0173] The second determining submodule is used to determine the torque application time information corresponding to the magnitude of the contact force based on the magnitude of the contact force.

[0174] The third determining submodule is used to determine the force-bearing area of ​​the target vehicle caused by the collision based on the torque application time information and by calling a preset calculation formula.

[0175] The fourth determination submodule is used to determine the target spatial coordinates based on the force-bearing area.

[0176] In this embodiment, the operations performed by the above-mentioned modules or units correspond one-to-one with the steps of the vehicle damage detection method of the aforementioned embodiments, and will not be repeated here.

[0177] In some optional implementations of this embodiment, the third determining module includes:

[0178] The first acquisition submodule is used to acquire multiple pre-divided spatial regions;

[0179] The filtering submodule is used to filter out the first spatial region corresponding to the target spatial coordinates from all the spatial regions based on the target spatial coordinates;

[0180] The second acquisition submodule is used to acquire the first priority value of the first spatial region;

[0181] The fifth determining submodule is used to determine the target damage spatial region from all the first spatial regions based on the first priority value.

[0182] In this embodiment, the operations performed by the above-mentioned modules or units correspond one-to-one with the steps of the vehicle damage detection method of the aforementioned embodiments, and will not be repeated here.

[0183] In some optional implementations of this embodiment, the fifth determining submodule includes:

[0184] The first acquisition unit is used for

[0185] A filtering unit is used to filter out the second priority value with the largest value from all the first priority values;

[0186] The second acquisition unit is used to acquire the second spatial region corresponding to the second priority value from all the first spatial regions;

[0187] A determining unit is used to identify the second spatial region as the target damage spatial region.

[0188] In this embodiment, the operations performed by the above-mentioned modules or units correspond one-to-one with the steps of the vehicle damage detection method of the aforementioned embodiments, and will not be repeated here.

[0189] In some optional implementations of this embodiment, the vehicle damage detection device further includes:

[0190] The fifth acquisition module is used to acquire the preset initial model;

[0191] The training module is used to take the training images as input to the initial model, take the damage site labels corresponding to the training images as output to the initial model, and train the initial model to obtain the trained initial model.

[0192] The judgment module is used to obtain the loss function of the trained initial model and determine whether the loss function satisfies the preset convergence condition.

[0193] The fourth determining module is used to, if so, use the trained initial model as the designated recognition model; wherein the designated recognition model is any one of all the recognition models.

[0194] In this embodiment, the operations performed by the above-mentioned modules or units correspond one-to-one with the steps of the vehicle damage detection method of the aforementioned embodiments, and will not be repeated here.

[0195] To address the aforementioned technical problems, embodiments of this application also provide a computer device. Please refer to [link / reference needed] for details. Figure 4 , Figure 4 This is a basic structural block diagram of the computer device in this embodiment.

[0196] The computer device 4 includes a memory 41, a processor 42, and a network interface 43 that are interconnected via a system bus. It should be noted that only the computer device 4 with components 41-43 is shown in the figure; however, it should be understood that it is not required to implement all the shown components, and more or fewer components can be implemented alternatively. Those skilled in the art will understand that the computer device described here is a device capable of automatically performing numerical calculations and / or information processing according to pre-set or stored instructions. Its hardware includes, but is not limited to, microprocessors, application-specific integrated circuits (ASICs), programmable gate arrays (FPGAs), digital digital processors (DSPs), embedded devices, etc.

[0197] The computer device can be a desktop computer, laptop, handheld computer, or cloud server, etc. The computer device can interact with the user via a keyboard, mouse, remote control, touchpad, or voice control.

[0198] The memory 41 includes at least one type of readable storage medium, including flash memory, hard disk, multimedia card, card-type memory (e.g., SD or DX memory), 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, magnetic disk, optical disk, etc. In some embodiments, the memory 41 may be an internal storage unit of the computer device 4, such as the hard disk or memory of the computer device 4. In other embodiments, the memory 41 may also be an external storage device of the computer device 4, such as a plug-in hard disk, smart memory card (SMC), secure digital card (SD) card, flash memory card, etc. of the computer device 4. Of course, the memory 41 may also include both the internal storage unit and the external storage device of the computer device 4. In this embodiment, the memory 41 is typically used to store the operating system and various application software installed on the computer device 4, such as computer-readable instructions for vehicle damage detection methods. In addition, the memory 41 can also be used to temporarily store various types of data that have been output or will be output.

