Vehicle-mounted reversing image automatic testing method, device, equipment and medium
By supporting multi-device parallel testing, CAN signal-based reverse command triggering and reverse picture comparison, the existing on-board reverse image testing problems are solved, and efficient and accurate test results and rapid fault diagnosis are achieved.
Patent Information
- Application Number
- CN202510044422.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-01-11
- Publication Date
- 2025-03-28
AI Technical Summary
The existing vehicle-mounted reversing image testing methods have low testing efficiency and are difficult to truly restore real car scenes. The test results are easily affected by the environment and are difficult to meet the needs of efficiency and accuracy.
By obtaining parallel test information of multiple devices, sending a reverse command based on the CAN signal, triggering the device to enter the reverse mode, collecting the reverse screen information, and comparing it with the trained reverse model. If it is abnormal, obtaining the entire process status mark information to confirm the cause of the failure.
Effectively reduce test time, improve test coverage, ensure the authenticity and credibility of test results, improve the integrity and quality of picture data, accurately locate fault problems, shorten fault detection time, and improve the efficiency, accuracy and fault handling capabilities of tests.
Smart Images
Figure CN119714810A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of vehicle-mounted reversing image testing, and in particular to a vehicle-mounted reversing image automated testing method, device, equipment and medium. Background Art
[0002] The in-vehicle reversing image display function is an important part of the smart cockpit, which is related to driving safety and user experience. Existing reversing image testing methods usually include building a test bench to simulate reversing instructions, integrating an automated test bench with an image recognition system, or relying on manual operation for manual testing. However, these methods have problems such as low test efficiency, difficulty in truly restoring the actual vehicle scene, and test results that are easily affected by the environment, making it difficult to meet the requirements of efficiency and accuracy. Summary of the invention In order to improve the efficiency and accuracy of reversing image testing, the present application provides a method, device, equipment and medium for automated vehicle reversing image testing.
[0003] The above-mentioned invention objective of the present application is achieved through the following technical solutions: An automated vehicle reversing image testing method, the automated vehicle reversing image testing method comprising: Get parallel testing information of multiple devices; Based on the parallel test information, reversing instruction information is sent through a CAN signal to trigger each device under test to enter a reversing mode and generate real-time reversing status information; According to the real-time reversing state information, collecting the reversing picture of each device under test to generate reversing picture information; Comparing the reversing picture information with the trained reversing model to generate a comparison result; If the comparison result information shows that the reversing screen information is abnormal, the reversing full process status mark information of each tested device is obtained, and the cause of the reversing failure is confirmed according to the reversing full process status mark information.
[0004] By adopting the above technical solutions and supporting parallel testing of multiple devices, the test time can be effectively reduced and the test coverage can be improved. The reversing command triggering method based on CAN signals is highly consistent with the actual vehicle reversing mode, ensuring the authenticity and credibility of the test results. The automatic collection and standardized processing of the reversing images can ensure the integrity and high quality of the image data, providing reliable data support for subsequent analysis. The trained reversing model can be used to quickly compare and identify anomalies in the images, accurately locate faults, and shorten fault detection time. By analyzing the status marking information of the entire process, the root cause of the anomaly can be deeply explored, providing a reliable basis for the rapid diagnosis and resolution of the problem, thereby comprehensively improving the efficiency, accuracy and fault handling capabilities of the vehicle-mounted reversing image test.
[0005] In a preferred example, the present application can be further configured as follows: based on the parallel test information, reversing instruction information is sent through a CAN signal to trigger each device under test to enter a reversing mode, and generate real-time reversing status information, including: Based on the parallel test information, the reversing instruction information is generated by an AI diagnostic algorithm, and the reversing instruction information is sent to each of the devices under test through the CAN signal; After successfully receiving the reversing instruction information, each of the tested devices enters the reversing mode and generates the real-time reversing status information.
[0006] By adopting the above technical solution, it is possible to accurately generate reversing command information based on parallel test information, and optimize the sending method of reversing commands through AI diagnostic algorithms to ensure that the command content is adapted to the operating status of each device. By using CAN signals to efficiently transmit reversing command information, it is possible to simulate the reversing trigger process in a real vehicle environment, thereby improving the real-time and accuracy of command sending. Each device under test automatically enters the reversing mode after receiving the reversing command, and can quickly generate real-time reversing status information, thereby achieving a comprehensive test of the reversing function and ensuring the efficiency of the test process and the authenticity of the test results.
[0007] In a preferred example, the present application may be further configured as follows: the reversing instruction information includes: The AI diagnostic algorithm analyzes the device status information in the parallel test information to obtain an analysis result; According to the analysis result, it is determined whether each of the devices under test meets the reversing instruction triggering condition. If each of the devices under test meets the reversing instruction triggering condition, the AI diagnostic algorithm generates the reversing instruction information for each of the devices under test.
[0008] By adopting the above technical solution, the AI diagnostic algorithm can be used to parse the device status information in the parallel test information, and the operating status and initialization conditions of each device can be accurately obtained, thereby ensuring real-time control of the device status during the test process; by judging whether the device meets the reversing command trigger conditions through the analysis results, it can effectively avoid test errors caused by devices that do not meet the conditions participating in the test. When the trigger conditions are met, the AI diagnostic algorithm generates reversing command information for each device to ensure that the command content is highly matched with the device status, thereby improving the adaptability and execution efficiency of the reversing command, and ultimately realizing the intelligence and efficiency of the reversing test.
