Vehicle damage detection method and device, computing equipment and computer program product

Abstract

The embodiment of the invention provides a vehicle damage detection method. The method comprises the following steps: acquiring historical driving data corresponding to a rented vehicle; analyzing the driving information indicated by the historical driving data to obtain early warning information indicating vehicle damage; further calling a first sensor associated with the rented vehicle based on the early warning information to carry out image acquisition so as to obtain first damage information; according to the first damage information, calling a second sensor associated with the rented vehicle to scan a vehicle structure so as to obtain second damage information; and the first damage information and the second damage information are combined to obtain damage detection information corresponding to the rented vehicle. Therefore, the multi-dimensional vehicle loss detection process is realized, the historical driving data is adopted to perform preliminary damage positioning, and comprehensive judgment is performed by combining the detection data of different dimensions according to the damage positioning, so that the uncertainty caused by the influence of a visual angle is avoided, and the accuracy of vehicle damage detection is improved.

Description

Technical Field

[0001] This specification relates to the field of computer application technologies. Specifically, it relates to the technology for detecting vehicle damage in the field of computer application technologies. More specifically, it relates to a method, device, computing device, and computer program product for detecting vehicle damage. Background Art

[0002] In vehicle rental services, vehicle damage assessment is a key link in ensuring operational cost control and user experience. Traditional vehicle damage assessment methods mainly rely on photos submitted by users.

[0003] However, the quality of photos is greatly affected by the shooting environment (such as light and angle), which may lead to misjudgment of the damage degree. In addition, the damage assessment process needs to be performed manually, and the results are easily interfered by subjective factors; thus affecting the accuracy of vehicle damage inspection. Summary of the Invention

[0004] Embodiments of this specification provide a method, device, computing device, and computer program product for detecting vehicle damage to improve the accuracy of vehicle damage inspection.

[0005] To achieve the above technical objectives, the embodiments of this specification provide the following technical solutions:

[0006] In a first aspect, an embodiment of this specification provides a method for detecting vehicle damage, including:

[0007] Responding to a return request for a rental vehicle, obtaining historical driving data corresponding to the rental vehicle;

[0008] Analyzing the driving information indicated by the historical driving data to obtain warning information indicating vehicle damage;

[0009] Based on the warning information, calling a first sensor associated with the rental vehicle to perform image acquisition to obtain first damage information;

[0010] According to the first damage information, calling a second sensor associated with the rental vehicle to perform vehicle structure scanning to obtain second damage information;

[0011] Combining the first damage information and the second damage information to obtain damage detection information corresponding to the rental vehicle.

[0012] Optionally, in some possible embodiments, the analyzing the driving information indicated by the historical driving data to obtain warning information indicating vehicle damage includes:

[0013] Analyzing the driving information indicated by the historical driving data to obtain vehicle speed change information;

[0014] Perform a collision determination based on the vehicle speed change information to obtain a collision warning;

[0015] Detect a collision event generated by a collision detection module configured in the rental vehicle;

[0016] Combine the collision warning and the collision event to determine warning information indicating vehicle damage.

[0017] Optionally, in some possible embodiments, the performing a collision determination based on the vehicle speed change information to obtain a collision warning includes:

[0018] Obtain a preset driving trajectory corresponding to the rental vehicle;

[0019] Perform a lateral acceleration detection based on the preset driving trajectory to obtain a lateral acceleration parameter;

[0020] If the lateral acceleration parameter is greater than a lateral threshold, generate the collision warning.

[0021] Optionally, in some possible embodiments, the parsing the driving information indicated by the historical driving data to obtain warning information indicating vehicle damage includes:

[0022] Parse the driving information indicated by the historical driving data to obtain violation information corresponding to the rental vehicle, where the violation information is determined based on the road where the rental vehicle is located;

[0023] Statistically analyze each abnormal operation indicated in the violation information to obtain violation statistical information;

[0024] Compare the violation statistical information with a violation threshold to obtain a dangerous driving parameter;

[0025] Generate warning information indicating vehicle damage based on the dangerous driving parameter.

[0026] Optionally, in some possible embodiments, the invoking a first sensor associated with the rental vehicle based on the warning information to perform image acquisition to obtain first damage information includes:

[0027] Determine a risk component corresponding to the rental vehicle based on the warning information;

[0028] Determine a target visual range corresponding to the risk component;

[0029] According to the target visual range, invoke a visual sensor associated with the rental vehicle to perform image acquisition to obtain the first damage information.

[0030] Optionally, in some possible embodiments, the step of invoking the visual sensor associated with the rental vehicle according to the target visual range for image acquisition to obtain the first damage information includes:

[0031] Trigger the rental vehicle to detect adjacent vehicles according to the target visual range;

[0032] Send an image assistance request to the detected adjacent vehicles, so that the adjacent vehicles determine the associated visual sensor to perform image acquisition on the target visual range according to the assistance request;

[0033] Obtain the captured images sent by the adjacent vehicles to obtain the first damage information.

[0034] Optionally, in some possible embodiments, the method further includes:

[0035] Obtain multiple captured images sent by multiple adjacent vehicles;

[0036] Based on the target visual range, review multiple captured images to update the first damage information.

[0037] Optionally, in some possible embodiments, the step of invoking the second sensor associated with the rental vehicle according to the first damage information to perform a vehicle structure scan to obtain the second damage information includes:

[0038] Send a radar assistance request to the adjacent vehicles around the rental vehicle according to the first damage information, so that the adjacent vehicles invoke the associated radar sensor to perform a vehicle structure scan on the rental vehicle;

[0039] Obtain the vehicle three-dimensional model scanned by the adjacent vehicles;

[0040] Perform a comparison of structural components according to the vehicle three-dimensional model to obtain the second damage information.

[0041] Optionally, in some possible embodiments, the method further includes:

[0042] Determine the damaged components indicated by the first damage information;

[0043] Determine the damaged structural components according to the second damage information;

[0044] Perform an association match on the damaged components and the damaged structural components to obtain association information;

[0045] Configure a confidence label for the second damage information based on the association information.

[0046] Optionally, in some possible embodiments, the method further includes:

[0047] Determine the vehicle component information indicated by the damage detection information;

[0048] In response to a viewing request for the rental vehicle, obtain the augmented reality scene corresponding to the viewing request;

[0049] Based on the vehicle component information, mark the damage of the rental vehicle in the augmented reality scene.

