Vehicle detection method and device, electronic equipment and storage medium

Abstract

The invention provides a vehicle detection method and device, electronic equipment and a storage medium, and the method comprises the steps: obtaining a vehicle image of a to-be-detected vehicle, and determining a vehicle model of the to-be-detected vehicle; determining a model tree structure corresponding to the to-be-detected vehicle according to the vehicle model; wherein the model tree structure comprises a plurality of anomaly detection models; determining a to-be-detected vehicle part in the vehicle image, determining a target anomaly detection model corresponding to the to-be-detected vehicle part in the model tree structure, and generating a model configuration file according to the target anomaly detection model; and based on the vehicle image, calling a target anomaly detection model according to the model configuration file to detect each to-be-detected vehicle part, and obtaining a corresponding detection result. By setting the model tree structure, the anomaly detection model corresponding to the to-be-detected vehicle part is quickly matched to perform anomaly detection on the vehicle image, and the automation degree and efficiency of anomaly detection are improved.

Description

Technical Field

[0001] The present invention relates to the field of vehicles and provides a vehicle detection method, device, electronic equipment and storage medium. Background Art

[0002] The rapid development of the rail transit industry has produced good economic and social benefits. At present, urban rail transit is getting closer and closer to the lives of the public and has become the preferred means of transportation for travel within and outside the city. Therefore, the daily maintenance of rail vehicles is also particularly important, and this process is an important guarantee for ensuring the safe operation of the entire vehicle.

[0003] In order to improve the degree of automation of vehicle operation processes and the efficiency of maintenance operations, intelligent operation and maintenance has become the development trend of train operation and maintenance. Trackside inspection equipment represented by 360 panoramic detection systems and inspection robots can identify abnormal conditions such as component fatigue, deformation, defects and foreign objects generated during the service process of vehicles, but the degree of automation and efficiency of abnormality detection are low. Summary of the invention

[0004] The embodiments of the present invention provide a vehicle detection method, device, electronic device and storage medium to solve the defects of low automation and efficiency of abnormality detection in related technologies, and to improve the automation and efficiency of abnormality detection.

[0005] The present invention provides a vehicle detection method, comprising: Acquire a vehicle image of a vehicle to be detected, and determine the vehicle model of the vehicle to be detected; Determine a model tree structure corresponding to the vehicle to be detected according to the vehicle model; wherein the model tree structure includes a plurality of anomaly detection models; Determine a vehicle component to be detected in the vehicle image, and determine a target anomaly detection model corresponding to the vehicle component to be detected in the model tree structure; Based on the vehicle image, each of the vehicle components to be detected is detected by the target anomaly detection model to obtain a corresponding detection result.

[0006] According to a vehicle detection method provided by the present invention, before determining the model tree structure corresponding to the vehicle to be detected according to the vehicle model, the method further includes: obtaining vehicles and vehicle maintenance procedures for multiple vehicle models, and obtaining multiple anomaly detection models; Disassemble the vehicles of each vehicle model to obtain vehicle components corresponding to each vehicle; According to the vehicle model and the vehicle components, a basic model tree structure corresponding to each vehicle model is constructed; wherein the basic model tree structure has the corresponding vehicle model as a root node, and each vehicle component is a node other than the root node; According to the vehicle maintenance regulations, determining the detection item points of each of the nodes in the corresponding basic model tree structure; Determining the abnormality detection model corresponding to each of the vehicle components according to the detection item points; The abnormality detection model corresponding to each of the vehicle components is configured into a corresponding node to obtain the model tree structure corresponding to each of the vehicle models.

[0007] According to a vehicle detection method provided by the present invention, the step of obtaining multiple anomaly detection models includes: The device data of a plurality of trackside detection devices are acquired, and an abnormality detection model possessed by each of the trackside detection devices is determined based on the device data.

[0008] According to a vehicle detection method provided by the present invention, based on the vehicle image, each of the vehicle components to be detected is detected by the target anomaly detection model to obtain a corresponding detection result, including: Performing image correction and image segmentation on the vehicle image to obtain a plurality of component images of the vehicle components to be detected; Each component image is input into the corresponding target anomaly detection model for detection to obtain a corresponding detection result.

