Camera anomaly detection method, electronic equipment and storage device

Abstract

The invention discloses a camera anomaly detection method, electronic equipment and a storage device, and the method comprises the steps: employing a camera of a to-be-detected vehicle in a target scene to carry out the collection of target identifiers matched with a plurality of target stations, and obtaining the collection positions of the target identifiers in a map coordinate system; based on the collection position and a reference position corresponding to the target identifier in the map coordinate system, obtaining a position deviation matched with the target identifier; based on the position deviation, deviation probability distribution matched with the to-be-detected vehicle is obtained; based on the deviation probability distribution and the reference probability distribution, obtaining a camera detection result matched with the to-be-detected vehicle; wherein the reference probability distribution is determined based on deviation probability distribution matched with a plurality of reference vehicles in the target scene. In this way, the accuracy of camera detection can be improved.

Description

Technical Field

[0001] The present application relates to the field of autonomous driving technology, and in particular to a camera anomaly detection method. Background Art

[0002] Automated Guided Vehicle (AGV) systems play an important role in modern manufacturing, and their operating efficiency and safety are highly dependent on the performance of sensing devices such as cameras. However, the cameras on AGVs may experience anomalies due to hardware failures, software problems, or environmental factors, such as blurred images or damaged photosensitive elements. Currently, anomaly detection methods are mainly based on manual detection or the combination of multiple sensors, which has the problems of low detection efficiency and high detection cost.

[0003] In view of this, how to propose a camera detection method with low cost and high detection efficiency has become an urgent problem to be solved. Summary of the invention

[0004] The main technical problem solved by the present application is to provide a camera anomaly detection method, an electronic device and a storage device, which can improve the accuracy of camera detection.

[0005] In order to solve the above technical problems, a technical solution adopted in the present application is: to provide a camera anomaly detection method, including: using the camera of the vehicle to be detected in the target scene to respectively collect target identifications matched with multiple target workstations, and obtain the collection position of the target identification in the map coordinate system; based on the collection position and the reference position corresponding to the target identification in the map coordinate system, obtain the position deviation matching the target identification; based on the position deviation, obtain the deviation probability distribution of the matching of the vehicle to be detected; based on the deviation probability distribution and the reference probability distribution, obtain the camera detection result of the matching of the vehicle to be detected; wherein, the reference probability distribution is determined based on the deviation probability distribution of multiple reference vehicle matchings in the target scene.

[0006] To solve the above technical problems, another technical solution adopted in the present application is: to provide an electronic device, comprising a memory and a processor coupled to each other, wherein the memory stores program instructions, and the processor is used to execute the program instructions to implement the camera abnormality detection method mentioned in the above technical solution.

[0007] In order to solve the above technical problems, another technical solution adopted in the present application is: providing a storage device storing program instructions that can be executed by a processor, wherein the program instructions are used to implement the camera anomaly detection method mentioned in the above technical solution.

[0008] The beneficial effects of the present application are as follows: Different from the prior art, the camera anomaly detection method proposed in the present application collects target identifiers when the vehicle to be detected is docked with multiple target workstations, obtains the collection positions of the corresponding target identifiers in the map coordinate system, and determines the deviation probability distribution based on all the collection positions. The reference probability distribution matched by the camera under normal circumstances is determined by the deviation probability distribution matched by multiple reference vehicles in the target scene. The camera detection result is obtained by comparing the deviation probability distribution matched by the vehicle to be detected with the reference probability distribution for similarity. This method reduces the cost of installing the detection device on the vehicle to be detected; and, the abnormality detection of the camera can be achieved during the operation of the vehicle to be detected, thereby improving the efficiency of detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work. Among them:

[0010] Figure 1 It is a flowchart of an implementation method of the camera abnormality detection method of the present application;

[0011] Figure 2 yes Figure 1 Step S101 corresponds to a flow chart of another implementation method;

[0012] Figure 3 yes Figure 1 A flow chart of an implementation method corresponding to step S104;

[0013] Figure 4 yes Figure 1 Step S104 corresponds to a flow chart of another implementation method;

[0014] Figure 5 yes Figure 4 Step S401 corresponds to a flow chart of another implementation method;

[0015] Figure 6 yes Figure 1 Step S102 corresponds to a flow chart of another implementation method;

[0016] Figure 7 It is a structural schematic diagram of an embodiment of the electronic device of the present application;

[0017] Figure 8 It is a structural diagram of an implementation method of a storage device of the present application. DETAILED DESCRIPTION

[0018] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments, and different embodiments can be adaptively combined. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0019] See also Figure 1 , Figure 1 : is a flow chart of an implementation method of a camera abnormality detection method of the present application, the method comprising:

[0020] S101: using a camera of a vehicle to be detected in a target scene to respectively collect target identifiers matching multiple target workstations, and obtaining collection positions of the target identifiers in a map coordinate system.