[0050] In a second aspect, an embodiment of the present specification provides a vehicle damage detection device, including:

[0051] An acquisition unit, configured to obtain the historical driving data corresponding to the rental vehicle in response to a return request for the vehicle;

[0052] An analysis unit, configured to analyze the driving information indicated by the historical driving data to obtain a warning information indicating vehicle damage;

[0053] A detection unit, configured to call a first sensor associated with the rental vehicle for image acquisition based on the warning information to obtain first damage information;

[0054] The detection unit is further configured to call a second sensor associated with the rental vehicle for vehicle structure scanning according to the first damage information to obtain second damage information;

[0055] The detection unit is further configured to combine the first damage information and the second damage information to obtain the damage detection information corresponding to the rental vehicle.

[0056] Optionally, in some possible embodiments, when the analysis unit analyzes the driving information indicated by the historical driving data to obtain a warning information indicating vehicle damage, the analysis unit analyzes the driving information indicated by the historical driving data to obtain vehicle speed change information; perform a collision judgment according to the vehicle speed change information to obtain a collision warning; detect a collision event generated by a collision detection module configured in the rental vehicle; combine the collision warning and the collision event to determine the warning information indicating vehicle damage.

[0057] Optionally, in some possible embodiments, when the analysis unit performs a collision judgment according to the vehicle speed change information to obtain a collision warning, the analysis unit obtains a preset driving trajectory corresponding to the rental vehicle; perform a lateral acceleration detection based on the preset driving trajectory to obtain a lateral acceleration parameter; if the lateral acceleration parameter is greater than a lateral threshold, generate the collision warning.

[0058] Optionally, in some possible embodiments, when parsing the driving information indicated by the historical driving data to obtain a warning information indicating vehicle damage, the parsing unit parses the driving information indicated by the historical driving data to obtain the violation information corresponding to the rental vehicle, where the violation information is determined based on the road where the rental vehicle is located; counts each abnormal operation indicated in the violation information to obtain violation statistics information; compares the violation statistics information with a violation threshold to obtain a dangerous driving parameter; and generates a warning information indicating vehicle damage based on the dangerous driving parameter.

[0059] Optionally, in some possible embodiments, when invoking the first sensor associated with the rental vehicle to collect images based on the warning information to obtain first damage information, the detection unit determines the risk components corresponding to the rental vehicle based on the warning information; determines the target visual range corresponding to the risk components; and invokes the visual sensor associated with the rental vehicle to collect images according to the target visual range to obtain the first damage information.

[0060] Optionally, in some possible embodiments, when invoking the visual sensor associated with the rental vehicle to collect images according to the target visual range to obtain the first damage information, the detection unit triggers the rental vehicle to detect adjacent vehicles according to the target visual range; sends an image assistance request to the detected adjacent vehicles, so that the adjacent vehicles determine the associated visual sensor to collect images of the target visual range according to the assistance request; and obtains the collected images sent by the adjacent vehicles to obtain the first damage information.

[0061] Optionally, in some possible embodiments, the detection unit is further configured to obtain multiple collected images sent by multiple adjacent vehicles; and based on the target visual range, review the multiple collected images to update the first damage information.

[0062] Optionally, in some possible embodiments, when invoking the second sensor associated with the rental vehicle to perform a vehicle structure scan based on the first damage information to obtain second damage information, the detection unit sends a radar assistance request to the adjacent vehicles around the rental vehicle according to the first damage information, so that the adjacent vehicles invoke the associated radar sensor to perform a vehicle structure scan on the rental vehicle; obtains the vehicle three-dimensional model scanned by the adjacent vehicles; and performs a structural component comparison according to the vehicle three-dimensional model to obtain the second damage information.

[0063] Optionally, in some possible embodiments, the detection unit is further configured to determine the damaged component indicated by the first damage information; determine the lost structural member according to the second damage information; perform a relevance match on the damaged component and the lost structural member to obtain association information; and configure a confidence label for the second damage information based on the association information.

[0064] Optionally, in some possible embodiments, the detection unit is further configured to determine the vehicle component information indicated by the damage detection information; in response to a viewing request for the rental vehicle, obtain the augmented reality scene corresponding to the viewing request; and perform damage marking on the rental vehicle in the augmented reality scene based on the vehicle component information.

[0065] In a third aspect, an embodiment of the present specification further provides a computing device, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the computer program, the vehicle damage detection method described above is implemented.

[0066] In a fourth aspect, an embodiment of the present specification further provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the vehicle damage detection method described above is implemented.

[0067] In a fifth aspect, an embodiment of the present specification provides a computer program product or a computer program. The computer program product includes a computer program, which can be stored in a computer-readable storage medium or in the cloud; a processor of the computer device reads the computer program, and when the processor executes the computer program, the steps of the vehicle damage detection method described above are implemented.

[0068] As can be seen from the above technical solution, the vehicle damage detection method provided by the embodiments of this specification obtains the historical driving data corresponding to the rental vehicle in response to the return request for the rental vehicle; then parses the driving information indicated by the historical driving data to obtain early warning information indicating vehicle damage; further calls the first sensor associated with the rental vehicle based on the early warning information for image acquisition to obtain the first damage information; and calls the second sensor associated with the rental vehicle based on the first damage information for vehicle structure scanning to obtain the second damage information; and then combines the first damage information and the second damage information to obtain the damage detection information corresponding to the rental vehicle. Thus, a multi-dimensional vehicle loss detection process is realized. The preliminary damage positioning is carried out through the historical driving data, reducing the scope of damage detection; and for the damage positioning, the first sensor is used to consider the appearance dimension, avoiding the uncertainty brought by the image quality and viewing angle, and further for the damage positioning, the second sensor is used to consider the structure dimension, making up for the lack of structural damage detection, so the accuracy of vehicle damage detection is improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0069] In order to more clearly illustrate the technical solutions in the embodiments of this specification or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings in the following description are only the embodiments of this specification. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained according to the provided drawings.