[0009] According to a vehicle detection method provided by the present invention, before determining the model tree structure corresponding to the vehicle to be detected according to the vehicle model, the method further includes: If a new anomaly detection model that does not exist in the model tree structure is received, performing compliance verification and availability verification on the new anomaly detection model; If the compliance check and the availability check of the new anomaly detection model pass, the new anomaly detection model is imported into the model tree structure.

[0010] The present invention also provides a vehicle detection device, comprising: A first acquisition module is configured to acquire a vehicle image of a vehicle to be detected and determine a vehicle model of the vehicle to be detected; A first determination module is configured to determine a model tree structure corresponding to the vehicle to be detected according to the vehicle model; wherein the model tree structure includes a plurality of anomaly detection models; A second determination module is configured to determine a vehicle component to be detected in the vehicle image, and determine a target anomaly detection model corresponding to the vehicle component to be detected in the model tree structure; The detection module is configured to detect each of the vehicle components to be detected based on the vehicle image through the target anomaly detection model to obtain corresponding detection results.

[0011] The present invention also provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the vehicle detection method as described in any one of the above items is implemented.

[0012] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements any of the vehicle detection methods described above.

[0013] According to the vehicle detection method, device, electronic device and storage medium provided by the present invention, a corresponding model tree structure is constructed for each vehicle model, wherein the model tree structure includes multiple anomaly detection models, and the model tree structure corresponding to the vehicle to be detected is determined according to the vehicle model of the vehicle to be detected, and then the vehicle parts to be detected in the vehicle image are determined, and the target anomaly detection model corresponding to the vehicle parts to be detected is determined in the model tree structure, and a corresponding model configuration file is generated, and the target anomaly detection model is called according to the model configuration file, and each of the vehicle parts to be detected is detected to obtain a corresponding detection result. The present invention sets a model tree structure, quickly matches the anomaly detection model corresponding to the vehicle parts to be detected, performs anomaly detection on the vehicle image, and improves the automation and efficiency of anomaly detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0015] Figure 1 It is a schematic flow chart of the vehicle detection method provided by the present invention.

[0016] Figure 2 It is a schematic diagram of the model tree structure in the embodiment provided by the present invention.

[0017] Figure 3 It is a schematic flow chart of constructing a model tree structure in an embodiment provided by the present invention.

[0018] Figure 4It is a schematic structural diagram of the vehicle detection device provided by the present invention.

[0019] Figure 5 It is a schematic structural diagram of the electronic device provided by the present invention. DETAILED DESCRIPTION

[0020] The following embodiments of the present invention are described in further detail in conjunction with the accompanying drawings and examples. The following examples are used to illustrate the present invention, but are not intended to limit the scope of the present invention.

[0021] Figure 1 FIG. 1 is a flow chart of a vehicle detection method according to an exemplary embodiment. Figure 1 As shown, in an exemplary embodiment, the vehicle detection method includes steps 110 to 140, which are described in detail as follows.

[0022] Step 110: Acquire a vehicle image of a vehicle to be detected, and determine the vehicle model of the vehicle to be detected.

[0023] In the embodiment of the present invention, a vehicle image of the vehicle to be detected is collected by a camera on the gantry, and the vehicle model of the vehicle to be detected is determined.

[0024] Step 120, determining a model tree structure corresponding to the vehicle to be detected according to the vehicle model; wherein the model tree structure includes a plurality of anomaly detection models.

[0025] In the embodiment of the present invention, a corresponding model tree structure is pre-established according to the vehicle model, and the model tree structure includes multiple anomaly detection models. Figure 2 As shown in Figure 2, the model tree structure is constructed by taking a rail vehicle as an example, and it is stipulated that the vehicle includes multiple carriages, that is, Figure 2 The model tree structure records the vehicle components in the corresponding vehicle and the anomaly detection model corresponding to each vehicle component. It also records the detection device provided by each anomaly detection model and the purpose of each anomaly detection model. In the model tree structure, there may be multiple anomaly detection models that perform the same type of detection on the same vehicle component.

[0026] The corresponding model tree structure can be determined according to the vehicle model of the vehicle to be detected.

[0027] Step 130, determining a vehicle component to be detected in the vehicle image, and determining a target anomaly detection model corresponding to the vehicle component to be detected in the model tree structure.

