Vehicle defect detection method, device and equipment and storage medium

Abstract

The invention relates to a vehicle defect detection method, device and equipment and a storage medium, and the method comprises the steps: firstly, receiving video stream data transmitted by a client; wherein the video stream data is acquired for a to-be-detected vehicle, and then vehicle defect detection is performed on the video stream data by using a vehicle defect detection model to obtain a defect detection result. According to the embodiment of the invention, the received video stream data is detected through the vehicle defect detection model, the defect detection result is obtained, the vehicle condition detection effect and detection efficiency of the vehicle are improved, and wrong detection and missing detection are avoided.

Description

Technical Field

[0001] The present disclosure relates to the field of computer processing technology, and in particular to a vehicle defect detection method, device, equipment and storage medium. Background Art

[0002] With the continuous development of the automobile industry, the used car business has gradually formed a trend of large-scale development, but it has also brought a series of challenges.

[0003] The existing solution for vehicle condition inspection in the used car business is to use manual inspection to evaluate the overall vehicle condition, including the appearance. Although this solution can meet the existing needs of used car condition inspection, due to the large time investment and labor costs, the overall inspection efficiency is low, and there are a lot of false detections and missed detections in manual judgments. Summary of the invention

[0004] In order to solve the above technical problems, the embodiments of the present disclosure provide a vehicle defect detection method, device, equipment and storage medium, which improve the vehicle condition detection effect and detection efficiency and avoid false detection and missed detection.

[0005] In a first aspect, an embodiment of the present disclosure provides a vehicle defect detection method, comprising:

[0006] Receiving video stream data transmitted by a client; wherein the video stream data is collected for a vehicle to be detected;

[0007] The vehicle defect detection model is used to perform vehicle defect detection on the video stream data to obtain a defect detection result.

[0008] In an optional implementation, the defect detection result includes defect location information, the vehicle defect detection model includes a defect location detection model, and the vehicle defect detection is performed on the video stream data using the vehicle defect detection model to obtain the defect detection result, including:

[0009] The first video frame in the video stream data is input into the defect position detection model, and after being processed by the defect position detection model, defect position information corresponding to the first video frame is obtained.

[0010] In an optional implementation, the defect detection result includes defect category information, the vehicle defect detection model also includes a defect category detection model, and the first video frame in the video stream data is input into the defect position detection model, and after being processed by the defect position detection model, the defect position information corresponding to the first video frame is obtained, and further includes:

[0011] Based on the defect position information, determining a defect area image from the first video frame;

[0012] The defect category detection model is used to perform defect category detection on the defect area image to obtain defect category information corresponding to the defect area image.

[0013] In an optional implementation manner, the defect detection result includes defective component information, the vehicle defect detection model also includes a defective component detection model, and after determining the defective area image from the first video frame based on the defect position information, the method further includes:

[0014] The defective component detection model is used to identify vehicle components in the defective region image to obtain defective component information corresponding to the defective region image; wherein the defective component information is used to characterize defective vehicle components.

[0015] In an optional implementation manner, the method further includes:

[0016] The defect category information and / or defect component information corresponding to the first video frame and the defect location information corresponding to the first video frame are transmitted to the client; wherein the defect location information is used to mark the vehicle defect on the video stream acquisition page of the client.

[0017] In an optional implementation manner, the method further includes:

[0018] If it is determined based on the defect detection result that there is a vehicle defect on the third video frame in the video stream data, predicting a moving trajectory of the vehicle defect to obtain a predicted trajectory of the vehicle defect;

[0019] After obtaining the real movement trajectory of the defect corresponding to the target video frame sequence, determining the matching degree between the real movement trajectory of the defect and the predicted trajectory; wherein the target video frame sequence includes a plurality of continuous video frames after the third video frame in the video stream data;

[0020] If it is determined that the matching degree between the actual moving trajectory of the defect and the predicted trajectory is greater than a set matching degree, the defect detection result corresponding to the third video frame is determined as the defect detection result corresponding to the target video frame sequence.

[0021] In an optional implementation manner, the method further includes:

[0022] The vehicle orientation recognition model is used to perform vehicle orientation recognition on the second video frame in the video stream data to obtain an orientation recognition result corresponding to the second video frame; wherein the orientation recognition result is used to characterize whether the second video frame belongs to a preset orientation image corresponding to any orientation identifier.

[0023] If the orientation recognition result corresponding to the second video frame includes an orientation identifier, the second video frame and the orientation identifier to which the second video frame belongs are transmitted to the client; wherein the client is used to display the second video based on the orientation identifier.

[0024] In an optional implementation manner, before the defect category detection is performed on the defect region image using the defect category detection model to obtain defect category information corresponding to the defect region image, the method further includes:

[0025] Performing image quality compression on a preset ratio of defective image samples in the first defective sample set to obtain a second defective sample set; wherein the preset ratio is determined from a preset ratio range;

[0026] performing data enhancement processing on the defect image samples in the second defect sample set to obtain a third defect sample set;

[0027] The defect image samples in the third defect sample set are used to train the defect category detection model to obtain a trained defect category detection model.

[0028] In a second aspect, an embodiment of the present disclosure provides a vehicle defect detection device, comprising:

[0029] A video stream data receiving module, used to receive video stream data transmitted by a client; wherein the video stream data is collected for a vehicle to be detected;

[0030] The vehicle defect detection module is used to perform vehicle defect detection on the video stream data using a vehicle defect detection model to obtain a defect detection result.

[0031] In a third aspect, an embodiment of the present disclosure provides an electronic device, the electronic device comprising:

[0032] one or more processors;

[0033] A storage device for storing one or more programs;

[0034] When the one or more programs are executed by the one or more processors, the one or more processors implement the vehicle defect detection method as described in any one of the first aspects above.

[0035] In a fourth aspect, an embodiment of the present disclosure provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the vehicle detection method as described in any one of the above-mentioned first aspects.