[0070] Figure 1 Schematic diagram of the application environment of the vehicle damage detection method provided by an embodiment of this specification;

[0071] Figure 2 Schematic diagram of the flow of a vehicle damage detection method provided by an embodiment of this specification;

[0072] Figure 3 Schematic diagram of the scenario of a vehicle damage detection method provided by an embodiment of this specification;

[0073] Figure 4 Schematic diagram of the scenario of another vehicle damage detection method provided by an embodiment of this specification;

[0074] Figure 5 Schematic diagram of the functional modules of the vehicle damage detection device provided by an embodiment of this specification;

[0075] Figure 6 Schematic diagram of the structure of a computing device provided by an embodiment of this specification. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0076] Unless otherwise defined, the technical terms or scientific terms used in the embodiments of this specification shall have the ordinary meanings understood by those of ordinary skill in the art to which this specification pertains. The terms "first", "second" and similar terms used in the embodiments of this specification do not denote any order, quantity or importance, but are only used to avoid confusion of components.

[0077] Unless the context otherwise requires, throughout this specification, "a plurality of" means "at least two", and "including" is interpreted as open and inclusive, that is, "including, but not limited to". In the description of the specification, the terms "one embodiment", "some embodiments", "exemplary embodiments", "examples", "specific examples" or "some examples" etc. are intended to indicate that specific features, structures, materials or characteristics related to the embodiment or example are included in at least one embodiment or example of this specification. The schematic representations of the above terms do not necessarily refer to the same embodiment or example.

[0078] Next, the technical solutions in the embodiments of this specification will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of this specification. Obviously, the described embodiments are only a part of the embodiments of this specification, rather than all of the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments in this specification without creative efforts shall fall within the scope of protection of this specification.

[0079] In the vehicle rental business, the damage assessment of vehicles is a key link to ensure operation cost control and user experience. The traditional vehicle damage assessment method mainly relies on the photos submitted by users.

[0080] However, the quality of the photos is greatly affected by the shooting environment (such as light, angle), which may lead to misjudgment of the damage degree. In addition, the damage assessment process needs to be carried out manually, and its result is easily interfered by subjective factors; thus affecting the accuracy of vehicle damage inspection.

[0081] To solve the above problems, the embodiments of this specification provide a vehicle damage detection system, and the vehicle damage detection method provided by the embodiments of this specification is applied to this vehicle damage detection system. This detection system realizes accurate and efficient vehicle damage identification by integrating historical driving data analysis and multi-perspective image acquisition, supplemented by sensing technology.

[0082] Specifically, this vehicle damage detection system may include Figure 1The operating environment formed by the client 110, the server 120, and the vehicle 130 therein. The client 110 communicates with the server 120 via a network. The client 110 is wirelessly connected to the vehicle 130. The vehicle 130 communicates with the server 120 via a network connection. Among them, the client 110 may be an electronic device with network access capabilities. Specifically, for example, the client 110 may be a desktop computer, a tablet computer, a laptop computer, a smart phone, a digital assistant, a smart wearable device, a shopping guide terminal, a television, a smart speaker, a microphone, etc. Among them, the smart wearable devices include but are not limited to smart bracelets, smart watches, smart glasses, smart helmets, smart necklaces, etc. Alternatively, the client 110 may also be software that can run on the electronic device. The server 120 may be an electronic device with certain computing and processing capabilities. It may have a network communication module, a processor, a memory, etc. Of course, the server 120 may also refer to the software running on the electronic device. The server 120 may also be a distributed server, which may be a system with multiple processors, memories, network communication modules, etc. operating in coordination. Alternatively, the server 120 may also be a cluster formed by several servers 120. Alternatively, with the development of science and technology, the server 120 may also be a new technical means capable of implementing the corresponding functions of the embodiments of the specification. For example, it may be a new form of "server" based on quantum computing.

[0083] Specifically, when performing damage detection based on the above detection system, in response to a return request for a rental vehicle, historical driving data corresponding to the rental vehicle is obtained; then, the driving information indicated by the historical driving data is parsed to obtain warning information indicating vehicle damage; further, based on the warning information, a first sensor associated with the rental vehicle is called to perform image acquisition to obtain first damage information; and based on the first damage information, a second sensor associated with the rental vehicle is called to perform a vehicle structure scan to obtain second damage information; and then, the first damage information and the second damage information are combined to obtain damage detection information corresponding to the rental vehicle. Thus, a multi-dimensional vehicle loss detection process is realized. Preliminary damage positioning is performed through historical driving data, reducing the scope of damage detection; and for damage positioning, a first sensor is used to consider the appearance dimension, avoiding the uncertainty caused by image quality and viewing angle effects, and further, for damage positioning, a second sensor is used to consider the structure dimension, making up for the lack of structural damage detection, so the accuracy of vehicle damage detection is improved.

[0084] Based on the above concept, the embodiments of the present specification provide a method for detecting vehicle damage. Next, the method for detecting vehicle damage provided by the embodiments of the present specification will be described exemplarily with reference to the accompanying drawings.

[0085] For application in Figure 1Taking the server in [the context] as an example, some embodiments of this specification will exemplarily illustrate the method for detecting damage to the leased vehicle, such as Figure 2 as shown Figure 2 is a schematic flowchart of a method for detecting vehicle damage provided by an embodiment of this specification; this vehicle damage detection method includes:

[0086] 201. In response to a return request for the leased vehicle, obtain the historical driving data corresponding to the leased vehicle.

[0087] In this embodiment, the return request is the request executed by the user after ending the car rental task. For example, when the vehicle reaches the designated return location, a return request from the user is received.

[0088] In addition, the historical driving data corresponding to the leased vehicle can be the historical driving data during the current rental period of the leased vehicle. This historical driving data includes the driving data recorded by various vehicle sensors or functional modules, such as vehicle speed, abnormal event detection, path information, environmental information, etc.

[0089] 202. Analyze the driving information indicated by the historical driving data to obtain early warning information indicating vehicle damage.

[0090] In this embodiment, the process of analyzing the driving information indicated by the historical driving data is the process of screening risk items, and thus early warning information indicating vehicle damage is generated based on the risk items; therefore, the early warning information can be used to determine whether the vehicle has collided or whether there is a situation of dangerous driving through the historical driving data, or it can be a combination of multiple risk items.

[0091] Specifically, for the composition of the early warning information, it can be a judgment of the collision possibility. That is, first analyze the driving information indicated by the historical driving data to obtain vehicle speed change information; then perform a collision judgment based on the vehicle speed change information to obtain a collision warning; and detect the collision events generated by the collision detection module configured in the leased vehicle; and then combine the collision warning and the collision events to determine the early warning information indicating vehicle damage.