[0009] In a preferred example, the present application may be further configured as follows: collecting the reversing picture of each device under test according to the real-time reversing state information to generate reversing picture information, including: According to the real-time reversing state information, obtaining the image output information of the reversing camera of each device under test, and converting the image output information into a continuous image data stream; According to the image processing rules, the noise information in the image data stream is filtered, and the effective visual features in the image data stream are extracted to obtain the effective visual feature information, and the reversing picture information is generated based on the extracted effective visual feature information.
[0010] By adopting the above technical solution, the image output information of the reversing camera of each device under test can be efficiently obtained based on the real-time reversing status information, and the image output information can be converted into a continuous image data stream to ensure the integrity and continuity of the image data; by applying image processing rules, the image data stream is subjected to noise filtering and effective visual feature extraction, which can effectively improve the clarity of the image data and the expression quality of key visual features, and the reversing image information is generated based on the extracted effective visual feature information, which can ensure that the image information has high accuracy and availability.
[0011] In a preferred example, the present application may be further configured as follows: comparing the reversing picture information with the trained reversing model to generate a comparison result includes: Calculating image feature values in the reversing picture information, and converting the image feature values into feature vectors; The feature vector is input into the trained reversing model, the matching degree between the feature vector and the standard feature vector in the trained reversing model is calculated, and the comparison result is generated according to the matching degree.
[0012] By adopting the above technical solution, the image features in the reversing picture information can be efficiently calculated, and a feature vector that can comprehensively describe the picture features can be generated, thereby ensuring the integrity and structured expression of the picture features. By inputting the generated feature vector into a trained reversing model and calculating the matching degree with the standard feature vector, the consistency between the picture and the standard reversing scene can be quickly determined, thereby achieving accurate detection of abnormal pictures.
[0013] In a preferred example, the present application can be further configured as follows: if the comparison result information shows that the reversing screen information is abnormal, the reversing full process status mark information of each device under test is obtained, and the reversing fault cause is confirmed according to the reversing full process status mark information, including: Extracting operation step information related to the abnormal node according to the abnormal node recorded in the state mark information of the whole reversing process; The extracted operation step information is compared with the preset abnormal rules to determine the fault type, and based on the fault type and the operation environment information of the abnormal node, the cause of the reversing fault is obtained.
[0014] By adopting the above technical solution, it is possible to accurately extract the operation step information related to the abnormality based on the abnormal nodes recorded in the status mark information of the entire reversing process, thereby achieving complete tracing of the problem occurrence process, and by comparing the extracted operation step information with the preset abnormal rules, the fault type can be quickly determined, the abnormal characteristics can be effectively identified, and the impact of environmental factors on the fault can be further analyzed in combination with the operating environment information of the abnormal node, and the root cause of the abnormal reversing screen can be accurately located, thereby providing a basis for rapid diagnosis and efficient solution of the problem, and improving the fault analysis efficiency and accuracy of the reversing test.
[0015] In a preferred example, the present application can be further configured as follows: comparing the extracted operation step information with the preset abnormal rule to determine the fault type, and obtaining the cause of the reversing fault based on the fault type and the operation environment information of the abnormal node, including: According to the operating environment information, extracting resource call sequence information, interface level priority allocation information and operation timing information recorded during the reversing process to generate operating environment feature information; According to the operating environment characteristic information, analyzing the conflict situation in the resource call sequence information, the abnormal state in the interface level priority allocation information, and the delayed node information in the operation timing information to generate link abnormality information; The cause of the reversing failure is generated by combining the abnormal information of the link with the historical fault record information of the equipment.
[0016] By adopting the above technical scheme, it is possible to accurately extract the resource calling sequence, interface level priority allocation and operation timing information recorded in the reversing process based on the operating environment information, generate comprehensive operating environment characteristic information, and thus provide structured data support for abnormal analysis. By analyzing the operating environment characteristic information, it is possible to effectively identify conflict problems in resource calling, abnormal states in interface priority allocation and delay nodes in operation timing, accurately locate the key links affecting the reversing function, and compare and analyze the abnormal information of the links with the historical fault records of the equipment. It is possible to quickly generate the cause of the reversing fault, thereby providing a clear basis for the fundamental diagnosis and efficient solution of the problem, and improving the reliability of the reversing test and the efficiency of troubleshooting.
[0017] The second object of the invention is achieved by the following technical solutions: An automatic vehicle reversing image testing device, comprising: Parallel test information collection module, used to obtain parallel test information of multiple devices; A reversing mode triggering and state generating module, used to send reversing instruction information through CAN signals based on the parallel test information, trigger each device under test to enter the reversing mode, and generate real-time reversing state information; A reversing picture acquisition and generation module, used to acquire the reversing picture of each device under test according to the real-time reversing state information, and generate reversing picture information; A picture comparison and result generation module, used for comparing the reversing picture information with the trained reversing model to generate a comparison result; The fault analysis and cause confirmation module is used to obtain the reversing full process status mark information of each tested device if the comparison result information shows that the reversing screen information is abnormal, and confirm the cause of the reversing fault according to the reversing full process status mark information.