[0028] In an embodiment of the present invention, the vehicle parts to be detected in the vehicle image are determined. The vehicle parts to be detected can be vehicle parts identified based on the vehicle image, or can be vehicle parts manually selected by the user. For example, if a maintenance map interface is provided, a model tree structure is displayed on the maintenance map interface, and the user checks one or more vehicle parts and the anomaly detection model to be enabled on the model tree. Specifically, a model configuration file (xml file) can be generated based on the confirmed target anomaly detection model. The model configuration file includes the ID of the anomaly detection model to be enabled. The unique anomaly detection model is determined by the ID, and the target anomaly detection model is called according to the model configuration file, thereby driving the execution of the corresponding detection task.

[0029] Step 140: Based on the vehicle image, each of the vehicle components to be detected is detected by using the target anomaly detection model to obtain a corresponding detection result.

[0030] In an embodiment of the present invention, a vehicle image is input into a target anomaly detection model to detect each vehicle component to be detected and obtain corresponding detection results. Specifically, the anomaly detection model detects the abnormal points of the vehicle components to be detected, and then matches the image corresponding to the abnormal point with the abnormal problem image in the sample library, thereby determining the problem existing in the vehicle to be detected. When calling the target anomaly detection model, a message queue is used to send a request message to the scheduling service to start image downloading and model calling, thereby realizing anomaly detection and improving the model detection efficiency and refined management and control of the model.

[0031] The technical solution provided by the present invention is to construct a corresponding model tree structure for each vehicle model, wherein the model tree structure includes multiple anomaly detection models, and the model tree structure corresponding to the vehicle to be detected is determined according to the vehicle model of the vehicle to be detected, and then the vehicle parts to be detected in the vehicle image are determined, and the target anomaly detection model corresponding to the vehicle parts to be detected is determined in the model tree structure, and a corresponding model configuration file is generated, and the target anomaly detection model is called according to the model configuration file, and each of the vehicle parts to be detected is detected to obtain a corresponding detection result. The present invention sets a model tree structure, quickly matches the anomaly detection model corresponding to the vehicle parts to be detected, performs anomaly detection on the vehicle image, and improves the automation and efficiency of anomaly detection.

[0032] In an exemplary embodiment of the present invention, Figure 3 As shown, before determining the model tree structure corresponding to the vehicle to be detected according to the vehicle model, the method also includes steps 310 to 360, which are described in detail as follows.

[0033] Step 310, obtaining vehicles and vehicle maintenance procedures of multiple vehicle models, and obtaining multiple anomaly detection models.

[0034] In an embodiment of the present invention, vehicles of multiple vehicle models and corresponding vehicle maintenance procedures are obtained. A vehicle maintenance procedure is a standard maintenance and repair procedure that has been approved and formed into a written document. It provides indicative instructions for how to perform vehicle maintenance and repair to ensure safe and efficient operation of the vehicle. Multiple anomaly detection models are obtained, and the anomaly detection model can be a model stored in the system or a model in an existing trackside detection device.

[0035] Step 320: disassemble the vehicles of each vehicle model to obtain vehicle components corresponding to each vehicle.

[0036] In the embodiment of the present invention, the vehicle is disassembled based on the vehicle model to obtain corresponding vehicle components, including the vehicle compartment and various components in the vehicle compartment. At the same time, the corresponding maintenance procedures of each vehicle component are determined.

[0037] Step 330, constructing a basic model tree structure corresponding to each of the vehicle models according to the vehicle model and the vehicle components; wherein the basic model tree structure has the corresponding vehicle model as a root node, and each of the vehicle components is a node other than the root node.

[0038] In the embodiment of the present invention, a basic model tree structure corresponding to each vehicle model is constructed according to the vehicle model and the vehicle components corresponding to the vehicle model. The basic model tree structure is a tree structure, with the corresponding vehicle model as the root node, and each vehicle component as other nodes except the root node. Other nodes are constructed according to the relationship between vehicle components, forming a tree structure as shown in the figure. Figure 2 The tree shape shown.

[0039] Step 340: Determine the detection item points of each of the nodes in the corresponding basic model tree structure according to the vehicle maintenance procedure.