[0199] In some embodiments, the processor 42 may be a central processing unit (CPU), controller, microcontroller, microprocessor, or other data processing chip. The processor 42 is typically used to control the overall operation of the computer device 4. In this embodiment, the processor 42 is used to execute computer-readable instructions stored in the memory 41 or to process data, for example, to execute computer-readable instructions for the vehicle damage detection method.

[0200] The network interface 43 may include a wireless network interface or a wired network interface, which is typically used to establish communication connections between the computer device 4 and other electronic devices.

[0201] Compared with the prior art, the embodiments of this application have the following main advantages:

[0202] In this embodiment, upon receiving a vehicle damage detection request triggered by a user, the application first obtains a vehicle damage image and user information from the request. Then, it performs pre-processing on the vehicle damage image to obtain a target vehicle damage image that meets pre-defined clarity conditions. Next, it obtains the user level corresponding to the user information. Subsequently, it obtains a target recognition model corresponding to the user level from multiple pre-defined recognition models. Finally, it performs recognition processing on the vehicle damage image based on the target recognition model to obtain the target damage location of the target vehicle corresponding to the target vehicle damage image. This embodiment can utilize a target recognition model to achieve fast and accurate recognition processing of the damage location of the target vehicle corresponding to the vehicle damage image, improving the efficiency of target vehicle damage location recognition and ensuring the accuracy of the generated target vehicle damage location results.

[0203] This application also provides another embodiment, namely, a computer-readable storage medium storing computer-readable instructions that can be executed by at least one processor to cause the at least one processor to perform the steps of the vehicle damage detection method described above.

[0204] Compared with the prior art, the embodiments of this application have the following main advantages:

[0205] In this embodiment, upon receiving a vehicle damage detection request triggered by a user, the application first obtains a vehicle damage image and user information from the request. Then, it performs pre-processing on the vehicle damage image to obtain a target vehicle damage image that meets pre-defined clarity conditions. Next, it obtains the user level corresponding to the user information. Subsequently, it obtains a target recognition model corresponding to the user level from multiple pre-defined recognition models. Finally, it performs recognition processing on the vehicle damage image based on the target recognition model to obtain the target damage location of the target vehicle corresponding to the target vehicle damage image. This embodiment can utilize a target recognition model to achieve fast and accurate recognition processing of the damage location of the target vehicle corresponding to the vehicle damage image, improving the efficiency of target vehicle damage location recognition and ensuring the accuracy of the generated target vehicle damage location results.

[0206] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), and includes several instructions to cause a terminal device (which may be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in the various embodiments of this application.

[0207] Obviously, the embodiments described above are only some embodiments of this application, not all embodiments. The accompanying drawings show preferred embodiments of this application, but do not limit the patent scope of this application. This application can be implemented in many different forms; rather, the purpose of providing these embodiments is to provide a more thorough and comprehensive understanding of the disclosure of this application. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing specific embodiments, or make equivalent substitutions for some of the technical features. Any equivalent structures made using the content of this application's specification and drawings, directly or indirectly applied to other related technical fields, are similarly within the scope of patent protection of this application.

Claims

1. A method for detecting vehicle damage, characterized in that, Includes the following steps: Receive a vehicle damage detection request triggered by a user; wherein the vehicle damage detection request carries an image of the damaged vehicle and the user's user information; Obtain the vehicle damage image and the user information from the vehicle damage detection request; The vehicle damage image is subjected to preset processing to obtain a target vehicle damage image that meets preset clarity conditions. Obtain the user level corresponding to the user information; Obtain the target recognition model corresponding to the user level from a set of preset recognition models; Based on the target recognition model, the damaged vehicle image is identified and processed to obtain the target damage part of the target vehicle corresponding to the damaged vehicle image. The step of identifying the damaged vehicle image based on the target recognition model to obtain the target damage location of the target vehicle corresponding to the damaged vehicle image further includes: The damaged images of the target vehicle are processed based on a preset loss detection model to obtain the corresponding vehicle damage information. Obtain the texture information of the target damaged area; Based on the vehicle damage information and the texture information, the magnitude of the contact force on the target vehicle during the collision is determined; The target spatial coordinates of the vehicle damage caused by the collision are determined based on the magnitude of the contact force. The target damage space area of ​​the target vehicle is determined based on the target spatial coordinates; The step of determining the target spatial coordinates of the vehicle damage caused by the collision based on the magnitude of the contact force specifically includes: Based on the magnitude of the contact force, the torque application time information corresponding to the magnitude of the contact force is determined; Based on the torque application time information, a preset calculation formula is used to determine the force-bearing area of ​​the target vehicle caused by the collision. The target spatial coordinates are determined based on the force-bearing region; The preset calculation formula is the calculation formula corresponding to the impulse moment principle; after obtaining the force area, the force area is transformed into the corresponding target spatial coordinates through the conversion relationship between the area and spatial coordinates.