[0092] Among them, the collision detection module has a collision detection function, which is a functional module in the vehicle-mounted system used to monitor whether the vehicle has a physical contact, and usually combines sensor data to judge whether an impact event has occurred. Therefore, for the above collision judgment process, it can be a comprehensive evaluation of whether events such as sudden braking, sudden reduction in vehicle speed, and activation of the vehicle's collision detection function have occurred during vehicle driving.

[0093] Further, since the vehicle can configure the corresponding driving trajectory according to the driving operation, collision determination can also be performed based on the lateral acceleration information at this time. That is, first obtain the preset driving trajectory corresponding to the rental vehicle; then perform lateral acceleration detection based on the preset driving trajectory to obtain the lateral acceleration parameter; if the lateral acceleration parameter is greater than the lateral threshold, a collision warning is generated. For example, when the vehicle does not drive according to the predetermined trajectory, it is detected that the lateral acceleration is greater than the lateral threshold (for example, 5m / s 2 ), a collision warning is generated.

[0094] It can be understood that collision determination based on the lateral speed information takes into account that vehicle accidents generally involve abnormal lateral movement, such as emergency lane changes, side scratches, etc.; that is, it simulates the operation of the user's possible sudden turning of the direction to avoid when an accident occurs to judge the possibility of collision at this time. Therefore, through the judgment of the lateral acceleration, the lateral collision scenario and the avoidance scenario that may have safety risks can be accurately identified, that is, representative collision prediction information can be obtained through this specific parameter.

[0095] In addition, the warning information can also be configured through dangerous driving behaviors. That is, first analyze the driving information indicated by the historical driving data to obtain the violation information corresponding to the rental vehicle. This violation information is determined based on the road where the rental vehicle is located. That is, the regulations for different roads are different. For example, a vehicle speed of 100m / s is normal on the highway, but it is speeding on the urban road and there are safety risks; then count each abnormal operation indicated in the violation information to obtain the violation statistical information, such as the time and number of times the vehicle speeds, the number of times the vehicle violates lane changes, etc.; and compare the violation statistical information with the violation threshold to obtain the dangerous driving parameter; and then generate a warning information indicating vehicle damage based on the dangerous driving parameter. That is, judge whether the time and number of times of the vehicle speeding meet the threshold conditions, whether the number of times of the vehicle violating lane changes meets the threshold conditions, and other abnormal operations. The higher the degree of exceeding the violation threshold, the higher the credibility of the warning information.

[0096] Through the above analysis process, the process of intelligent warning is realized, that is, by analyzing the historical driving data, potential risk points are predicted in advance, and the inspection intensity is strengthened in a targeted manner.

[0097] 203. Call the first sensor associated with the rental vehicle based on the warning information to perform image acquisition to obtain the first damage information.

[0098] In this embodiment, the first sensor is used to detect the appearance of the vehicle, and it can be a visual sensor, that is, a sensor using optical imaging. It can be configured with a visible light camera, such as a monocular camera, a binocular / multi - camera, and an infrared camera, etc.

[0099] It can be understood that the process of invoking the visual sensor associated with the rental vehicle based on the warning information for image acquisition is a targeted detection process for possible damage risk items. Therefore, the risk components corresponding to the rental vehicle can be determined first based on the warning information. For example, if the warning information indicates a collision risk, the risk component is the front bumper of the vehicle. Then, the target visual range corresponding to the risk component is determined. That is, since the sizes and shapes of components of different vehicle models are different, the corresponding visual range can be marked at this time. And the visual sensor associated with the rental vehicle is invoked according to the target visual range for image acquisition to obtain the first damage information. The invocation process of this visual sensor can be a combination of one or more. For example, image acquisition is performed on the target visual range from different perspectives. Specifically, this visual sensor can be a sensor with visual sensing such as a camera.

[0100] In a possible scenario, the visual sensor associated with the rental vehicle can be the camera of the vehicle itself or the camera of other rental vehicles. That is, considering that there are blind spots in the vehicle's own field of vision and it may be impossible to complete image acquisition completely. At this time, due to the characteristic of unified management of rental vehicles, an interface for mutual communication can be configured between rental vehicles. Therefore, other rental vehicles around the vehicle can be detected and a request for adjacent vehicle assistance in acquisition can be sent. As Figure 3 shown, Figure 3 This is a schematic diagram of the scenario of a vehicle damage detection method provided by an embodiment of this specification. That is, other rental vehicles around the vehicle are detected. When there are other rental vehicles to be rented, pictures of the parts that cannot be acquired on this vehicle are acquired through the cameras of other rental vehicles to be rented. Damage detection is performed based on all the pictures acquired above. If damage is detected, such as surface damage such as dents and scratches is detected.

[0101] Specifically, for the interaction process between rental vehicles, when it is determined that a specific component cannot be completely imaged, the rental vehicle can be triggered to detect adjacent vehicles according to the target visual range. And an image assistance request is sent to the detected adjacent vehicles so that the adjacent vehicles can determine the associated visual sensor to perform image acquisition on the target visual range according to the assistance request. Furthermore, the acquired images sent by the adjacent vehicles are obtained to get the first damage information, that is, surface damage such as dents and scratches is detected.

[0102] Optionally, since rental vehicles are concentrated at the return location, there may be multiple adjacent vehicles at this time. In order to further reduce the influence of light and perspective on the acquired images, multiple acquired images sent by multiple adjacent vehicles can be obtained. Then, the multiple acquired images are reviewed based on the target visual range to update the first damage information, thereby improving the accuracy of damage detection.

[0103] In addition, specific service stations can be set up with professional tools for in-depth inspections, such as customized image acquisition parking spaces, etc.

[0104] Specifically, for the process of equipping specific service stations, it can be achieved by statistically analyzing common exterior damage locations, setting up a damage heat map, and then making targeted multi-perspective configurations based on the damage heat map, thereby improving the usage efficiency of cameras while ensuring the accuracy of image acquisition.

[0105] Through the above-mentioned targeted image acquisition process based on warning information, comprehensive coverage of risk items is achieved. It utilizes multiple camera perspectives plus remote scanning technology to ensure that every corner can be carefully inspected. On the one hand, it reduces the loss assessment error caused by poor picture quality, and on the other hand, it solves the problem of supplementing information on vehicle body parts that are difficult to capture from a single perspective.

[0106] 204. Invoke a second sensor associated with the rental vehicle according to the first damage information to perform a vehicle structure scan to obtain second damage information.