[0036] The disclosed embodiments provide a vehicle defect detection method, apparatus, device, storage medium and program product, the method comprising: first, receiving video stream data transmitted by a client; wherein the video stream data is collected for a vehicle to be detected, and then, using a vehicle defect detection model to perform vehicle defect detection on the video stream data to obtain a defect detection result. The disclosed embodiments detect the received video stream data through a vehicle defect detection model to obtain a defect detection result, thereby improving the vehicle condition detection effect and detection efficiency, and avoiding false detection and missed detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] The above and other features, advantages and aspects of the embodiments of the present disclosure will become more apparent with reference to the following detailed description in conjunction with the accompanying drawings. Throughout the accompanying drawings, the same or similar reference numerals represent the same or similar elements. It should be understood that the drawings are schematic and the originals and elements are not necessarily drawn to scale.

[0038] Figure 1 An architecture diagram of a vehicle defect detection scenario provided by an embodiment of the present disclosure is shown;

[0039] Figure 2 A schematic diagram of a vehicle defect detection method in an embodiment of the present disclosure;

[0040] Figure 3 is a schematic diagram of a vehicle orientation in an embodiment of the present disclosure;

[0041] Figure 4a is a schematic diagram of a vehicle defect detection process in an embodiment of the present disclosure;

[0042] Figure 4b is a schematic diagram of a vehicle defect detection process in an embodiment of the present disclosure;

[0043] Figure 5 is a structural schematic diagram of a vehicle defect detection device in an embodiment of the present disclosure;

[0044] Figure 6 It is a structural schematic diagram of an electronic device in an embodiment of the present disclosure. DETAILED DESCRIPTION

[0045] Embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although certain embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be construed as being limited to the embodiments described herein, which are instead provided for a more thorough and complete understanding of the present disclosure. It should be understood that the drawings and embodiments of the present disclosure are only for exemplary purposes and are not intended to limit the scope of protection of the present disclosure.

[0046] It should be understood that the various steps described in the method embodiments of the present disclosure may be performed in different orders and / or in parallel. In addition, the method embodiments may include additional steps and / or omit the steps shown. The scope of the present disclosure is not limited in this respect.

[0047] The term "including" and its variations used herein are open inclusions, i.e., "including but not limited to". The term "based on" means "based at least in part on". The term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one additional embodiment"; the term "some embodiments" means "at least some embodiments". The relevant definitions of other terms will be given in the following description.

[0048] It should be noted that the concepts such as "first" and "second" mentioned in the present disclosure are only used to distinguish different devices, modules or units, and are not used to limit the order or interdependence of the functions performed by these devices, modules or units.

[0049] It should be noted that the modifications of "one" and "plurality" mentioned in the present disclosure are illustrative rather than restrictive, and those skilled in the art should understand that unless otherwise clearly indicated in the context, it should be understood as "one or more".

[0050] The names of the messages or information exchanged between multiple devices in the embodiments of the present disclosure are only used for illustrative purposes and are not used to limit the scope of these messages or information.

[0051] Before further describing the embodiments of the present disclosure in detail, the nouns and terms involved in the embodiments of the present disclosure are described. The nouns and terms involved in the embodiments of the present disclosure are subject to the following interpretations.

[0052] In response, it is used to indicate the conditions or states on which the executed operations depend. When the dependent conditions or states are met, one or more operations executed can be in real time or have a set delay. Unless otherwise specified, there is no restriction on the order of execution of the multiple operations executed.

[0053] It is understandable that before using the technical solutions disclosed in the embodiments of the present disclosure, the types, scope of use, usage scenarios, etc. of the personal information involved in the present disclosure should be informed to the user and the user's authorization should be obtained in an appropriate manner in accordance with relevant laws and regulations.

[0054] For example, in response to receiving an active request from a user, a prompt message is sent to the user to clearly prompt the user that the operation requested to be performed will require obtaining and using the user's personal information. Thus, the user can autonomously choose whether to provide personal information to software or hardware such as an electronic device, application, server, or storage medium that performs the operation of the technical solution of the present disclosure according to the prompt message.

[0055] The vehicle defect detection method provided in the embodiments of the present disclosure is described in detail below in conjunction with the accompanying drawings.

[0056] Figure 1 A system that can be used to implement the vehicle defect detection method provided by the embodiment of the present disclosure. Figure 1 As shown, the system 100 may include a plurality of user terminals 110, a network 120, a server 130, and a database 140. For example, the user terminal 110 may be used to implement the vehicle defect detection method described in any embodiment of the present disclosure.

[0057] It will be appreciated that the user terminal 110 may be any other type of electronic device capable of performing data processing, which may include but is not limited to: a mobile phone, a station, a unit, a device, a multimedia computer, a multimedia tablet, an Internet node, a communicator, a desktop computer, a laptop computer, a notebook computer, a netbook computer, a tablet computer, a personal communication system (PCS) device, a personal navigation device, a personal digital assistant (PDA), an audio / video player, a digital camera / camcorder, a positioning device, a television receiver, a radio broadcast receiver, an e-book device, a gaming device, or any combination thereof, including accessories and peripherals of these devices or any combination thereof.

[0058] The embodiments of the present disclosure have no restrictions on the hardware system and software system of the user terminal 110. For example, the user terminal 110 can be based on a processor such as ARM, X86, etc., can have input / output devices such as a camera, a touch screen, and a microphone, and can run an operating system such as Windows, iOS, Linux, Android, and Hongmeng OS.

[0059] The user terminal 110 can implement the vehicle defect detection method provided by the embodiment of the present disclosure by running a process or thread. In some examples, the user terminal 110 can use its built-in application to execute the vehicle defect detection method. In other examples, the user terminal 110 can execute the vehicle defect detection method by calling an application stored externally to the user terminal 110.

[0060] The network 120 may be a single network or a combination of at least two different networks. For example, the network 120 may include, but is not limited to, a local area network, a wide area network, a public network, a private network, etc., or a combination of several of them. The network 120 may be a computer network such as the Internet and / or various telecommunication networks (e.g., 3G / 4G / 5G mobile communication networks, Wi-Fi, Bluetooth, ZigBee, etc.), and the embodiments of the present disclosure are not limited thereto.