[0018] By adopting the above technical solutions and supporting parallel testing of multiple devices, the test time can be effectively reduced and the test coverage can be improved. The reversing command triggering method based on CAN signals is highly consistent with the actual vehicle reversing mode, ensuring the authenticity and credibility of the test results. The automatic collection and standardized processing of the reversing images can ensure the integrity and high quality of the image data, providing reliable data support for subsequent analysis. The trained reversing model can be used to quickly compare and identify anomalies in the images, accurately locate faults, and shorten fault detection time. By analyzing the status marking information of the entire process, the root cause of the anomaly can be deeply explored, providing a reliable basis for the rapid diagnosis and resolution of the problem, thereby comprehensively improving the efficiency, accuracy and fault handling capabilities of the vehicle-mounted reversing image test.
[0019] The third objective of the present application is achieved through the following technical solutions: A computer device comprises a memory, a processor and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of the above-mentioned vehicle-mounted reversing image automatic testing method are implemented.
[0020] The fourth objective of the present application is achieved through the following technical solutions: A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps of the above-mentioned vehicle-mounted reversing image automatic testing method.
[0021] In summary, the present application includes at least one of the following beneficial technical effects: 1. By supporting parallel testing of multiple devices, the test time is effectively reduced and the test coverage is improved. The reversing command triggering method based on CAN signals is highly consistent with the actual vehicle reversing mode, ensuring the authenticity and credibility of the test results. The automatic collection and standardized processing of the reversing image can ensure the integrity and high quality of the image data, providing reliable data support for subsequent analysis. The trained reversing model can quickly compare and identify abnormalities on the image, accurately locate the fault problem, shorten the fault detection time, and deeply explore the root cause of the abnormality by analyzing the status mark information of the whole process, providing a reliable basis for the rapid diagnosis and resolution of the problem, thereby comprehensively improving the efficiency, accuracy and fault handling capabilities of the vehicle reversing image test; 2. Based on the abnormal nodes recorded in the status mark information of the whole reversing process, the system can accurately extract the operation step information related to the abnormality, so as to realize the complete tracing of the problem occurrence process. By comparing the extracted operation step information with the preset abnormal rules, the fault type can be quickly determined, the abnormal characteristics can be effectively identified, and the impact of environmental factors on the fault can be further analyzed by combining the operation environment information of the abnormal node. The root cause of the abnormal reversing screen can be accurately located, thereby providing a basis for the rapid diagnosis and efficient solution of the problem, and improving the fault analysis efficiency and accuracy of the reversing test; 3. Based on the operating environment information, it can accurately extract the resource call sequence, interface level priority allocation and operation timing information recorded during the reversing process, generate comprehensive operating environment feature information, and thus provide structured data support for abnormal analysis. Through the analysis of the operating environment feature information, it can effectively identify the conflict problems in resource calls, the abnormal state in the interface priority allocation and the delay nodes in the operation sequence, accurately locate the key links that affect the reversing function, and compare and analyze the link abnormal information with the historical fault records of the equipment. It can quickly generate the cause of the reversing fault, thereby providing a clear basis for the fundamental diagnosis and efficient solution of the problem, and improving the reliability of the reversing test and the efficiency of troubleshooting. BRIEF DESCRIPTION OF THE DRAWINGS
[0022] Figure 1 is a flow chart of an automated vehicle reversing image testing method in one embodiment of the present application; Figure 2 This is a flowchart for implementing step S20 in the vehicle-mounted reversing image automated testing method in one embodiment of the present application; Figure 3 It is a flowchart for implementing step S202 in the vehicle-mounted reversing image automated testing method in one embodiment of the present application; Figure 4 This is a flowchart for implementing step S30 in the vehicle-mounted reversing image automated testing method in one embodiment of the present application; Figure 5 This is a flowchart for implementing step S40 in the vehicle-mounted reversing image automated testing method in one embodiment of the present application; Figure 6 This is a flowchart for implementing step S50 in the vehicle-mounted reversing image automated testing method in one embodiment of the present application; Figure 7 This is a flowchart of the implementation after step S502 in the vehicle-mounted reversing image automatic testing method in one embodiment of the present application; Figure 8 This is a principle block diagram of a vehicle-mounted reversing image automatic testing device in one embodiment of the present application; Fig. 9 It is a schematic diagram of a device in an embodiment of the present application. DETAILED DESCRIPTION
[0023] The present application is further described in detail below in conjunction with the accompanying drawings.
[0024] In one embodiment, if Figure 1 As shown, the present application discloses an automated vehicle reversing image testing method, which specifically includes the following steps: S10: Acquire parallel test information of multiple devices.
[0025] Specifically, by building a test system that supports the connection of multiple devices, a data communication channel is established with the hard-wired interface and CAN bus of each device under test. At the same time, the status information of the device is collected, including the hardware connection status, CAN signal communication status and the initialization status of the camera. The real-time operation data of each device is obtained in parallel through multi-threading technology, and the communication stability and environmental parameters of each device are comprehensively analyzed, such as whether there is signal interference or initialization failure, etc., to finally generate parallel test information of multiple devices.