[0107] In this embodiment, the second sensor is used to detect the structure of the vehicle. It can be a radar sensor. Considering that there may be parts of the vehicle damage that are not visible on the exterior, such as structural damage, a radar sensor can be combined for detection at this time. The radar sensor includes: millimeter-wave radar or lidar. Among them, millimeter-wave radar is a technology that uses millimeter-wave band electromagnetic waves for detection and can accurately measure information such as the distance and speed of the target, and has wide applications in the field of autonomous driving. Lidar is a technology that obtains three-dimensional information of the surrounding environment by emitting laser beams and receiving the reflected signals, and has high-precision and long-distance detection capabilities.

[0108] Specifically, for the process of performing a vehicle structure scan through a radar sensor to obtain second damage information, first, a radar assistance request can be sent to adjacent vehicles around the rental vehicle according to the first damage information, so that the adjacent vehicles can invoke the associated radar sensor to perform a vehicle structure scan on the rental vehicle; then, obtain the three-dimensional model of the vehicle scanned by the adjacent vehicles; and perform a comparison of structural components based on the three-dimensional model of the vehicle to obtain second damage information.

[0109] Among them, the first damage information has a guiding effect on the second damage information, that is, there is a correlation between vehicle damages. The internal structural damage (second damage information) can be targeted by the damage on the surface appearance (first damage information). That is, the millimeter-wave radar (or lidar and other devices) of other vehicles to be rented is used to scan this vehicle and construct a three-dimensional model, and it is judged whether there is structural damage (such as diamond damage, vehicle body distortion) to this vehicle through the three-dimensional model.

[0110] Optionally, for the process of scanning the vehicle by the millimeter-wave radar of other vehicles to be rented and constructing a three-dimensional model, and determining whether there is structural damage to the vehicle through the three-dimensional model; the auxiliary vehicle can be determined through the interaction between the vehicle and the surrounding vehicles; that is, in order to improve the accuracy of structural damage detection, vehicles close to the side of structural damage are preferably selected for scanning.

[0111] Specifically, for the process of determining the auxiliary vehicle, the possible appearance damage can be first determined according to the first damage information, and then the possible damage mode can be determined through the appearance damage, so as to combine the damage mode and the location of the appearance damage to determine the best scanning perspective. For example, if the first damage information indicates that the front mouth is sunken, the possible damage mode is rear-end collision, which may cause damage to the front bumper. Therefore, the best scanning perspective is the front perspective; at this time, the detection of the auxiliary vehicle is carried out, and the vehicle in the front perspective is called for auxiliary scanning, so as to improve the accuracy of structural scanning.

[0112] Furthermore, through the correlation relationship between the first damage information and the second damage information, a confidence label can also be configured for the second damage information, that is, the marking of the confidence level; first, determine the damaged component indicated by the first damage information; then determine the damaged structural component according to the second damage information; and perform correlation matching on the damaged component and the damaged structural component to obtain the correlation information; furthermore, based on the correlation information, a confidence label is configured for the second damage information. Thus, the damage in the previous rental cycle is prevented from being introduced into the current rental cycle, and the accuracy of damage detection is improved.

[0113] Through the above process of radar detection, the comprehensive evaluation of the vehicle state based on multi-source data is realized, and the comprehensiveness and accuracy of damage assessment are improved.

[0114] 205. Combine the first damage information and the second damage information to obtain the damage detection information corresponding to the rental vehicle.

[0115] In this embodiment, the damage detection information is the set of the first damage information and the second damage information. All damage data can be uploaded to the platform for compensation appraisal. If there is no damage, the vehicle return task is executed.

[0116] Specifically, for the processing of compensation appraisal, the AI model can be continuously trained to better understand the characteristics of different types of damage, so as to improve the recognition rate and classification accuracy.

[0117] In a possible scenario, for the process of training the AI model to accurately identify the damage characteristics of different vehicle types, it can be carried out based on three dimensions: data construction, knowledge fusion, and model optimization.

[0118] First, construct a vehicle damage knowledge graph; this process uses multi-modal data fusion: damaged images (more than 100,000) of more than 20 vehicle models including sedans, trucks, SUVs, etc., lidar point clouds (accuracy ±2 cm), and maintenance record texts (such as insurance loss assessment reports) can be collected to construct a cross-modal alignment dataset. The mapping relationship between the local damage area of the image (such as a door dent) and the text description ("sheet metal deformation depth > 5 cm") is established through the attention mechanism (Cross-Modality Transformer). And structural mechanics feature injection is carried out, that is, key structural parameters (such as the A-pillar thickness of 2.5 mm and the bumper material of PP+GF30) are extracted based on the vehicle CAD model to construct the node features of the graph neural network (GNN), enabling the model to understand the impact resistance thresholds of different parts (such as when the deformation of the engine compartment exceeds 8 mm, structural correction is required).

[0119] Then, perform decoupled learning of damage features, that is, a hierarchical decoding architecture can be configured, using a cascaded U-Net structure. In the first stage, the damage area is located (IoU > 0.85), in the second stage, the damage morphology (scratches, dents, fractures) is analyzed through deformable convolution (Deformable Conv), and in the third stage, the damage type and repair level are classified in combination with the vehicle model library (such as the VMMR dataset).

[0120] Furthermore, for the above model architecture, a contrastive learning strategy is adopted, and a triplet loss function is designed to bring closer samples of the same damage degree (such as a 1 cm vs 1.2 cm dent) and push away different damage types (dent vs glass crack), forming a damage feature metric standard in the latent space (F1-score is increased by 12%).

[0121] In addition, based on the above model configuration, a physical enhancement training mechanism can also be carried out, that is, synthetic data generation is performed. The Blender physical engine is used to simulate different collision scenarios (speed 20 - 80 km / h, angle 0 - 180°) to generate damage images with real mechanical parameters (PSNR > 32 dB), solving the problem of the long-tail distribution of real data (the coverage rate of rare damage types is increased to 95%). And domain adaptation optimization is configured, using CycleGAN for environmental migration such as rain, fog, and low light at night, and cooperating with the gradient reversal layer (GRL) to reduce the weather domain difference, so that the fluctuation of the damage recognition accuracy of the model under extreme conditions is controlled within ±3%.

[0122] After the above model configuration, dynamic evaluation and iteration can also be performed, that is, a quantitative evaluation system is defined with multi-level evaluation indicators: the basic layer (mIoU of pixel-level damage segmentation), the functional layer (prediction error of maintenance cost < 15%), and the safety layer (false detection rate of structural damage < 0.1%). And an online learning framework is configured, and a federated learning system is deployed. New vehicle damage data (daily average increment of 500+ samples) is continuously collected through the 4S store terminal, and the Elastic Weight Consolidation (EWC) algorithm is used to prevent catastrophic forgetting, realizing monthly iterative update of the model.