[0061] The server 130 may be a single server, or a server group, or a cloud server, and each server in the server group is connected via a wired or wireless network. A server group may be centralized, such as a data center, or distributed. The server 130 may be local or remote. The server 130 may communicate with the user terminal 110 via a wired or wireless network. The embodiments of the present disclosure do not limit the hardware system and software system of the server 130.

[0062] The database 140 may generally refer to a device with a storage function. The database 140 is mainly used to store various data used, generated and output by the user terminal 110 and the server 130 during work. The database 140 may be local or remote. The database 140 may include various memories, such as random access memory (RAM), read-only memory (ROM), etc. The storage devices mentioned above are just some examples, and the storage devices that can be used by the system 100 are not limited to this. The embodiments of the present disclosure do not limit the hardware system and software system of the database 140. For example, it may be a relational database or a non-relational database.

[0063] The database 140 may be connected or communicated with the server 130 or a part thereof via the network 120 , or directly connected or communicated with the server 130 , or a combination of the above two methods.

[0064] In some examples, the database 140 may be an independent device. In other examples, the database 140 may also be integrated in at least one of the user terminal 110 and the server 130. For example, the database 140 may be set on the user terminal 110 or on the server 130. For another example, the database 140 may also be distributed, with a portion of it being set on the user terminal 110 and another portion being set on the server 130.

[0065] Figure 2is a flow chart of a vehicle defect detection method in an embodiment of the present disclosure. This embodiment is applicable to the case of performing vehicle defect assessment. The method can be executed by a vehicle defect detection device. The vehicle defect detection device can be implemented in software and / or hardware. The vehicle defect detection device can be configured as follows: Figure 1 In the electronic device.

[0066] like Figure 2 As shown, the vehicle defect detection method provided by the embodiment of the present disclosure mainly includes steps S101-S102.

[0067] S101: Receive video stream data transmitted by a client.

[0068] The video stream data is collected for the vehicle to be detected.

[0069] In the embodiment of the present disclosure, the video stream data can be understood as video data transmitted in the form of a stream. Among them, the video stream data can be understood as the tester holding the test device (for example Figure 1 The electronic device described in the above) drives around the vehicle for one cycle, and the video stream data is generated in the process of driving around the vehicle. Further, the video stream data refers to the video stream data (such as the video stream data) collected by the inspector holding the inspection device starting from the left front position of the vehicle to be inspected and driving around the vehicle from the left side (the order of driving around the vehicle is: left front, left side, left rear, rear, right rear, right side, right front, front and roof). Figure 3 in the direction of the large arrow shown in the figure).

[0070] In one embodiment of the present disclosure, performing video capture on a vehicle to be inspected to obtain video stream data may include: in response to an inspector starting an inspection application, controlling a camera on the inspection device to turn on, and obtaining video stream data for the vehicle to be inspected in real time through the camera.

[0071] S102: Perform vehicle defect detection on the video stream data using a vehicle defect detection model to obtain a defect detection result.

[0072] In the disclosed embodiment, the vehicle defect detection model refers to a pre-trained model for evaluating the vehicle condition. The defect detection result can be understood as the defect detection result for evaluating the vehicle condition of the vehicle to be detected after the vehicle defect detection model detects the video stream data.

[0073] In the disclosed embodiment, the video stream data acquired in step S101 is transmitted from the client to the vehicle defect detection model in real time, so that the video stream data is detected by using the vehicle defect detection model.

[0074] In the disclosed embodiment, video stream data refers to a series of video frames arranged in a set order. In the disclosed embodiment, the video stream data is sequentially input into the vehicle detection model in the order of the picture frames, so as to detect the image frames using the vehicle detection model. It should be noted that the video stream data does not need to be input into the vehicle defect detection model after the acquisition is completed, but the camera acquires a frame of image frames, that is, the image frames are synchronously input into the vehicle defect detection model, so as to perform defect detection on the vehicle in real time.

[0075] In one embodiment of the present disclosure, since a series of image frames included in the video stream data have a large overlap of image contents in adjacent image frames, a frame extraction method can be used to extract target image frames from the video stream data and input them into the vehicle defect detection model. In other words, one image frame can be extracted after each preset number of image frames as the target image frame, wherein the preset number can be set according to actual conditions, for example, the preset number is 3. That is, one image frame is obtained as the target image frame every 3 frames.

[0076] In one embodiment of the present disclosure, the defect detection result includes defect location information, and further, on this basis, may also include at least one of defect category information and defect component information. When the vehicle to be detected does not have any defects, the defect detection result is used to indicate that the vehicle to be detected has no defects.

[0077] In one embodiment of the present disclosure, the vehicle defect detection method also includes: using a vehicle orientation recognition model to perform vehicle orientation recognition on a video frame in the video stream data to obtain an orientation recognition result corresponding to the video frame, wherein the orientation recognition result is used to characterize whether the video frame belongs to a preset orientation image corresponding to any orientation identifier.

[0078] In the disclosed embodiment, the preset orientation image is a pre-set vehicle orientation image corresponding to the standard orientation, which can better reflect the orientation characteristics of the corresponding standard orientation. For example, if the standard orientation is the left front, the preset orientation image corresponding to the left front can better reflect the left front characteristics of the vehicle.

[0079] Specifically, the vehicle orientation recognition model is used to perform vehicle orientation recognition on the second video frame in the video stream data, and after obtaining the orientation recognition result corresponding to the second video frame, if the orientation recognition result corresponding to the second video frame includes an orientation identifier, the second video frame and the orientation identifier to which the second video frame belongs are transmitted to the client; wherein the client is used to display the second video frame based on the orientation identifier. Specifically, when the client receives the second video frame and the orientation identifier to which the second video frame belongs, it can display the icon corresponding to the orientation identifier in a preset manner, such as lighting up the icon, and at the same time display the second video frame at the position corresponding to the orientation identifier.