[0026] S20: Based on the parallel test information, reversing instruction information is sent through the CAN signal to trigger each device under test to enter the reversing mode and generate real-time reversing status information.
[0027] Specifically, according to the status identifier in the parallel test information, corresponding reversing command information is generated for each device. The reversing command information includes the CAN signal message content, the sending timestamp and the target device address. The reversing command is sent to each device through the CAN bus according to the message priority. After the device receives the command, real-time reversing status information is generated according to the feedback. The feedback information includes the successful identifier of the device entering the reversing mode, the initialization status of the reversing screen display module and the real-time response result of the data communication. The real-time reversing status information is generated by combining this information.
[0028] S30: According to the real-time reversing state information, the reversing picture of each device under test is collected to generate reversing picture information.
[0029] Specifically, by analyzing the camera initialization flag in the real-time reversing status information, the device that can capture the image is determined, and the image acquisition interface is called to obtain the reversing image data stream. During the acquisition process, the image restoration algorithm is used to filter out the image noise caused by insufficient ambient light or interference signals. At the same time, the color balance technology is used to adjust the image contrast. Key area identification and marking operations are performed for each frame to ensure that the image contains all necessary reversing scene elements. Finally, the processed image data is converted into standardized reversing image information.
[0030] S40: Compare the reversing image information with the trained reversing model to generate a comparison result.
[0031] Specifically, the reversing image information is input into the trained reversing model. The model uses a built-in feature extraction algorithm to analyze key areas in the image, such as lane edges, obstacle contours, and depth information of the rear-view environment. The generated multi-dimensional feature vectors are compared one by one with the normal reversing scene feature library in the model. The accuracy of the image is judged by calculating the similarity score. For example, when the recognition accuracy of the lane edge in the image is lower than the set threshold, the system will mark the image as abnormal. Finally, a comparison report is generated based on each comparison result, including the distribution of normal and abnormal images and the corresponding feature descriptions.
[0032] S50: If the comparison result information shows that the reversing screen information is abnormal, the reversing full process status mark information of each device under test is obtained, and the reversing fault cause is confirmed according to the reversing full process status mark information.
[0033] Specifically, when the comparison result information shows that the reversing screen is abnormal, the status mark information of the entire process from the reversing command being sent to the screen display is extracted. The status mark information includes the command sending time, the camera startup delay time, the screen transmission frame loss rate, and the interface resource conflict record. For example, in the Bluetooth call or video playback scenario, the interface priority allocation is analyzed to see whether there is an abnormality, and it is determined whether the screen loading failure is caused by the resource call conflict. In the case of resource conflict, the specific abnormal link is further deduced according to the device's operation log and operation record. For example, the subsequent screen loading failed due to the failure to release the surface resource call, thereby confirming the cause of the fault that caused the abnormal reversing function.
[0034] In one embodiment, if Figure 2 As shown, in step S20, based on the parallel test information, the reversing instruction information is sent through the CAN signal to trigger each device under test to enter the reversing mode and generate real-time reversing status information, including: S201: Based on the parallel test information, reversing command information is generated through the AI diagnostic algorithm, and the reversing command information is sent to each device under test through a CAN signal.
[0035] In this embodiment, the AI diagnostic algorithm refers to an intelligent algorithm used to generate and send reversing command information. By comprehensively analyzing the parallel test information of multiple devices, the algorithm automatically generates reversing commands that are adapted to the status of each device. At the same time, the AI diagnostic algorithm uses the optimized CAN signal transmission mechanism to efficiently send commands and parses the status data returned by the device in real time to ensure the triggering of the reversing mode and the accuracy of the status record, further providing support for fault analysis and cause location.
[0036] Specifically, according to the communication status, hardware connection status and reversing mode initialization status of each device under test recorded in the parallel test information, the AI diagnostic algorithm is used to generate reversing command information. The reversing command information includes the command type of the CAN message, the message ID, the target device address and the specific parameter value for triggering the reversing mode, such as the activation flag and initialization parameters of the reversing camera. In the process of generating the reversing command information, the AI diagnostic algorithm dynamically adjusts the message sending order according to the communication priority of the device, and optimizes the sending time interval based on the real-time status of the message queue to ensure that all devices can receive the instructions quickly, and sends the generated reversing command information to each device under test one by one through the CAN bus. Among them, the CAN message is supplemented with a check bit during the sending process, which is used to automatically verify the data integrity after the device receives the message, thereby ensuring the reliability of the command information.
[0037] S202: After successfully receiving the reversing instruction information, each device under test enters the reversing mode and generates real-time reversing status information.
[0038] Specifically, after receiving the reversing command information, each device under test parses the CAN message content in the command, including the reversing camera start signal, the screen display loading command and the reversing function initialization parameters, and performs operations in sequence according to the command content. First, the reversing camera is activated and the shooting function is initialized to ensure the stability of the video stream output. Secondly, the reversing screen is loaded to the display interface, and the screen loading delay is avoided by adjusting the resource allocation priority. After the above operations are completed, the working status of the camera, the initialization completion time of the display interface and related parameters are recorded, and real-time reversing status information is generated in combination with the device's operation log. The status information includes the camera start flag, screen display flag and response delay, which is used for subsequent comprehensive testing of the reversing function.