[0040] In the embodiment of the present invention, according to the vehicle maintenance regulations, the system sorts out the detection points corresponding to each node in the basic model tree structure. A point refers to a specific detail or key point in a plan, project or task, and correspondingly, a detection point is a point that needs to be detected.

[0041] Step 350: Determine the abnormality detection model corresponding to each of the vehicle components according to the detection item points.

[0042] In the embodiment of the present invention, the trackside detection equipment and the anomaly detection model are configured according to the detection item points, and the anomaly detection model corresponding to the anomaly detection of each vehicle component is determined.

[0043] Step 360: configure the abnormality detection model corresponding to each of the vehicle components into the corresponding node to obtain the model tree structure corresponding to each of the vehicle models.

[0044] In the embodiment of the present invention, the abnormality detection model corresponding to each vehicle component is configured into the corresponding node to obtain the model tree structure corresponding to each vehicle model. Figure 2 As shown, in the model tree structure, the bogie, a vehicle component, corresponds to the bogie inspection module SD1.0 and the bogie inspection module BS2.0 for performing appearance inspection on the bogie, as well as the bolt anti-loosening inspection model and the bolt 3D loosening detection model for inspecting the bolts in the bogie.

[0045] In an exemplary embodiment of the present invention, the obtaining of multiple anomaly detection models includes the following steps, which are described in detail as follows.

[0046] The device data of a plurality of trackside detection devices are acquired, and an abnormality detection model possessed by each of the trackside detection devices is determined based on the device data.

[0047] In the embodiment of the present invention, the trackside detection equipment is a high-tech equipment used in the railway system to monitor and diagnose trains and tracks in real time, such as a 360-degree panoramic detection system and a patrol robot. According to the detection capability of each trackside detection equipment, the abnormality detection model in each trackside detection equipment is determined. When it is necessary to detect through the abnormality detection model in the trackside detection equipment, the model contained in the trackside detection equipment is accessed according to the reserved standard interface.

[0048] In an exemplary embodiment of the present invention, based on the vehicle image, each of the vehicle components to be detected is detected by the target anomaly detection model to obtain the corresponding detection results, including the following steps, which are described in detail as follows.

[0049] The vehicle image is subjected to image correction and image segmentation to obtain a plurality of component images of the vehicle components to be detected.

[0050] Each component image is input into the corresponding target anomaly detection model for detection to obtain a corresponding detection result.

[0051] In the embodiment of the present invention, the vehicle image is corrected to facilitate subsequent processing of the vehicle image. An image segmentation model is pre-set, and the corrected vehicle image is segmented using the image segmentation model to obtain component images of multiple vehicle components to be detected. Each component image is input into the corresponding target anomaly detection model for processing to obtain a corresponding detection result.

[0052] In an exemplary embodiment of the present invention, the abnormality detection model includes at least two of a bolt loss detection model, a lock buckle abnormality detection model, a foreign object detection model, a cable loosening detection model, and a nameplate loss detection model.

[0053] In an embodiment of the present invention, a bolt loss detection model is used to detect whether the bolts on the vehicle to be inspected are lost or loose; a lock buckle abnormality detection model is used to detect whether the lock buckle on the vehicle to be inspected is abnormal; a foreign object detection model is used to detect whether there is foreign object on the surface of the vehicle to be inspected; a cable loose detection model is used to detect whether the cables in the vehicle to be inspected are loose; and a nameplate loss detection model is used to detect whether there is a nameplate on the vehicle to be inspected, and detection is performed through the above-mentioned various abnormality detection models.

[0054] In the embodiment of the present invention, the abnormality detection model also includes a bogie detection model, a blindfold appearance detection model, a blindfold height measurement model, etc. The bogie detection model is used to detect whether the bogie of the vehicle is abnormal; the blindfold appearance detection model is used to detect whether the appearance of the blindfold of the vehicle is abnormal; and the blindfold height measurement model is used to detect whether the height of the blindfold of the vehicle meets the requirements.

[0055] In an exemplary embodiment of the present invention, before determining the model tree structure corresponding to the vehicle to be detected according to the vehicle model, the method further includes the following steps, which are described in detail as follows.

[0056] If a new anomaly detection model that does not exist in the model tree structure is received, a compliance check and an availability check are performed on the new anomaly detection model.