[0123] Through the above multi-dimensional damage training process, an efficient damage recognition process can be achieved, and further combined with the above first damage information and second damage information for review and correction, improving the accuracy of damage detection.

[0124] In a possible scenario, combining the above damage detection process, Figure 4 the execution process shown Figure 4 is a schematic diagram of a scenario of another vehicle damage detection method provided for an embodiment of this specification; when the vehicle arrives at the designated return location and receives the user's return request, the vehicle's historical driving data is obtained, and it is judged whether the vehicle has collided or there is a situation of dangerous driving through the historical driving data. If not, the vehicle's camera is used to collect the body image, and vehicle damage detection is performed based on the body image; if it is judged that the vehicle may have collided or there is a situation of dangerous driving, first, the vehicle's camera is used to collect the body image, and other rental vehicles around the vehicle are detected. When there are other rental vehicles, pictures of the parts that cannot be collected on this vehicle are collected through the cameras of other rental vehicles, and damage detection is performed based on all the above collected pictures. If damage is detected (usually only dents and scratches can be detected), it is also necessary to scan this vehicle with the millimeter-wave radar (or lidar and other devices) of other rental vehicles and construct a three-dimensional model, and judge whether there is structural damage (such as diamond damage, body distortion) on this vehicle through the three-dimensional model. If there is also structural damage, all damage data is uploaded to the platform for compensation appraisal. If there is no damage, the return vehicle task is executed.

[0125] Through the above automated processing flow, human intervention is reduced, and efficiency and fairness are improved; the whole process is highly automated, reducing the uncertainty brought by manual participation, and at the same time accelerating the processing speed. And all operations are based on objective facts and scientific basis, ensuring that the rights and interests of both parties are not infringed.

[0126] Further, this embodiment can also be combined with augmented reality for assisting in damage assessment. That is, first, determine the vehicle component information indicated by the damage detection information; then, in response to a viewing request for the rental vehicle, obtain the augmented reality scene corresponding to the viewing request. This viewing request can be a status check when returning the vehicle (the damage detection information is the current data), or a status check when renting the vehicle (the damage detection information is the cumulative data); then, based on the vehicle component information, mark the damage on the rental vehicle in the augmented reality scene. Therefore, the user can view the vehicle condition in real time through the smartphone camera, and superimpose virtual marks to indicate the suspicious areas, so as to accurately understand the status of the rental vehicle.

[0127] In summary, in this embodiment, in response to a return request for a rental vehicle, historical driving data corresponding to the rental vehicle is obtained; then, the driving information indicated by the historical driving data is analyzed to obtain a warning information indicating vehicle damage; further, based on the warning information, a first sensor associated with the rental vehicle is called to collect images to obtain first damage information; and according to the first damage information, a second sensor associated with the rental vehicle is called to perform a vehicle structure scan to obtain second damage information; furthermore, the first damage information and the second damage information are combined to obtain damage detection information corresponding to the rental vehicle. Thus, a multi-dimensional vehicle loss detection process is realized. Preliminary damage positioning is performed through historical driving data, reducing the scope of damage detection; and for damage positioning, the first sensor is used to consider the appearance dimension, avoiding the uncertainty caused by image quality and viewing angle, and further, for damage positioning, the second sensor is used to consider the structure dimension, making up for the lack of structural damage detection, so the accuracy of vehicle damage detection is improved.

[0128] It should be noted that each of the multiple embodiments in this specification emphasizes the parts different from other embodiments. The embodiments can be mutually explained. Any combination of the multiple embodiments in this specification by those skilled in the art is covered by the disclosure scope of this specification based on general technical knowledge.

[0129] In an exemplary embodiment of this specification, a detection device 600 for vehicle damage is also provided, as Figure 5 shown, Figure 5 is a schematic diagram of the functional modules of the detection device for vehicle damage provided by an embodiment of this specification. The detection device 500 includes:

[0130] An acquisition unit 501, configured to obtain historical driving data corresponding to the rental vehicle in response to a return request for the vehicle;

[0131] An analysis unit 502, configured to analyze the driving information indicated by the historical driving data to obtain a warning information indicating vehicle damage;

[0132] A detection unit 503, configured to call a first sensor associated with the rental vehicle based on the warning information to collect images, so as to obtain first damage information;

[0133] The detection unit 503 is further configured to call a second sensor associated with the rental vehicle to perform a vehicle structure scan according to the first damage information, so as to obtain second damage information;

[0134] The detection unit 503 is further configured to combine the first damage information and the second damage information to obtain damage detection information corresponding to the rental vehicle.

[0135] Optionally, in some possible embodiments, when parsing the driving information indicated by the historical driving data to obtain a warning information indicating vehicle damage, the parsing unit 502 is configured to parse the driving information indicated by the historical driving data to obtain vehicle speed change information; perform a collision determination according to the vehicle speed change information to obtain a collision warning; detect a collision event generated by a collision detection module configured in the rental vehicle; and combine the collision warning and the collision event to determine a warning information indicating vehicle damage.

[0136] Optionally, in some possible embodiments, when performing a collision determination according to the vehicle speed change information to obtain a collision warning, the parsing unit 502 is configured to obtain a preset driving trajectory corresponding to the rental vehicle; perform a lateral acceleration detection based on the preset driving trajectory to obtain a lateral acceleration parameter; and if the lateral acceleration parameter is greater than a lateral threshold, generate the collision warning.

[0137] Optionally, in some possible embodiments, when parsing the driving information indicated by the historical driving data to obtain a warning information indicating vehicle damage, the parsing unit 502 is configured to parse the driving information indicated by the historical driving data to obtain violation information corresponding to the rental vehicle, where the violation information is determined based on a road where the rental vehicle is located; count each abnormal operation indicated in the violation information to obtain violation statistical information; compare the violation statistical information with a violation threshold to obtain a dangerous driving parameter; and generate a warning information indicating vehicle damage based on the dangerous driving parameter.