[0039] In one embodiment, if Figure 3 As shown, in step S202, the reversing instruction information includes: S2021: The AI diagnostic algorithm parses the device status information in the parallel test information to obtain the parsing result.
[0040] Specifically, the AI diagnostic algorithm reads the device status data recorded in the parallel test information one by one, and analyzes the communication status, hardware status and operating environment parameters of the device in combination with the preset parsing rules. The communication status includes the interface connection stability and response time, the hardware status includes the device initialization completion mark and the camera activation mark, and the operating environment parameters include the temperature and humidity around the device and the signal interference intensity. The AI diagnostic algorithm cross-validates the above information, for example, it detects whether the communication interface of the device is working normally and determines whether the hardware module meets the operating standards. When a parameter is abnormal, it is marked as pending status, and finally generates the parsing result of the device. The parsing result includes the device identification that meets the conditions and the specific device status with the abnormality.
[0041] S2022: Based on the analysis results, determine whether each device under test meets the reversing command triggering condition. If each device under test meets the reversing command triggering condition, the AI diagnostic algorithm generates reversing command information for each device under test.
[0042] Specifically, according to the device identification information in the parsing results, each device is checked one by one to see whether it meets the reversing command triggering conditions. The judgment basis includes whether the communication interface status of the device is stable, whether the camera is successfully initialized, and whether the device environmental parameters are within the preset range. For example, when the ambient temperature is higher or lower than the set threshold, the device will be marked as not meeting the conditions. If the device meets the conditions, the AI diagnostic algorithm generates reversing command information based on the device status information. The reversing command information includes the command type of the CAN message, the target device address, and the initialization parameters of the reversing mode, such as specifying the resolution of the camera and the resource allocation priority loaded on the display interface, to ensure that the command can take into account both real-time and accuracy during execution.
[0043] In one embodiment, if Figure 4 As shown, in step S30, the reversing picture of each device under test is collected according to the real-time reversing state information to generate reversing picture information, including: S301: According to the real-time reversing state information, the image output information of the reversing camera of each device under test is obtained, and the image output information is converted into a continuous image data stream.
[0044] Specifically, according to the camera working status of each device under test recorded in the real-time reversing status information, the camera interface is called to obtain real-time screen output information. The screen output information includes continuous video frames and image metadata. During the image metadata acquisition process, the frame sequence correction algorithm is used to complete the incomplete frame sequence for possible frame loss or delay in the screen output. At the same time, dynamic data smoothing technology is used to eliminate possible frame skipping or freezes in the screen. After the stable screen output data is collected, it is converted into a continuous image data stream that conforms to a predetermined format. After encoding, the data stream contains timestamp information, screen resolution identifier, and frame rate parameters, providing a basis for subsequent image analysis and comparison.
[0045] S302: According to the image processing rules, the noise information in the image data stream is filtered, and the effective visual features in the image data stream are extracted to obtain the effective visual feature information, and the reversing picture information is generated based on the extracted effective visual feature information.
[0046] Specifically, the collected image data stream is processed frame by frame according to the image processing rules. First, the random noise in the data stream is removed by the frequency domain filtering method, such as brightness fluctuations caused by sensor interference or ambient light changes. Secondly, the sharpening enhancement algorithm is applied to the edge data in the image to improve the clarity and make the key features in the picture more obvious, such as lane lines and obstacle contours. Then, the region segmentation technology is used to extract the visual areas with significant features in the image, such as the reversing trajectory area and dynamic objects in the rear-view scene. The extracted visual feature information is formatted and stored to generate effective visual feature information. On this basis, combined with the context information in the frame sequence, the complete reversing picture information is generated by constructing a feature combination model.
[0047] In one embodiment, if Figure 5 As shown, in step S40, the reversing picture information is compared with the trained reversing model to generate a comparison result, including: S401: Calculate image feature values in the reversing picture information, and convert the image feature values into feature vectors.
[0048] Specifically, by performing feature extraction operations on each frame of the reversing picture information, the edge detection algorithm is first used to extract the boundary line data in the picture, and the contour features of the reversing trajectory, lane edge and rear-view obstacles are analyzed. Then, the color histograms of different areas in the image are calculated through color distribution analysis, and the color features of key areas are extracted. Subsequently, the texture analysis method is combined, such as the grayscale co-occurrence matrix technology, to calculate the texture change pattern in the picture, and the above feature results are quantized and normalized according to preset coding rules to generate a set of image feature values for each frame, and these sets are converted into standardized feature vectors through a feature vectorization algorithm.
[0049] S402: Input the feature vector into the trained reversing model, calculate the matching degree between the feature vector and the standard feature vector in the trained reversing model, and generate a comparison result according to the matching degree.