[0057] If the compliance check and the availability check of the new anomaly detection model pass, the new anomaly detection model is imported into the model tree structure.

[0058] In the embodiment of the present invention, for each anomaly detection model, when the corresponding model is uploaded, a verification message will be sent to the model background to verify the compliance of the model and verify whether the model is available. Only the model that passes the verification is allowed to be imported into the model tree structure and used for subsequent anomaly detection, thereby ensuring the compliance of the uploaded model.

[0059] In the embodiment of the present invention, the model type, model version, and review and release status of the anomaly detection model involved in the anomaly detection process are centrally managed. The new model is updated by manual import, the model change information is recorded, and the model information is provided for the model tree structure.

[0060] In an exemplary embodiment of the present invention, after detecting each of the vehicle components to be detected by the target anomaly detection model based on the vehicle image, the method further includes the following steps, which are described in detail as follows.

[0061] Determine the running progress and execution time of each of the target anomaly detection models, and display the running progress and execution time on a preset display page.

[0062] In the embodiment of the present invention, during the operation, the processes involved, such as image acquisition, image correction, image segmentation, anomaly detection and result handling, are monitored to obtain corresponding operation progress, error information, execution time and other information. The operation progress and execution time of the target anomaly detection model for anomaly detection are centrally displayed on the front-end software interface, so that users can grasp the operation status in real time.

[0063] In an exemplary embodiment of the present invention, a vehicle detection system is provided for implementing the above-mentioned vehicle detection method, and the vehicle detection system includes a model management platform and an image processing platform. The model management platform stores anomaly detection models and corresponding model description files. The model description file is used to record the file paths of the various components of the model, describe the type and detailed functions of the model, and guide the service to call in the correct way. The model description file is named in the form of model name + ".xml". The model description file includes XML metadata, model name, model language and library file type, library file path, functions, model parameters, and model report files. Functions include supported vehicle models, supported cameras, supported components, supported targets, etc. During the operation of the model, one or more report files will be created for transferring intermediate data.

[0064] The model management platform and the image processing platform are interconnected. The image processing platform includes an image correction unit and an image segmentation unit. The image processing platform inputs the collected vehicle images into the model management platform after processing them, and concurrently calls the anomaly detection model in the model management platform to detect the vehicle images, and each outputs the results and summarizes them.

[0065] The model management platform is based on modular design ideas. It can be interconnected with the image correction unit and the image segmentation unit through inter-process communication (running) or cmd window calling, to exchange information, realize multi-model modularization, refined management and control, and fast integrated calling. The model management platform also has certain scalability and compatibility, and reserves standard interfaces to meet the model access and centralized management requirements of trackside inspection equipment such as inspection robots.

[0066] The vehicle detection device provided by the present invention is described below, and the vehicle detection device described below and the vehicle detection method described above can be referred to each other. It should be noted that the device provided in the following embodiment and the method provided in the above embodiment belong to the same concept, and the specific manner in which each module and unit performs the operation has been described in detail in the method embodiment, which will not be repeated here.

[0067] In an exemplary embodiment of the present invention, see Figure 4 , Figure 4A vehicle detection device according to an exemplary embodiment includes: A first acquisition module 410 is configured to acquire a vehicle image of a vehicle to be detected and determine a vehicle model of the vehicle to be detected; A first determination module 420 is configured to determine a model tree structure corresponding to the vehicle to be detected according to the vehicle model; wherein the model tree structure includes a plurality of anomaly detection models; A second determination module 430 is configured to determine a vehicle component to be detected in the vehicle image, and determine a target anomaly detection model corresponding to the vehicle component to be detected in the model tree structure; The detection module 440 is configured to detect each of the to-be-detected vehicle components based on the vehicle image by using the target anomaly detection model to obtain a corresponding detection result.