[0138] Optionally, in some possible embodiments, when the detection unit 503 is used to call the first sensor associated with the rental vehicle based on the warning information to collect images to obtain the first damage information, the detection unit 503 is configured to determine the risk components corresponding to the rental vehicle based on the warning information; determine the target visual range corresponding to the risk components; and call the visual sensor associated with the rental vehicle to collect images according to the target visual range to obtain the first damage information.

[0139] Optionally, in some possible embodiments, when the detection unit 503 is used to call the visual sensor associated with the rental vehicle to collect images according to the target visual range to obtain the first damage information, the detection unit 503 is configured to trigger the rental vehicle to detect adjacent vehicles according to the target visual range; send an image assistance request to the detected adjacent vehicles, so that the adjacent vehicles determine the associated visual sensors to collect images of the target visual range according to the assistance request; and obtain the collected images sent by the adjacent vehicles to obtain the first damage information.

[0140] Optionally, in some possible embodiments, the detection unit 503 is further configured to obtain a plurality of collected images sent by the plurality of adjacent vehicles; and perform a review on the plurality of collected images based on the target visual range to update the first damage information.

[0141] Optionally, in some possible embodiments, when the detection unit 503 is used to call the second sensor associated with the rental vehicle to perform a vehicle structure scan according to the first damage information to obtain the second damage information, the detection unit 503 is configured to send a radar assistance request to the adjacent vehicles around the rental vehicle according to the first damage information, so that the adjacent vehicles call the associated radar sensors to perform a vehicle structure scan on the rental vehicle; obtain the vehicle three-dimensional model scanned by the adjacent vehicles; and perform a structural component comparison according to the vehicle three-dimensional model to obtain the second damage information.

[0142] Optionally, in some possible embodiments, the detection unit 503 is further configured to determine the damaged components indicated by the first damage information; determine the damaged structural components according to the second damage information; perform an association matching on the damaged components and the damaged structural components to obtain association information; and configure a confidence label for the second damage information based on the association information.

[0143] Optionally, in some possible embodiments, the detection unit 503 is further configured to determine the vehicle component information indicated by the damage detection information; respond to a viewing request for the rental vehicle, and obtain the augmented reality scene corresponding to the viewing request; and perform damage marking on the rental vehicle in the augmented reality scene based on the vehicle component information.

[0144] Specifically, the processing unit and the interaction unit in this embodiment can correspond to physical components. For example, the processing unit can be a processing module such as a CPU, GPU, FPGA, etc., and the interaction unit can be an interaction module such as a screen, speaker, projector, etc. The selection of specific physical components can be any component and combination of components with the above functions, and the specific method depends on the actual scenario and is not limited here.

[0145] The above detection device obtains historical driving data corresponding to the rental vehicle by responding to the return request for the rental vehicle; then parses the driving information indicated by the historical driving data to obtain early warning information indicating vehicle damage; further calls the first sensor associated with the rental vehicle based on the early warning information for image acquisition to obtain first damage information; and calls the second sensor associated with the rental vehicle based on the first damage information for vehicle structure scanning to obtain second damage information; and then combines the first damage information and the second damage information to obtain damage detection information corresponding to the rental vehicle. Thus, a multi-dimensional vehicle damage detection process is realized. Preliminary damage positioning is performed through historical driving data, reducing the scope of damage detection; and for damage positioning, the first sensor is used to consider the external dimension, avoiding the uncertainty caused by image quality and viewing angle, and further for damage positioning, the second sensor is used to consider the structural dimension, making up for the lack of structural damage detection, so the accuracy of vehicle damage detection is improved.

[0146] For the specific limitations of the detection device for vehicle damage, reference can be made to the limitations of the detection method for vehicle damage in the above text and will not be elaborated here. Each unit module in the above detection device for vehicle damage can be implemented in whole or in part by software, hardware, and their combination. The above modules can be embedded in the processor in the computer device in hardware form or independent of it, or stored in the memory in the computer device in software form, so that the processor can call and execute the operations corresponding to the above modules.

[0147] Another embodiment of this application also proposes a computing device. Refer to Figure 6 As shown, an exemplary embodiment of this specification also provides a computing device, including: a memory and a processor, the memory stores a computer program, and when the processor executes the computer program, it executes the steps in the detection method for vehicle damage according to various embodiments of this specification described in the above embodiments of this specification.

[0148] The internal structure of this computing device can be as Figure 6As shown, the computing device includes a processor, a memory, a network interface, and an input device connected via a system bus. Among them, the processor of the computing device is used to provide computing and control capabilities. The memory of the computing device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The network interface of the computing device is used to communicate with an external terminal via a network connection. When the computer program is executed by the processor, it performs the steps in the vehicle damage detection method according to various embodiments of this specification described in the above embodiments of this specification.

[0149] The processor may include a main processor and may also include a baseband chip, a modem, etc.

[0150] The memory stores a program for implementing the technical solution of the present invention and may also store an operating system and other key services. Specifically, the program may include program code, and the program code includes computer operation instructions. More specifically, the memory may include a read-only memory (ROM), other types of static storage devices that can store static information and instructions, a random access memory (RAM), other types of dynamic storage devices that can store information and instructions, a disk memory, a flash memory, etc.

[0151] The processor may be a general-purpose processor, such as a general-purpose central processing unit (CPU), a microprocessor, etc., or an application-specific integrated circuit (ASIC), or one or more integrated circuits for controlling the execution of the program of the present invention. It may also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components.

[0152] The input device may include a device for receiving user input data and information, such as a keyboard, a mouse, a camera, a scanner, a light pen, a voice input device, a touch screen, a pedometer, or a gravity sensor, etc.

[0153] The output device may include a device for allowing output of information to the user, such as a display screen, a printer, a speaker, etc.

[0154] The communication interface may include any transceiver-like device for communicating with other devices or communication networks, such as Ethernet, a radio access network (RAN), a wireless local area network (WLAN), etc.

[0155] The processor executes the program stored in the memory and calls other devices, which can be used to implement each step of any one of the vehicle damage detection methods provided in the above embodiments of the present application.

[0156] The computing device may further include a display component and a voice component. The display component may be a liquid crystal display screen or an electronic ink display screen. The input device of the computing device may be a touch layer covered on the display component, or a button, a trackball or a touchpad provided on the housing of the computing device, or an external keyboard, touchpad or mouse, etc.

[0157] Those skilled in the art can understand that Figure 6 the structure shown in

[0158] is only a block diagram of some structures related to the solution of this specification, and does not constitute a limitation on the computing device to which the solution of this specification is applied. The specific computing device may include more or fewer components than those shown in the figure, or combine some components, or have different component arrangements.