[0050] Specifically, the converted feature vectors are input into the trained reversing model in sequence. The reversing model calls the internal feature matching algorithm to compare the input vectors with the standard feature vectors stored in the feature library one by one. The cosine similarity calculation is used to evaluate the matching degree between each group of vectors. For example, the shape overlap between the trajectory lines in the reversing picture and the standard trajectory model, the spatial position difference of the obstacle contour and the consistency of the picture texture distribution are analyzed. According to the matching score, the comparison result is classified as normal or abnormal. If the matching degree is lower than the preset threshold, the detailed information of the abnormal feature vector is recorded and the possible fault characteristics are marked. Finally, the comparison result is output for subsequent analysis and processing.
[0051] In one embodiment, if Figure 6 As shown, in step S50, if the comparison result information shows that the reversing screen information is abnormal, the reversing full process status mark information of each device under test is obtained, and the reversing fault cause is confirmed according to the reversing full process status mark information, including: S501: Extracting operation step information related to the abnormal node according to the abnormal node recorded in the state mark information of the whole reversing process.
[0052] Specifically, by analyzing the abnormal nodes in the status mark information of the entire reversing process, the specific steps where the abnormality occurs are located, such as the reversing camera not being activated, the screen loading delay or the data transmission interruption. For the located abnormal nodes, the associated operation step information is extracted, including the timestamp of the abnormal node, the execution status of the operation instruction and the detailed records of the related resource calls. During the extraction process, the context information of the abnormal node is analyzed, such as the completion status of the previous and next steps and the resource usage log, to ensure the integrity and accuracy of the operation step information.
[0053] S502: Compare the extracted operation step information with the preset abnormal rules to determine the fault type, and obtain the cause of the reversing fault based on the fault type and the operation environment information of the abnormal node.
[0054] Specifically, by comparing the extracted operation step information with the preset exception rules one by one, check whether there are features in the operation steps that match the exception rules, such as signal loss when the camera fails to start, resource scheduling conflicts when the screen loads time out, and frame rate drops when data transmission is interrupted. When the operation step information successfully matches an exception rule, determine the specific fault type based on the fault type marked in the rule, such as resource call failure, interface priority error, or operation timing conflict. Combined with the operating environment information of the abnormal node, such as resource usage records in the test scenario, system response logs, and environmental interference data, analyze the impact of the environment on the fault, and finally generate the cause of the reversing fault.
[0055] In one embodiment, if Figure 7 As shown, in step S502, the extracted operation step information is compared with the preset abnormal rules to determine the fault type. Based on the fault type and the operation environment information of the abnormal node, the cause of the reversing fault is obtained, including: S5021: According to the operating environment information, the resource calling sequence information, the interface level priority allocation information and the operation timing information recorded during the reversing process are extracted to generate the operating environment feature information.
[0056] Specifically, according to the resource usage log and instruction execution record recorded in the operating environment information, the timestamp and calling order of each resource call in the reversing process are parsed, such as the trigger time of the camera initialization instruction and the execution order of the screen loading instruction. At the same time, the interface level priority allocation information is analyzed to check whether the reversing screen fails to load correctly due to the occupation of other interface resources. For example, the priority conflict between the Bluetooth call interface and the reversing screen causes the resource scheduling failure. Combined with the operation timing information, it is identified whether the execution interval of key instructions meets the preset timing standards, such as whether the delay between camera startup and screen loading exceeds the set threshold. The above-mentioned extracted information is formatted into operating environment feature information. The feature information includes resource call chain, interface allocation priority and operation timing characteristics.
[0057] S5022: Analyze the conflict situation in the resource call sequence information, the abnormal state in the interface level priority allocation information, and the delayed node information in the operation timing information according to the operating environment characteristic information to generate link abnormality information.
[0058] Specifically, with respect to the resource call sequence information, check whether there is a call conflict. For example, during the loading of the reversing screen, other applications simultaneously request the same hardware resources, resulting in a call failure. The interface level priority allocation information is analyzed to identify abnormal states caused by incorrect priority settings. For example, the reversing screen fails to cover other non-critical interfaces. The delay nodes in the operation timing information are identified, and the deviation between the actual execution time of key operations and the standard value is calculated. For example, the delay in starting the reversing camera causes the screen display to time out. The abnormal features identified in the above analysis are integrated into link abnormality information. The abnormal information includes the identifier of the conflicting resource, the name of the interface with the wrong priority allocation, and the time stamp of the delay node.
[0059] S5023: Combine the link abnormality information and the historical fault record information of the equipment to generate the reversing fault cause.
[0060] Specifically, by comparing the abnormal features in the link abnormal information with the historical fault records of the equipment, confirm whether the abnormal features match the known fault types, such as whether the resource call conflict caused by the failure to start the camera is consistent with the hardware initialization problem in the history. When the match is successful, supplement the details of the abnormal features according to the fault description in the historical fault record, such as explaining in detail the specific reasons for the resource call failure. If there is no match, generate a new fault type identification based on the new features in the link abnormal information, such as recording the delayed node information as a potential performance bottleneck. Finally, the analysis results are integrated into the cause of the reversing failure, and the specific link that caused the abnormality and the root cause of the reversing failure are clarified.
[0061] It should be understood that the size of the serial numbers of the steps in the above embodiments does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.