[0021] In one embodiment, the vehicle to be detected is located in the target scene, and a camera is provided on the vehicle to be detected to help the vehicle to be detected to achieve automatic docking with at least part of the target stations in the target scene. During the actual working process of the vehicle to be detected, the vehicle to be detected collects the target identification that matches the target station through the camera to verify the position deviation between it and the target station. Therefore, the accuracy of the camera on the vehicle to be detected will directly affect the operation of the car to be detected. In order to detect whether the camera on the above-mentioned vehicle to be detected is abnormal, when the car to be detected is docked with each target station, the camera is used to collect the corresponding target identification, and the collection position of the target identification in the map coordinate system is calculated. Among them, the collection position is the posture of the corresponding target identification in the map coordinate system, which includes the position and direction of the target identification. In addition, the specific position of the target identification matching the target station can be set according to the actual situation.

[0022] In a specific application scenario, the vehicle to be detected is an AGV, which is used to perform handling work in the target scene, and the target station is a shelf. When the goods on the target station need to be transported, the vehicle to be detected moves to the corresponding area and collects the target identifier that matches the target station. In order to verify whether there is an abnormality in the camera on the AGV, when the AGV docks with the corresponding shelf, the collection position obtained by the camera of the AGV to collect the target identifier that matches the shelf is obtained. Among them, the above-mentioned target identifier can be a QR code.

[0023] In another embodiment, in order to ensure the integrity of the data during the camera detection process, for each target station in the target scene, the camera of the vehicle to be detected is used to collect the corresponding target identification, and the collection position of the corresponding target identification in the map coordinate system is obtained. It should be noted that when the vehicle to be detected is not docked with the target station in the target scene, the corresponding collection position is set to a special character, which represents a non-numeric value.

[0024] In a specific application scenario, the special character is NaN (Not a Number).

[0025] S102: Based on the acquisition position and the reference position of the corresponding target identifier in the map coordinate system, obtain a position deviation that matches the target identifier.

[0026] In one embodiment, for each target station matched by the vehicle to be detected, the reference position of the target identifier in the map coordinate system is obtained based on the actual position of the target identifier in the target scene, wherein the actual position of the target identifier in the target scene is predetermined.

[0027] Furthermore, for each target identifier, the coordinate difference between the corresponding acquisition position and the corresponding reference position is used as the position deviation.

[0028] In one implementation scenario, for each target identification matched to the target station where the vehicle to be detected has docked, the corresponding reference position and acquisition position are converted into coordinate representations, and the difference between the coordinates corresponding to the acquisition position and the coordinates corresponding to the reference position is used as the position deviation. The coordinate representation corresponding to the reference position or the acquisition position includes a horizontal coordinate value, a vertical coordinate value, and an angle value.

[0029] In addition, it should be noted that when the vehicle to be detected does not dock with a target workstation in the target scene, the corresponding position deviation is also set to a special character to represent a non-numeric value.

[0030] S103: Based on the position deviation, obtain the deviation probability distribution of the matching of the vehicle to be detected.

[0031] In one embodiment, in response to the vehicle to be detected docking with multiple target stations and obtaining a corresponding position deviation when docking with each target station, the corresponding deviation probability distribution is calculated according to all position deviations corresponding to the vehicle to be detected.

[0032] In one implementation scenario, all position deviations corresponding to the vehicle to be detected are deemed to obey a normal distribution, and the corresponding density function is calculated as the deviation probability distribution corresponding to the vehicle to be detected.

[0033] In another embodiment, a plurality of position deviations are selected according to the docking sequence of the vehicle to be detected to different target workstations, and the deviation probability distribution is calculated based on the selected position deviations.