[0080] In one embodiment of the present disclosure, the vehicle position recognition model is a multi-classification model. Specifically, the vehicle position recognition model can be a classification model. In one possible embodiment, the vehicle position recognition model is obtained after training with ResNet18 as the backbone network. Among them, the basic meaning of ResNet18 is that the basic architecture of the network is ResNet, and the depth of the network is 18 layers. The network depth here refers to the weight layer of the network, that is, including pooling, activation, and linear layers.

[0081] In the embodiment of the present disclosure, the number of directions of the direction mark can be set according to actual conditions. Figure 3 As shown, in the embodiment of the present disclosure, 9 standard orientation marks are set, namely, the left front, left side, left rear, rear, right rear, right side, right front, front and roof of the vehicle. Each standard orientation mark corresponds to a preset orientation image.

[0082] In one embodiment of the present disclosure, if after the vehicle position recognition is completed, the detected position mark is less than the number of positions marked, such as less than the above-mentioned number of positions 9, it indicates that there is an error in the vehicle position recognition process, and the vehicle position detection needs to be re-performed on the vehicle to be detected.

[0083] In one embodiment of the present disclosure, the detection device of the client collects multiple pictures around 9 positions, namely, the left front, left side, left rear, rear, right rear, right side, right front, front and roof, to construct a vehicle standard position data set, and uses the points that need to be recorded in the vehicle standard position data set to perform training and recognition of classification tasks to complete the training of the vehicle position recognition model. During use, the relative position of the current camera and the vehicle is located in real time, and 9 standard positions are recognized in turn through the left front, left side, left rear, rear, right rear, right side, right front, front and roof.

[0084] In the embodiment of the present disclosure, it should be noted that the nine standard orientations are identified in sequence, and the image frames corresponding to the standard orientations are processed into pictures as corresponding preset orientation images and saved.

[0085] In one embodiment of the present disclosure, the vehicle defect detection model includes a defect position detection model, and the vehicle defect detection is performed on the video stream data using the vehicle defect detection model to obtain a defect detection result, including: inputting a first video frame in the video stream data into the defect position detection model, and obtaining defect position information corresponding to the first video frame after being processed by the defect position detection model. The first video frame can be any frame in the video stream data.

[0086] In one embodiment of the present disclosure, the vehicle defect detection model includes a defect category detection model, which is trained using a defect data set; after obtaining the defect position information corresponding to the first video frame, the defect area image is determined from the first video frame based on the defect position information; the defect category detection model is used to perform defect category detection on the defect area image to obtain the defect category information corresponding to the defect area image.

[0087] In one embodiment of the present disclosure, the defect category detection model is a trained single-stage target detection network Yolov7.

[0088] In one embodiment of the present disclosure, the vehicle defect information includes defect category information, which includes at least one of seven types: dents, scrapes, scratches, scuffs, cracks, breakages, and rust.

[0089] In one embodiment of the present disclosure, in addition to the above-mentioned seven defect categories, an "other" category is added, and data from a variety of scenarios, including "lossless" vehicle data under strong light, are used as negative samples for training to avoid interference in vehicle paint defect detection caused by reflections, making the model more robust, thereby achieving effective filtering of background noise in different scenarios, preventing false detection of the model to a certain extent, and greatly improving the model's domain generalization ability.

[0090] In one embodiment of the present disclosure, each defect sample image in the defect data set is subjected to quality compression processing.

[0091] In the disclosed embodiment, during the training phase, in order to achieve high recall of defects under limited resolution, a data enhancement method of image quality compression is used for defect sample images, and the defect image samples in the first defect sample set are subjected to random image quality compression of not less than 20% of the original quality to obtain a second defect sample set. Then, the defect image samples in the second defect sample set are subjected to data enhancement processing to obtain a third defect sample set. The defect image samples in the third defect sample set can also be used to train a defect category detection model to obtain a trained defect category detection model, so as to overcome the problem that the model is difficult to take effect due to severe loss of original data information caused by data fluctuations under real-time video streaming transmission, thereby effectively improving the robustness of the model.

[0092] During the defect dataset construction phase, the defective areas were manually labeled using the existing second-hand car inspection data. Since the distribution of vehicle defects is relatively complex, they may appear in various locations on the vehicle body and vary in shape and size, resulting in large errors and omissions in the manually divided defect category labels. To further ensure the accurate determination of the defect category, the manually labeled inspection areas were extracted and then subjected to secondary data cleaning to construct a defect category dataset. The defect category dataset was used to train a defect classification model based on ResNet18 separately, and the overall recall and accuracy were improved by first detecting the defect area and then classifying it. At the same time, the adjustment and subdivision of the defect category standard only requires the training of a lighter classification model without the need to train the detection network again, which can effectively shorten the model iteration cycle for updating defect categories.

[0093] In one embodiment of the present disclosure, the vehicle defect detection model also includes: a defective component detection model; after determining the defect area image from the first video frame based on the defect position information, it also includes: using the defective component detection model to identify vehicle components in the defect area image to obtain defective component information corresponding to the defective area image; wherein the defective component information is used to characterize defective vehicle components.

[0094] In the disclosed embodiment, by performing component area matching on the defective area image, the defective component of the vehicle corresponding to the defective area image is obtained, so that the inspector can intuitively see the specific component with defects.

[0095] In one embodiment of the present disclosure, the defect sample image is subjected to random data enhancement processing, wherein the random data enhancement processing method includes one or more of the following: image rotation, image translation, image shearing, image scaling, and image perspective transformation.

[0096] In the disclosed embodiment, a random data enhancement method is proposed for the detection of parts to which defects belong, which integrates data enhancement means such as rotation (-30°~30°), translation, shearing (-30°~30°), scaling (0.5~1.6) and perspective transformation, and eliminates the flipping enhancement of data, thereby effectively enhancing the recognition of local features of vehicle parts and avoiding confusing recognition of left and right directions of vehicle parts (such as left front door and right front door). It enables the inspector to locate the current vehicle part while approaching the defective area, and further converts the problem into a matching problem between the defective area and the component area, so as to realize the automatic matching output of defects and their corresponding parts during the vehicle condition inspection process.