[0062] In one embodiment, a vehicle-mounted reversing image automatic testing device is provided, and the vehicle-mounted reversing image automatic testing device corresponds one-to-one to the vehicle-mounted reversing image automatic testing method in the above embodiment. Figure 8 As shown, the vehicle-mounted reversing image automatic test device includes a parallel test information acquisition module, a reversing mode trigger and state generation module, a reversing image acquisition and generation module, an image comparison and result generation module, and a fault analysis and cause confirmation module. The detailed description of each functional module is as follows: Parallel test information collection module, used to obtain parallel test information of multiple devices; The reversing mode triggering and status generation module is used to send reversing command information through CAN signals based on parallel test information, trigger each device under test to enter the reversing mode, and generate real-time reversing status information; The reversing picture acquisition and generation module is used to collect the reversing picture of each device under test according to the real-time reversing state information and generate the reversing picture information; The image comparison and result generation module is used to compare the reversing image information with the trained reversing model and generate a comparison result; The fault analysis and cause confirmation module is used to obtain the reversing full process status mark information of each device under test if the comparison result information shows that the reversing screen information is abnormal, and confirm the cause of the reversing fault based on the reversing full process status mark information.
[0063] Optionally, the reverse mode triggering and state generating module includes: The reversing command generation and sending submodule is used to generate reversing command information through the AI diagnostic algorithm based on the parallel test information, and send the reversing command information to each device under test through the CAN signal; The reversing mode triggering and status generating submodule is used for each device under test to enter the reversing mode and generate real-time reversing status information after successfully receiving the reversing command information.
[0064] Optionally, the reverse mode triggering and state generation submodule includes: The device status analysis unit is used for the AI diagnosis algorithm to analyze the device status information in the parallel test information and obtain the analysis result; The reversing instruction generation unit is used to determine whether each device under test meets the reversing instruction triggering condition based on the analysis results. If each device under test meets the reversing instruction triggering condition, the AI diagnostic algorithm generates reversing instruction information for each device under test.
[0065] Optionally, the reversing image acquisition and generation module includes: The image data stream generation submodule is used to obtain the image output information of the reversing camera of each device under test according to the real-time reversing state information, and convert the image output information into a continuous image data stream; The image processing and reversing screen generation submodule is used to filter the noise information in the image data stream according to the image processing rules, and extract the effective visual features in the image data stream to obtain the effective visual feature information, and generate the reversing screen information based on the extracted effective visual feature information.
[0066] Optionally, the picture comparison and result generation module includes: The image feature value calculation and feature vector generation submodule is used to calculate the image feature value in the reversing picture information and convert the image feature value into a feature vector; The feature vector matching and comparison result generation submodule is used to input the feature vector into the trained reversing model, calculate the matching degree between the feature vector and the standard feature vector in the trained reversing model, and generate a comparison result according to the matching degree.
[0067] Optional, fault analysis and cause confirmation module includes: The abnormal node operation step extraction submodule is used to extract the operation step information related to the abnormal node according to the abnormal node recorded in the state mark information of the whole reversing process; The fault type determination and fault cause generation submodule is used to compare the extracted operation step information with the preset abnormal rules to determine the fault type, and based on the fault type and the operating environment information of the abnormal node, obtain the reversing fault cause.
[0068] Optionally, the fault type determination and fault cause generation submodule includes: An operating environment characteristic information extraction unit is used to extract resource call sequence information, interface level priority allocation information and operation timing information recorded during the reversing process according to the operating environment information, and generate operating environment characteristic information; A link abnormality information generating unit is used to analyze the conflict situation in the resource call sequence information, the abnormal state in the interface level priority allocation information and the delay node information in the operation timing information according to the operation environment characteristic information, so as to generate link abnormality information; The reversing fault cause generating unit is used to generate the reversing fault cause by combining the link abnormality information and the equipment historical fault record information.
[0069] The specific definition of the vehicle-mounted reversing image automated test device can be found in the definition of the vehicle-mounted reversing image automated test method above, which will not be repeated here. Each module in the above-mentioned vehicle-mounted reversing image automated test device can be implemented in whole or in part by software, hardware and a combination thereof. The above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.
[0070] In one embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as follows: Fig. 9As shown. The computer device includes a processor, a memory, a network interface and a database connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a vehicle-mounted reversing image automatic test method is implemented.
[0071] In one embodiment, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the following steps when executing the computer program: Get parallel testing information of multiple devices; Based on the parallel test information, the reversing command information is sent through the CAN signal to trigger each device under test to enter the reversing mode and generate real-time reversing status information; According to the real-time reversing status information, the reversing picture of each device under test is collected to generate reversing picture information; Compare the reversing image information with the trained reversing model to generate a comparison result; If the comparison result information shows that the reversing screen information is abnormal, the reversing full process status mark information of each device under test is obtained, and the cause of the reversing failure is confirmed based on the reversing full process status mark information.
[0072] In one embodiment, a computer readable storage medium is provided, on which a computer program is stored, and when the computer program is executed by a processor, the following steps are implemented: Get parallel testing information of multiple devices; Based on the parallel test information, the reversing command information is sent through the CAN signal to trigger each device under test to enter the reversing mode and generate real-time reversing status information; According to the real-time reversing status information, the reversing picture of each device under test is collected to generate reversing picture information; Compare the reversing image information with the trained reversing model to generate a comparison result; If the comparison result information shows that the reversing screen information is abnormal, the reversing full process status mark information of each device under test is obtained, and the cause of the reversing failure is confirmed based on the reversing full process status mark information.