2. The vehicle damage detection method according to claim 1, characterized in that, The step of identifying the damaged vehicle image based on the target recognition model to obtain the target damage location of the target vehicle corresponding to the damaged vehicle image specifically includes: Feature extraction is performed on the damaged vehicle images to obtain corresponding image features; Generate a corresponding feature matrix based on the image features; The feature matrix is ​​input into the target recognition model, the target recognition model performs recognition processing on the feature matrix, and outputs the recognition result corresponding to the damaged image of the target vehicle; The identification result is used as the target damage location of the target vehicle.

3. The vehicle damage detection method according to claim 1, characterized in that, The step of determining the target damage spatial region of the target vehicle based on the target spatial coordinates specifically includes: Obtain multiple pre-segmented spatial regions; Based on the target spatial coordinates, a first spatial region corresponding to the target spatial coordinates is selected from all the spatial regions; Obtain the first priority value of the first spatial region; The target damage spatial region is determined from all the first spatial regions based on the first priority value.

4. The vehicle damage detection method according to claim 3, characterized in that, The step of determining the target damage spatial region from all the first spatial regions based on the first priority value specifically includes: Select the second priority value with the largest value from all first priority values; Obtain the second spatial region corresponding to the second priority value from all the first spatial regions; The second spatial region is taken as the target damage spatial region.

5. The vehicle damage detection method according to claim 1, characterized in that, Before the step of obtaining the target recognition model corresponding to the user level from a plurality of preset recognition models, the method further includes: Acquire pre-collected training data; wherein, the training data includes training images of damaged parts of several vehicles, and damage part labels corresponding one-to-one with each training image. Obtain the preset initial model; The training images are used as input to the initial model, and the damage site labels corresponding to the training images are used as output to train the initial model, thereby obtaining the trained initial model. Obtain the loss function of the trained initial model and determine whether the loss function satisfies the preset convergence condition; If so, the trained initial model is used as the designated recognition model; wherein, the designated recognition model is any one of all the recognition models.

6. A vehicle damage detection device, characterized in that, include: The receiving module is used to receive a vehicle damage detection request triggered by a user; wherein the vehicle damage detection request carries a picture of the vehicle damage and the user's user information; The first acquisition module is used to acquire the vehicle damage image and the user information from the vehicle damage detection request; The first processing module is used to perform preset processing on the vehicle damage image to obtain a target vehicle damage image that meets the preset clarity conditions. The second acquisition module is used to acquire the user level corresponding to the user information; The third acquisition module is used to acquire the target recognition model corresponding to the user level from a plurality of preset recognition models; The second processing module is used to identify and process the damaged vehicle image based on the target recognition model to obtain the target damage part of the target vehicle corresponding to the damaged vehicle image. The vehicle damage detection device also includes: The third processing module is used to process the damaged image of the target vehicle based on a preset loss detection model to obtain the corresponding vehicle damage information. The fourth acquisition module is used to acquire the texture information of the target damaged area; The first determining module is used to determine the magnitude of the contact force of the target vehicle during the collision process based on the vehicle damage information and the texture information. The second determining module is used to determine the target spatial coordinates of the vehicle damage caused by the collision based on the magnitude of the contact force. The third determining module is used to determine the target damage space area of ​​the target vehicle based on the target space coordinates; The second determining module includes: The second determining submodule is used to determine the torque application time information corresponding to the magnitude of the contact force based on the magnitude of the contact force. The third determining submodule is used to determine the force-bearing area of ​​the target vehicle caused by the collision based on the torque application time information and by calling a preset calculation formula. The fourth determining submodule is used to determine the target spatial coordinates based on the force-bearing area; The preset calculation formula is the calculation formula corresponding to the impulse moment principle; after obtaining the force area, the force area is transformed into the corresponding target spatial coordinates through the conversion relationship between the area and spatial coordinates.

7. A computer device comprising a memory and a processor, the memory storing computer-readable instructions, wherein the processor, when executing the computer-readable instructions, implements the steps of the vehicle damage detection method as described in any one of claims 1 to 5.

8. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-readable instructions, which, when executed by a processor, implement the steps of the vehicle damage detection method as described in any one of claims 1 to 5.

Images