[0097] In one embodiment of the present disclosure, the vehicle defect detection result includes defect location information, defect category information and defect component information. After the vehicle defect detection model is used to detect the image frames in the video stream data and obtain the vehicle defect result, the following further includes:

[0098] If it is determined based on the defect detection result that there is a vehicle defect on the third video frame in the video stream data, the moving trajectory of the vehicle defect is predicted to obtain the predicted trajectory of the vehicle defect; after obtaining the real moving trajectory of the defect corresponding to the target video frame sequence, the matching degree between the real moving trajectory of the defect and the predicted trajectory is determined; wherein the target video frame sequence includes a plurality of consecutive video frames after the third video frame in the video stream data; if it is determined that the matching degree between the real moving trajectory of the defect and the predicted trajectory is greater than a set matching degree, the defect detection result corresponding to the third video frame is determined as the defect detection result corresponding to the target video frame sequence.

[0099] In the embodiment of the present disclosure, in order to avoid redundant storage caused by repeated identification of the same defect area in different image frames, the vehicle defect obtained by identifying the current video frame through the vehicle defect detection model is tracked to obtain the actual movement trajectory of the vehicle defect, and the movement trajectory of the vehicle defect is predicted to obtain the predicted trajectory. Then, by determining whether the matching degree between the actual movement trajectory of the defect and the predicted trajectory is greater than the set matching degree, it is determined whether it needs to be uniquely stored.

[0100] The disclosed embodiment provides a vehicle defect detection method, which includes: first, receiving video stream data transmitted by a client; wherein the video stream data is collected for a vehicle to be detected, and then using a vehicle defect detection model to perform vehicle defect detection on the video stream data to obtain a defect detection result. The disclosed embodiment detects the received video stream data through a vehicle defect detection model to obtain a defect detection result, thereby improving the vehicle condition detection effect and detection efficiency and avoiding false detection and missed detection.

[0101] like Figure 4aAs shown, the vehicle defect detection process provided in the embodiment of the present disclosure mainly includes: the inspector turns on the detection system in the detection equipment, controls the camera in the detection equipment to turn on, and the inspector starts from the left front position and walks around the car from the left side. In the process of walking around the car, the camera collects video stream data in real time, and transmits the video stream data to the vehicle defect detection model. The defective vehicle detection model infers the video stream data to obtain and save the vehicle orientation picture and vehicle defect picture. The inspector independently chooses whether to upload the obtained vehicle orientation picture and vehicle defect picture to the cloud. In response to receiving an instruction to upload to the cloud, the vehicle orientation picture and vehicle defect picture are uploaded to the cloud database, and the detection ends. In response to receiving an instruction not to upload to the cloud, the detection ends.

[0102] like Figure 4b As shown, the process of the vehicle defect detection model reasoning about the video stream data includes: the vehicle defect detection model obtains the video stream data, pre-processes the video stream data, uses the vehicle orientation recognition model to process the video stream data to obtain the orientation recognition result, uses the vehicle defect detection model to process the video stream data to obtain the defect detection result, and after post-processing the orientation recognition result and the defect detection result, obtains the vehicle orientation picture and the vehicle defect picture.

[0103] Figure 5 FIG. 1 is a schematic diagram of the structure of a vehicle defect detection device in an embodiment of the present disclosure. The present embodiment is applicable to the case of detecting the vehicle condition. The vehicle defect detection device can be implemented in software and / or hardware. The vehicle defect detection device can be configured as follows: Figure 1 In the electronic device.

[0104] like Figure 5 As shown, the vehicle detection device 50 provided in the embodiment of the present disclosure mainly includes: a video stream data receiving module 51 and a vehicle defect detection module 52.

[0105] The video stream data receiving module 51 is used to receive the video stream data transmitted by the client; wherein the video stream data is collected for the vehicle to be detected;

[0106] The vehicle defect detection module 52 is used to perform vehicle defect detection on the video stream data using a vehicle defect detection model to obtain a defect detection result.

[0107] In one embodiment of the present disclosure, the defect detection result includes defect location information, the vehicle defect detection model includes a defect location detection model, and the vehicle defect detection module is specifically used to:

[0108] The first video frame in the video stream data is input into the defect position detection model, and after being processed by the defect position detection model, defect position information corresponding to the first video frame is obtained.

[0109] In one embodiment of the present disclosure, the defect detection result includes defect category information, the vehicle defect detection model also includes a defect category detection model, and the device further includes:

[0110] A defect area determination module, configured to determine a defect area image from the first video frame based on the defect position information;

[0111] The defect category detection module is used to perform defect category detection on the defect area image using the defect category detection model to obtain defect category information corresponding to the defect area image.

[0112] In one embodiment of the present disclosure, the defect detection result includes defective component information, the vehicle defect detection model also includes a defective component detection model, and the device further includes:

[0113] The component recognition module is used to use the defective component detection model to identify vehicle components on the defective area image to obtain defective component information corresponding to the defective area image; wherein the defective component information is used to characterize defective vehicle components.

[0114] In one embodiment of the present disclosure, the device further comprises:

[0115] An information transmission module is used to transmit defect category information and / or defective component information corresponding to the first video frame, and defect location information corresponding to the first video frame to the client; wherein the defect location information is used to mark vehicle defects on the video stream acquisition page of the client.

[0116] In one embodiment of the present disclosure, the device further comprises:

[0117] A trajectory prediction module, configured to predict a moving trajectory of the vehicle defect if it is determined based on the defect detection result that there is a vehicle defect on the third video frame in the video stream data, so as to obtain a predicted trajectory of the vehicle defect;

[0118] A trajectory matching module, configured to determine the degree of matching between the real movement trajectory of the defect and the predicted trajectory after obtaining the real movement trajectory of the defect corresponding to the target video frame sequence; wherein the target video frame sequence includes a plurality of continuous video frames after the third video frame in the video stream data;

[0119] The detection result determination module is used to determine the defect detection result corresponding to the third video frame as the defect detection result corresponding to the target video frame sequence if it is determined that the matching degree between the actual movement trajectory of the defect and the predicted trajectory is greater than a set matching degree.