[0068] In an exemplary embodiment of the present invention, the vehicle detection device further includes: a second acquisition module configured to acquire vehicles and vehicle maintenance procedures of multiple vehicle models, and to acquire multiple anomaly detection models; A disassembly module, configured to disassemble the vehicle of each vehicle model to obtain vehicle components corresponding to each vehicle; A construction module, configured to construct a basic model tree structure corresponding to each vehicle model according to the vehicle model and the vehicle component; wherein the basic model tree structure has the corresponding vehicle model as a root node, and each vehicle component is a node other than the root node; A third determination module is configured to determine the detection item points of each of the nodes in the corresponding basic model tree structure according to the vehicle maintenance regulations; A fourth determination module is configured to determine the abnormality detection model corresponding to each of the vehicle components according to the detection item points; The configuration module is configured to configure the abnormality detection model corresponding to each of the vehicle components into the corresponding node to obtain the model tree structure corresponding to each of the vehicle models.

[0069] In an exemplary embodiment of the present invention, the second acquisition module includes: The acquisition submodule is configured to acquire device data of a plurality of trackside detection devices, and determine an abnormality detection model possessed by each of the trackside detection devices based on the device data.

[0070] In an exemplary embodiment of the present invention, the detection module 440 includes: An image processing module, configured to perform image correction and image segmentation on the vehicle image to obtain a plurality of component images of the vehicle components to be detected; The input module is configured to input each of the component images into the corresponding target anomaly detection model for detection to obtain corresponding detection results.

[0071] In an exemplary embodiment of the present invention, the abnormality detection model includes at least two of a bolt loss detection model, a lock buckle abnormality detection model, a foreign object detection model, a cable loosening detection model, and a nameplate loss detection model.

[0072] In an exemplary embodiment of the present invention, the vehicle detection device further includes: A verification module, configured to, if a new anomaly detection model that does not exist in the model tree structure is received, perform compliance verification and availability verification on the new anomaly detection model; The import module is configured to import the new anomaly detection model into the model tree structure if the compliance check and the availability check of the new anomaly detection model pass.

[0073] In an exemplary embodiment of the present invention, the vehicle detection device further includes: The fifth determination module is configured to determine the running progress and execution time of each of the target anomaly detection models, and display the running progress and the execution time on a preset display page.

[0074] Figure 5 An example of a physical structure diagram of an electronic device is shown in FIG. Figure 5 As shown, the electronic device may include: a processor 510, a communications interface 520, a memory 530 and a communication bus 540, wherein the processor 510, the communications interface 520 and the memory 530 communicate with each other via the communication bus 540. The processor 510 may call the logic instructions in the memory 530 to execute the vehicle detection method, which includes: Acquire a vehicle image of a vehicle to be detected, and determine the vehicle model of the vehicle to be detected; Determine a model tree structure corresponding to the vehicle to be detected according to the vehicle model; wherein the model tree structure includes a plurality of anomaly detection models; Determine a vehicle component to be detected in the vehicle image, and determine a target anomaly detection model corresponding to the vehicle component to be detected in the model tree structure; Based on the vehicle image, each of the vehicle components to be detected is detected by the target anomaly detection model to obtain a corresponding detection result.

[0075] In addition, the logic instructions in the above-mentioned memory 530 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when it is sold or used as an independent product. Based on this understanding, the technical solution of the present invention, or the part that contributes to the relevant technology or the part of the technical solution, can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods of each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc. Various media that can store program codes.

[0076] An embodiment of the present invention discloses a computer program product, which includes a computer program stored on a non-transitory computer-readable storage medium. The computer program includes program instructions. When the program instructions are executed by a computer, the computer can execute the vehicle detection method provided by the above-mentioned method embodiments, for example, including: Acquire a vehicle image of a vehicle to be detected, and determine the vehicle model of the vehicle to be detected; Determine a model tree structure corresponding to the vehicle to be detected according to the vehicle model; wherein the model tree structure includes a plurality of anomaly detection models; Determine a vehicle component to be detected in the vehicle image, and determine a target anomaly detection model corresponding to the vehicle component to be detected in the model tree structure; Based on the vehicle image, each of the vehicle components to be detected is detected by the target anomaly detection model to obtain a corresponding detection result.

[0077] On the other hand, an embodiment of the present invention further provides a non-transitory computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the vehicle detection method provided in the above embodiments is implemented, for example, including: Acquire a vehicle image of a vehicle to be detected, and determine the vehicle model of the vehicle to be detected; Determine a model tree structure corresponding to the vehicle to be detected according to the vehicle model; wherein the model tree structure includes a plurality of anomaly detection models; Determine a vehicle component to be detected in the vehicle image, and determine a target anomaly detection model corresponding to the vehicle component to be detected in the model tree structure; Based on the vehicle image, each of the vehicle components to be detected is detected by the target anomaly detection model to obtain a corresponding detection result.