[0159] The computer program product may be written in any combination of one or more programming languages for the program code to perform the operations of the embodiments of this specification. The programming languages include object-oriented programming languages such as Java, C++, etc., and also include conventional procedural programming languages such as the "C" language or similar programming languages. The program code may be executed entirely on the user computing device, partially on the user device, executed as an independent software package, partially on the user computing device and partially on a remote computing device, or entirely on a remote computing device or server.

[0160] In addition, the embodiments of this specification further provide a computer-readable storage medium, on which a computer program is stored, and the computer program is executed by the processor to perform the steps in the vehicle damage detection method according to various embodiments of this specification described in the above "Exemplary Method" section.

[0161] Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be completed by instructing relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above methods. Among them, any reference to a memory, storage, database, or other medium used in the embodiments provided in this specification can include non-volatile and / or volatile memories. Non-volatile memories can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memories can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and Rambus dynamic RAM (RDRAM), etc.

[0162] The technical features of the above embodiments can be combined arbitrarily. For the sake of brevity of description, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered as the scope described in this specification.

[0163] The above-described embodiments merely represent several implementation manners of this specification. The description is relatively specific and detailed, but it should not be construed as a limitation on the scope of the solutions provided by the embodiments of this specification. It should be noted that for those of ordinary skill in the art, without departing from the concept of this specification, several modifications and improvements can still be made, and these all belong to the protection scope of this specification. Therefore, the protection scope of the patent of this specification should be subject to the appended claims.

Claims

1. A method for detecting vehicle damage, characterized in that, Including: In response to a vehicle return request for a rental vehicle, obtaining historical driving data corresponding to the rental vehicle; Analyzing the driving information indicated by the historical driving data to obtain early warning information indicating vehicle damage; Based on the early warning information, calling a first sensor associated with the rental vehicle to perform image acquisition to obtain first damage information; According to the first damage information, calling a second sensor associated with the rental vehicle to perform vehicle structure scanning to obtain second damage information; Combining the first damage information and the second damage information to obtain damage detection information corresponding to the rental vehicle.

2. The method according to claim 1, wherein The analyzing the driving information indicated by the historical driving data to obtain early warning information indicating vehicle damage includes: Analyzing the driving information indicated by the historical driving data to obtain vehicle speed change information; Performing a collision judgment according to the vehicle speed change information to obtain a collision warning; Detecting a collision event generated by a collision detection module configured in the rental vehicle; Combining the collision warning and the collision event to determine early warning information indicating vehicle damage.

3. The method according to claim 2, wherein The performing a collision judgment according to the vehicle speed change information to obtain a collision warning includes: Obtaining a preset driving trajectory corresponding to the rental vehicle; Performing lateral acceleration detection based on the preset driving trajectory to obtain a lateral acceleration parameter; If the lateral acceleration parameter is greater than a lateral threshold, generating the collision warning.

4. The method according to claim 1, characterized in that, The analyzing the driving information indicated by the historical driving data to obtain early warning information indicating vehicle damage includes: Analyzing the driving information indicated by the historical driving data to obtain violation information corresponding to the rental vehicle, where the violation information is determined based on the road where the rental vehicle is located; Counting each abnormal operation indicated in the violation information to obtain violation statistical information; Comparing the violation statistical information with a violation threshold to obtain a dangerous driving parameter; Generating early warning information indicating vehicle damage based on the dangerous driving parameter.

5. The method according to claim 1, characterized in that The first sensor is a vision sensor, and the based on the early warning information, calling a first sensor associated with the rental vehicle to perform image acquisition to obtain first damage information includes: Determining risk components corresponding to the rental vehicle based on the early warning information; Determining a target visual range corresponding to the risk components; According to the target visual range, calling a vision sensor associated with the rental vehicle to perform image acquisition to obtain the first damage information.

6. The method according to claim 5, characterized in that, The according to the target visual range, calling a vision sensor associated with the rental vehicle to perform image acquisition to obtain the first damage information includes: Triggering the rental vehicle to detect adjacent vehicles according to the target visual range; Sending an image assistance request to the detected adjacent vehicles, so that the adjacent vehicles determine an associated vision sensor to perform image acquisition on the target visual range according to the assistance request; Obtaining the acquired images sent by the adjacent vehicles to obtain the first damage information.

7. The method according to claim 6, wherein The method further includes: Obtaining a plurality of acquired images sent by the plurality of adjacent vehicles; Review the multiple collected images based on the target visual range to update the first damage information.

8. The method according to claim 1, characterized in that, The second sensor is a radar sensor. Call the second sensor associated with the rental vehicle according to the first damage information to perform a vehicle structure scan to obtain second damage information, including: Send a radar assistance request to adjacent vehicles around the rental vehicle according to the first damage information, so that the adjacent vehicles call the associated radar sensors to perform a vehicle structure scan on the rental vehicle; Obtain the three-dimensional vehicle model scanned by the adjacent vehicles; Perform a comparison of structural components based on the three-dimensional vehicle model to obtain the second damage information.

9. The method according to claim 8, wherein The method further includes: Determine the damaged components indicated by the first damage information; Determine the damaged structural components according to the second damage information; Perform an association match on the damaged components and the damaged structural components to obtain association information; Configure a confidence label for the second damage information based on the association information.

10. The method according to any one of claims 1-9, characterized in that, The method further includes: Determine the vehicle component information indicated by the damage detection information; In response to a viewing request for the rental vehicle, obtain the augmented reality scene corresponding to the viewing request; Mark the damage of the rental vehicle in the augmented reality scene based on the vehicle component information.

11. A detection device for vehicle damage, characterized in that, It includes: An acquisition unit, configured to obtain the historical driving data corresponding to the rental vehicle in response to a return request for the vehicle; An analysis unit, configured to analyze the driving information indicated by the historical driving data to obtain early warning information indicating vehicle damage; A detection unit, configured to call a first sensor associated with the rental vehicle to perform image acquisition based on the early warning information to obtain first damage information; The detection unit is further configured to call a second sensor associated with the rental vehicle to perform a vehicle structure scan according to the first damage information to obtain second damage information; The detection unit is further configured to combine the first damage information and the second damage information to obtain the damage detection information corresponding to the rental vehicle.

12. A vehicle, characterized in that, The vehicle executes the method for detecting vehicle damage according to any one of claims 1-10.

Images