[0073] Those of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be implemented by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.
[0074] Those skilled in the art can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example. In actual applications, the above-mentioned functions can be distributed and completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above.
[0075] The embodiments described above are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, a person skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such 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 application, and should all be included in the protection scope of the present application.
Claims
1. A method for automatically testing a vehicle-mounted reversing image, characterized in that: The vehicle-mounted reversing image automatic testing method comprises: Get parallel testing information of multiple devices; Based on the parallel test information, reversing instruction information is sent through a CAN signal to trigger each device under test to enter a reversing mode and generate real-time reversing status information; According to the real-time reversing state information, collecting the reversing picture of each device under test to generate reversing picture information; Comparing the reversing picture information with the trained reversing model to generate a comparison result; If the comparison result information shows that the reversing screen information is abnormal, the reversing full process status mark information of each tested device is obtained, and the cause of the reversing failure is confirmed according to the reversing full process status mark information.
2. The vehicle-mounted reversing image automated testing method according to claim 1, characterized in that: Based on the parallel test information, reversing instruction information is sent through a CAN signal to trigger each device under test to enter a reversing mode and generate real-time reversing status information, including: Based on the parallel test information, the reversing instruction information is generated by an AI diagnostic algorithm, and the reversing instruction information is sent to each of the devices under test through the CAN signal; After successfully receiving the reversing instruction information, each of the tested devices enters the reversing mode and generates the real-time reversing status information.
3. The vehicle-mounted reversing image automated testing method according to claim 2, characterized in that: The reversing instruction information includes: The AI diagnostic algorithm analyzes the device status information in the parallel test information to obtain an analysis result; According to the analysis result, it is determined whether each of the devices under test meets the reversing instruction triggering condition. If each of the devices under test meets the reversing instruction triggering condition, the AI diagnostic algorithm generates the reversing instruction information for each of the devices under test.
4. The vehicle-mounted reversing image automated testing method according to claim 1, characterized in that: The step of collecting the reversing picture of each device under test according to the real-time reversing state information and generating reversing picture information includes: According to the real-time reversing state information, obtaining the image output information of the reversing camera of each device under test, and converting the image output information into a continuous image data stream; According to the image processing rules, the noise information in the image data stream is filtered, and the effective visual features in the image data stream are extracted to obtain the effective visual feature information, and the reversing picture information is generated based on the extracted effective visual feature information.
5. The vehicle-mounted reversing image automated testing method according to claim 1, characterized in that: The step of comparing the reversing picture information with the trained reversing model to generate a comparison result includes: Calculating image feature values in the reversing picture information, and converting the image feature values into feature vectors; The feature vector is input into the trained reversing model, the matching degree between the feature vector and the standard feature vector in the trained reversing model is calculated, and the comparison result is generated according to the matching degree.
6. The vehicle-mounted reversing image automated testing method according to claim 1, characterized in that: If the comparison result information shows that the reversing screen information is abnormal, obtaining the reversing full process status mark information of each device under test, and confirming the reversing fault cause according to the reversing full process status mark information, including: Extracting operation step information related to the abnormal node according to the abnormal node recorded in the state mark information of the whole reversing process; The extracted operation step information is compared with the preset abnormal rules to determine the fault type, and based on the fault type and the operation environment information of the abnormal node, the cause of the reversing fault is obtained.
7. The vehicle-mounted reversing image automatic testing method according to claim 6, characterized in that: The extracted operation step information is compared with a preset abnormal rule to determine the fault type, and based on the fault type and in combination with the operation environment information of the abnormal node, the cause of the reversing fault is obtained, including: According to the operating environment information, extracting resource call sequence information, interface level priority allocation information and operation timing information recorded during the reversing process to generate operating environment feature information; According to the operating environment characteristic information, analyzing the conflict situation in the resource call sequence information, the abnormal state in the interface level priority allocation information, and the delayed node information in the operation timing information to generate link abnormality information; The cause of the reversing failure is generated by combining the abnormal information of the link with the historical fault record information of the equipment.
8. An automatic vehicle reversing image testing device, characterized in that: The vehicle-mounted reversing image automatic testing device comprises: Parallel test information collection module, used to obtain parallel test information of multiple devices; A reversing mode triggering and state generating module, used to send reversing instruction information through CAN signals based on the parallel test information, trigger each device under test to enter the reversing mode, and generate real-time reversing state information; A reversing picture acquisition and generation module, used to acquire the reversing picture of each device under test according to the real-time reversing state information, and generate reversing picture information; A picture comparison and result generation module, used for comparing the reversing picture information with the trained reversing model to generate a comparison result; The fault analysis and cause confirmation module is used to obtain the reversing full process status mark information of each tested device if the comparison result information shows that the reversing screen information is abnormal, and confirm the cause of the reversing fault according to the reversing full process status mark information.
9. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the steps of the vehicle-mounted reversing image automated testing method as described in any one of claims 1 to 7 are implemented.
10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of the vehicle-mounted reversing image automated testing method as claimed in any one of claims 1 to 7 are implemented.