[0120] In one embodiment of the present disclosure, the device further comprises:

[0121] an orientation recognition module, configured to perform vehicle orientation recognition on a second video frame in the video stream data using a vehicle orientation recognition model, and obtain an orientation recognition result corresponding to the second video frame; wherein the orientation recognition result is used to indicate whether the second video frame belongs to a preset orientation image corresponding to any orientation identifier;

[0122] An orientation transmission module is used to transmit the second video frame and the orientation identifier to which the second video frame belongs to to the client if the orientation recognition result corresponding to the second video frame includes an orientation identifier; wherein the client is used to display the second video frame based on the orientation identifier.

[0123] In one embodiment of the present disclosure, the device further comprises:

[0124] A quality compression module, used for performing image quality compression on a preset ratio of defective image samples in the first defective sample set to obtain a second defective sample set; wherein the preset ratio is determined from a preset ratio range;

[0125] A data enhancement module, configured to perform data enhancement processing on the defect image samples in the second defect sample set to obtain a third defect sample set;

[0126] The model training module is used to train the defect category detection model using the defect image samples in the third defect sample set to obtain a trained defect category detection model.

[0127] The vehicle defect detection device provided in the embodiment of the present disclosure can execute the steps executed in the vehicle defect detection method provided in the method embodiment of the present disclosure, and the execution steps and beneficial effects are no longer repeated here.

[0128] Figure 6 Schematic diagram of the structure of an electronic device in the embodiment of the present disclosure. Figure 6, which shows a schematic diagram of the structure of an electronic device 600 suitable for implementing the embodiment of the present disclosure. The electronic device 600 in the embodiment of the present disclosure may include, but is not limited to, mobile terminals such as mobile phones, laptop computers, digital broadcast receivers, PDAs (personal digital assistants), PADs (tablet computers), PMPs (portable multimedia players), vehicle terminals (such as vehicle navigation terminals), wearable terminal devices, etc., and fixed terminals such as digital TVs, desktop computers, smart home devices, etc. Figure 6 The electronic device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present disclosure.

[0129] like Figure 6 As shown, the electronic device 600 may include a processing device (e.g., a central processing unit, a graphics processing unit, etc.) 601, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 602 or a program loaded from a storage device 608 to a random access memory (RAM) 603 to implement the vehicle defect detection method of the embodiment described in the present disclosure. In the RAM 603, various programs and data required for the operation of the terminal device 600 are also stored. The processing device 601, the ROM 602, and the RAM 603 are connected to each other via a bus 604. An input / output (I / O) interface 605 is also connected to the bus 604.

[0130] Typically, the following devices may be connected to the I / O interface 605: an input device 606 including, for example, a touch screen, a touch pad, a keyboard, a mouse, a camera, a microphone, an accelerometer, a gyroscope, etc.; an output device 607 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; a storage device 608 including, for example, a magnetic tape, a hard disk, etc.; and a communication device 609. The communication device 609 may allow the terminal device 600 to communicate with other devices wirelessly or by wire to exchange data. Although Figure 6 The terminal device 600 is shown to have various devices, but it should be understood that it is not required to implement or possess all the devices shown. More or fewer devices may be implemented or possessed instead.

[0131] In particular, according to an embodiment of the present disclosure, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present disclosure includes a computer program product, which includes a computer program carried on a non-transitory computer-readable medium, and the computer program contains program code for executing the method shown in the flowchart, thereby implementing the vehicle defect detection method as described above. In such an embodiment, the computer program can be downloaded and installed from the network through the communication device 609, or installed from the storage device 608, or installed from the ROM 602. When the computer program is executed by the processing device 601, the above-mentioned functions defined in the method of the embodiment of the present disclosure are executed.

[0132] It should be noted that the computer-readable medium disclosed above may be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present disclosure, a computer-readable storage medium may be any tangible medium containing or storing a program that may be used by or in combination with an instruction execution system, device or device. In the present disclosure, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, in which a computer-readable program code is carried. This propagated data signal may take a variety of forms, including but not limited to an electromagnetic signal, an optical signal, or any suitable combination of the above. The computer readable signal medium may also be any computer readable medium other than a computer readable storage medium, which may send, propagate or transmit a program for use by or in conjunction with an instruction execution system, apparatus or device. The program code contained on the computer readable medium may be transmitted using any suitable medium, including but not limited to: wires, optical cables, RF (radio frequency), etc., or any suitable combination of the above.

[0133] In some embodiments, the client and the server may communicate using any currently known or future developed network protocol such as HTTP (HyperText Transfer Protocol), and may be interconnected with any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include a local area network ("LAN"), a wide area network ("WAN"), an internet (e.g., the Internet), and a peer-to-peer network (e.g., an ad hoc peer-to-peer network), as well as any currently known or future developed network.

[0134] The computer-readable medium may be included in the electronic device, or may exist independently without being incorporated into the electronic device.

[0135] The above-mentioned computer-readable medium carries one or more programs. When the above-mentioned one or more programs are executed by the terminal device, the terminal device: receives video stream data transmitted by the client; wherein the video stream data is collected for the vehicle to be detected; and uses the vehicle defect detection model to perform vehicle defect detection on the video stream data to obtain a defect detection result.

[0136] In a possible implementation manner of the present disclosure, when the above one or more programs are executed by the terminal device, the terminal device may also execute other steps described in the above embodiments.

[0137] Computer program code for performing the operations of the present disclosure may be written in one or more programming languages ​​or a combination thereof, including, but not limited to, object-oriented programming languages, such as Java, Smalltalk, C++, and conventional procedural programming languages, such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., through the Internet using an Internet service provider).

[0138] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present disclosure. In this regard, each square box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some implementations as replacements, the functions marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two square boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0139] The units involved in the embodiments described in the present disclosure may be implemented by software or hardware, wherein the name of a unit does not, in some cases, limit the unit itself.

[0140] The functions described above herein may be performed at least in part by one or more hardware logic components. For example, without limitation, exemplary types of hardware logic components that may be used include: field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), complex programmable logic devices (CPLDs), and the like.