[0078] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0079] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the relevant technology can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiment.

[0080] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A vehicle detection method, characterized in that: include: Acquire a vehicle image of a vehicle to be detected, and determine the vehicle model of the vehicle to be detected; Determine a model tree structure corresponding to the vehicle to be detected according to the vehicle model; wherein the model tree structure includes a plurality of anomaly detection models; Determine a vehicle component to be detected in the vehicle image, and determine a target anomaly detection model corresponding to the vehicle component to be detected in the model tree structure; Based on the vehicle image, each of the vehicle components to be detected is detected by the target anomaly detection model to obtain a corresponding detection result.

2. The vehicle detection method according to claim 1, characterized in that: Before determining the model tree structure corresponding to the vehicle to be detected according to the vehicle model, the method further includes: obtaining vehicles and vehicle maintenance procedures for multiple vehicle models, and obtaining multiple anomaly detection models; Disassembling the vehicles of each vehicle model to obtain vehicle components corresponding to each vehicle; According to the vehicle model and the vehicle components, a basic model tree structure corresponding to each vehicle model is constructed; wherein the basic model tree structure has the corresponding vehicle model as a root node, and each vehicle component is a node other than the root node; According to the vehicle maintenance regulations, determining the detection item points of each of the nodes in the corresponding basic model tree structure; Determining the abnormality detection model corresponding to each of the vehicle components according to the detection item points; The abnormality detection model corresponding to each of the vehicle components is configured into a corresponding node to obtain the model tree structure corresponding to each of the vehicle models.

3. The vehicle detection method according to claim 2, characterized in that: The obtaining of multiple anomaly detection models includes: The device data of a plurality of trackside detection devices are acquired, and an abnormality detection model possessed by each of the trackside detection devices is determined based on the device data.

4. The vehicle detection method according to claim 1, characterized in that: The detecting each of the to-be-detected vehicle components by the target anomaly detection model based on the vehicle image to obtain a corresponding detection result includes: Performing image correction and image segmentation on the vehicle image to obtain a plurality of component images of the vehicle components to be detected; Each component image is input into the corresponding target anomaly detection model for detection to obtain a corresponding detection result.

5. The vehicle detection method according to any one of claims 1 to 4, characterized in that: The abnormality detection model includes at least two of a bolt loss detection model, a lock buckle abnormality detection model, a foreign object detection model, a cable loosening detection model, and a nameplate loss detection model.

6. The vehicle detection method according to any one of claims 1 to 4, characterized in that: Before determining the model tree structure corresponding to the vehicle to be detected according to the vehicle model, the method further includes: If a new anomaly detection model that does not exist in the model tree structure is received, performing compliance verification and availability verification on the new anomaly detection model; If the compliance check and the availability check of the new anomaly detection model pass, the new anomaly detection model is imported into the model tree structure.

7. The vehicle detection method according to any one of claims 1 to 4, characterized in that: After detecting each of the to-be-detected vehicle components by using the target anomaly detection model based on the vehicle image, the method further includes: Determine the running progress and execution time of each of the target anomaly detection models, and display the running progress and execution time on a preset display page.

8. A vehicle detection device, characterized in that: include: A first acquisition module is configured to acquire a vehicle image of a vehicle to be detected and determine a vehicle model of the vehicle to be detected; A first determination module is configured to determine a model tree structure corresponding to the vehicle to be detected according to the vehicle model; wherein the model tree structure includes a plurality of anomaly detection models; A second determination module is configured to determine a vehicle component to be detected in the vehicle image, and determine a target anomaly detection model corresponding to the vehicle component to be detected in the model tree structure; The detection module is configured to detect each of the vehicle components to be detected based on the vehicle image through the target anomaly detection model to obtain corresponding detection results.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the program, the vehicle detection method according to any one of claims 1 to 7 is implemented.

10. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the vehicle detection method according to any one of claims 1 to 7 is implemented.

Images