[0141] In the context of the present disclosure, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, device, or equipment. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or equipment, or any suitable combination of the foregoing. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0142] According to one or more embodiments of the present disclosure, the present disclosure provides a vehicle defect detection method, comprising: receiving video stream data transmitted by a client; wherein the video stream data is collected for a vehicle to be detected, and then, using a vehicle defect detection model to perform vehicle defect detection on the video stream data to obtain a defect detection result.

[0143] According to one or more embodiments of the present disclosure, the present disclosure provides a vehicle defect detection method, wherein the vehicle detection model includes a vehicle defect detection model, and the vehicle defect detection model is trained using a defect data set; the vehicle detection model is used to detect image frames in the video stream data to obtain a vehicle detection result, including: using the vehicle defect detection model to detect image frames in the video stream data in sequence to obtain vehicle defect information; processing the picture frames corresponding to the vehicle defect information to obtain a vehicle defect picture corresponding to the vehicle defect information.

[0144] According to one or more embodiments of the present disclosure, the present disclosure provides a vehicle defect detection method, wherein the vehicle defect detection model includes a defect position detection model and a defect category detection model; the vehicle defect detection model is used to detect image frames in the video stream data in sequence to obtain vehicle defect information, including: for image frames in the video stream data, using the defect position detection model to perform defect position detection on the image frames to obtain a defect area image; using the defect category detection model to perform defect category detection on the defect area image to obtain a defect category corresponding to the defect area image.

[0145] According to one or more embodiments of the present disclosure, the present disclosure provides a vehicle defect detection method, wherein the vehicle defect detection model also includes: a defective component detection model; after using the defect position detection model to perform defect position detection on the image frame to obtain a defect area image, the method also includes: using the defective component detection model to perform component area matching on the defect area image to obtain the vehicle defective component corresponding to the defect area image.

[0146] According to one or more embodiments of the present disclosure, the present disclosure provides a vehicle defect detection method, wherein the vehicle defect information includes defect location information, and the vehicle defect detection model includes a target tracking sub-model; after using the vehicle defect detection model to sequentially detect image frames in the video stream data and obtain the vehicle defect information, the method further includes: using the target tracking sub-model to track the defect location information in the video stream data.

[0147] According to one or more embodiments of the present disclosure, the present disclosure provides a vehicle defect detection method, wherein each defect sample image in the defect data set is subjected to quality compression processing; and / or the defect sample image is subjected to random data enhancement processing, wherein the random data enhancement processing method includes one or more of the following: image rotation, image translation, image shearing, image scaling, and image perspective transformation.

[0148] According to one or more embodiments of the present disclosure, the present disclosure provides a vehicle defect detection method, wherein the vehicle detection model includes: a vehicle orientation recognition model, wherein the vehicle detection model is used to detect image frames in the video stream data to obtain a vehicle detection result, including: using the vehicle orientation recognition model to sequentially recognize image frames in the video stream data to obtain a vehicle orientation recognition result; processing the image frames corresponding to the vehicle orientation recognition result to obtain a vehicle preset orientation image corresponding to the vehicle orientation recognition result.

[0149] According to one or more embodiments of the present disclosure, the present disclosure provides a vehicle defect detection device, which includes: a video stream data receiving module, used to receive video stream data transmitted by a client; wherein the video stream data is collected for a vehicle to be detected; and a vehicle defect detection module, used to perform vehicle defect detection on the video stream data using a vehicle defect detection model to obtain a defect detection result.

[0150] According to one or more embodiments of the present disclosure, the present disclosure provides a vehicle defect detection device, wherein the vehicle detection model includes a vehicle defect detection model, and the vehicle defect detection model is trained using a defect data set; a vehicle detection result determination module includes: a vehicle defect information determination unit, used to use the vehicle defect detection model to detect image frames in the video stream data in sequence to obtain vehicle defect information; a vehicle defect picture determination unit, used to process picture frames corresponding to the vehicle defect information to obtain vehicle defect pictures corresponding to the vehicle defect information.

[0151] According to one or more embodiments of the present disclosure, the present disclosure provides a vehicle detection device, wherein the vehicle defect detection model includes a defect position detection model and a defect category detection model; a vehicle defect information determination unit is specifically used to perform defect position detection on the image frame in the video stream data using the defect area detection sub-model to obtain a defect area image; and perform defect category detection on the defect area image using the defect category detection sub-model to obtain a defect category corresponding to the defect area image.

[0152] According to one or more embodiments of the present disclosure, the present disclosure provides a vehicle detection device, wherein the vehicle defect detection model further includes: a defective component recognition sub-model; and a vehicle defect information determination unit, which is also used for performing defect position detection on the image frame using the defective area detection sub-model, and after obtaining the defective area image, performing component area matching on the defective area image using the defective component recognition sub-model to obtain the vehicle defective component corresponding to the defective area image.

[0153] According to one or more embodiments of the present disclosure, the present disclosure provides a vehicle detection device, wherein the vehicle defect information includes defect location information, and the vehicle defect detection model includes a target tracking sub-model; a vehicle defect information determination unit is further used to use the vehicle defect detection model to sequentially detect image frames in the video stream data, and after obtaining the vehicle defect information, use the target tracking sub-model to track the defect location information in the video stream data.

[0154] According to one or more embodiments of the present disclosure, the present disclosure provides a vehicle detection device, wherein each defect sample image in the defect data set is subjected to quality compression processing; and / or the defect sample image is subjected to random data enhancement processing, wherein the random data enhancement processing method includes one or more of the following: image rotation, image translation, image shearing, image scaling, and image perspective transformation.

[0155] According to one or more embodiments of the present disclosure, the present disclosure provides a vehicle detection device, wherein the vehicle detection model includes: a vehicle position recognition model, a vehicle detection result determination module, including: a vehicle recognition result determination unit, for using the vehicle position recognition model to sequentially recognize image frames in the video stream data to obtain a vehicle position recognition result; a vehicle preset position image determination unit, for processing the image frames corresponding to the vehicle position recognition result to obtain a vehicle preset position image corresponding to the vehicle position recognition result.

[0156] According to one or more embodiments of the present disclosure, an electronic device is provided. The electronic device includes:

[0157] one or more processors;

[0158] A storage device for storing one or more programs;

[0159] When the one or more programs are executed by the one or more processors, the one or more processors implement the vehicle defect detection method as described in any one of the first aspects above.

[0160] According to one or more embodiments of the present disclosure, an embodiment of the present disclosure provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements a vehicle defect detection method as described in any one of the first aspects above.

[0161] According to one or more embodiments of the present disclosure, an embodiment of the present disclosure provides a computer program product, which includes a computer program or instructions, and when the computer program or instructions are executed by a processor, the vehicle defect detection method as described in any one of the first aspects above is implemented.

[0162] The above description is only a preferred embodiment of the present disclosure and an explanation of the technical principles used. Those skilled in the art should understand that the scope of disclosure involved in the present disclosure is not limited to the technical solutions formed by a specific combination of the above technical features, but should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the above disclosed concept. For example, the above features are replaced with the technical features with similar functions disclosed in the present disclosure (but not limited to) by each other to form a technical solution.

[0163] In addition, although each operation is described in a specific order, this should not be required to perform these operations in the specific order shown or in a sequential order. Under certain circumstances, multitasking and parallel processing may be advantageous. Similarly, although some specific implementation details are included in the above discussion, these should not be interpreted as limiting the scope of the present disclosure. Some features described in the context of a separate embodiment can also be implemented in a single embodiment in combination. On the contrary, the various features described in the context of a single embodiment can also be implemented in multiple embodiments individually or in any suitable sub-combination mode.

[0164] Although the subject matter has been described in language specific to structural features and / or methodological logical actions, it should be understood that the subject matter defined in the appended claims is not necessarily limited to the specific features or actions described above. On the contrary, the specific features and actions described above are merely example forms of implementing the claims.

Claims

1. A vehicle defect detection method, characterized in that: The method comprises: Receiving video stream data transmitted by a client; wherein the video stream data is collected for a vehicle to be detected; The vehicle defect detection model is used to perform vehicle defect detection on the video stream data to obtain a defect detection result.

2. The method according to claim 1, characterized in that The defect detection result includes defect location information, the vehicle defect detection model includes a defect location detection model, and the vehicle defect detection is performed on the video stream data using the vehicle defect detection model to obtain a defect detection result, including: The first video frame in the video stream data is input into the defect position detection model, and after being processed by the defect position detection model, defect position information corresponding to the first video frame is obtained.

3. The method according to claim 2, characterized in that The defect detection result includes defect category information, the vehicle defect detection model also includes a defect category detection model, and the first video frame in the video stream data is input into the defect position detection model, and after being processed by the defect position detection model, the defect position information corresponding to the first video frame is obtained, and further includes: Based on the defect position information, determining a defect area image from the first video frame; The defect category detection model is used to perform defect category detection on the defect area image to obtain defect category information corresponding to the defect area image.

4. The method according to claim 3, characterized in that The defect detection result includes defective component information, the vehicle defect detection model also includes a defective component detection model, and after determining the defective area image from the first video frame based on the defect position information, the method further includes: The defective component detection model is used to identify vehicle components in the defective region image to obtain defective component information corresponding to the defective region image; wherein the defective component information is used to characterize defective vehicle components.

5. The method according to any one of claims 2 to 4, characterized in that: The method further comprises: The defect category information and / or defect component information corresponding to the first video frame and the defect location information corresponding to the first video frame are transmitted to the client; wherein the defect location information is used to mark the vehicle defect on the video stream acquisition page of the client.

6. The method according to claim 1, characterized in that The method further comprises: If it is determined based on the defect detection result that there is a vehicle defect on the third video frame in the video stream data, predicting a moving trajectory of the vehicle defect to obtain a predicted trajectory of the vehicle defect; After obtaining the real movement trajectory of the defect corresponding to the target video frame sequence, determining the matching degree between the real movement trajectory of the defect and the predicted trajectory; wherein the target video frame sequence includes a plurality of continuous video frames after the third video frame in the video stream data; If it is determined that the matching degree between the actual moving trajectory of the defect and the predicted trajectory is greater than a set matching degree, the defect detection result corresponding to the third video frame is determined as the defect detection result corresponding to the target video frame sequence.

7. The method according to claim 1, characterized in that The method further comprises: Using the vehicle orientation recognition model to perform vehicle orientation recognition on the second video frame in the video stream data, and obtain an orientation recognition result corresponding to the second video frame; wherein the orientation recognition result is used to indicate whether the second video frame belongs to a preset orientation image corresponding to any orientation identifier; If the orientation recognition result corresponding to the second video frame includes an orientation identifier, the second video frame and the orientation identifier to which the second video frame belongs are transmitted to the client; wherein the client is used to display the second video frame based on the orientation identifier.

8. The method according to claim 3, characterized in that Before performing defect category detection on the defect region image using the defect category detection model to obtain defect category information corresponding to the defect region image, the method further includes: Performing image quality compression on a preset ratio of defective image samples in the first defective sample set to obtain a second defective sample set; wherein the preset ratio is determined from a preset ratio range; performing data enhancement processing on the defect image samples in the second defect sample set to obtain a third defect sample set; The defect image samples in the third defect sample set are used to train the defect category detection model to obtain a trained defect category detection model.

9. A vehicle defect detection device, characterized in that: include: A video stream data receiving module, used to receive video stream data transmitted by a client; wherein the video stream data is collected for a vehicle to be detected; The vehicle defect detection module is used to perform vehicle defect detection on the video stream data using a vehicle defect detection model to obtain a defect detection result.

10. An electronic device, characterized in that: The electronic device comprises: one or more processors; A storage device for storing one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the vehicle defect detection method as described in any one of claims 1-8.

11. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the vehicle defect detection method as described in any one of claims 1 to 8 is implemented.

Images