[0034] In one implementation scenario, according to the docking sequence, the position deviations of the first number of target stations that are closer to the detection time are selected to respectively match, and the deviation probability distribution is calculated. For example, when the vehicle to be detected docks with target station a, target station b, target station c, and target station d, the deviation probability distribution is calculated based on the position deviations obtained when the vehicle to be detected docks with target station b, target station c, and target station d respectively.

[0035] Alternatively, within a preset time period before the detection moment, a corresponding deviation probability distribution is obtained according to the position deviation obtained when the vehicle to be detected docks with the target workstation.

[0036] S104: Based on the deviation probability distribution and the reference probability distribution, obtaining a camera detection result matching the vehicle to be detected; wherein the reference probability distribution is determined based on the deviation probability distribution of multiple reference vehicle matches in the target scene.

[0037] In one embodiment, in response to the target scene including multiple reference vehicles, a deviation probability distribution matched by each reference vehicle is obtained, and a reference probability distribution is determined based on the deviation probability distributions matched by all reference vehicles.

[0038] Furthermore, based on the similarity between the deviation probability distribution of the matching of the vehicle to be detected and the above-mentioned reference probability distribution, the camera detection result of the matching of the vehicle to be detected is obtained.

[0039] The camera anomaly detection method proposed in the present application collects target identifiers when the vehicle to be detected is docked with multiple target workstations, obtains the collection position of the corresponding target identifier in the map coordinate system, and determines the deviation probability distribution based on all the collection positions. The reference probability distribution matched by the camera under normal circumstances is determined by the deviation probability distribution matched by multiple reference vehicles in the target scene. The camera detection result is obtained by comparing the deviation probability distribution matched by the vehicle to be detected with the reference probability distribution for similarity. This method reduces the cost of installing the detection device on the vehicle to be detected; and the abnormality detection of the camera can be realized during the operation of the vehicle to be detected, thereby improving the efficiency of detection.

[0040] See also Figure 2 , Figure 2 yes Figure 1 Step S101 in the flowchart corresponds to another embodiment. Specifically, the implementation process of step S101 includes:

[0041] S201: In response to the vehicle to be detected docking with the target workstation, a first position of the vehicle to be detected in the map coordinate system is obtained.

[0042] In one embodiment, a certain position is arbitrarily selected from the target scene as the origin to construct a map coordinate system, and when the vehicle to be detected docks with the target workstation, the first position of the vehicle to be detected in the map coordinate system is obtained. The first position is the posture of the vehicle to be detected in the map coordinate system, which includes the position and direction of the vehicle to be detected.

[0043] In one implementation scenario, the vehicle to be detected measures road signs in the target scene through corresponding sensors to calculate its own first position in the target scene.

[0044] In another embodiment, a plurality of positioning blocks are installed on the ground in the target scene, and the vehicle to be detected determines its own first position by collecting signals from the positioning blocks.

[0045] S202: Using the camera of the vehicle to be detected to obtain a second position of the target identifier in the camera coordinate system.

[0046] In one embodiment, after the vehicle to be detected arrives at the target station, the target identifier matching the target station is collected by the camera to obtain a second position in the camera coordinate system, wherein the second position is the position and orientation of the corresponding target identifier in the camera coordinate system.

[0047] It should be noted that, in actual application, the first position and the second position may be obtained in another order. For example, step S201 and step S202 are performed simultaneously to obtain the first position and the second position simultaneously.

[0048] S203: Based on the calibration parameters between the camera coordinate system and the map coordinate system, the first position and the second position, obtain a collection position corresponding to the target identifier.

[0049] In one embodiment, the calibration parameters between the camera coordinate system matched by the camera on the vehicle to be detected and the constructed map coordinate system are obtained, and the second position corresponding to the target identifier is converted into the map coordinate system according to the above calibration parameters, the first position and the second position, so as to obtain the collection position of the corresponding target identifier in the map coordinate system. By obtaining the collection position, it is helpful to determine the deviation between the position of the target identifier collected by the vehicle to be detected and the actual position of the target identifier. Among them, the calibration parameters corresponding to the camera on the vehicle to be detected are determined by calibration, and the specific calibration process can refer to the prior art and will not be elaborated in detail here.

[0050] In one implementation scenario, the specific calculation formula for the collection position corresponding to the target identifier is as follows:

[0051]

[0052] in, Indicates the collection position of the target mark in the map coordinate system, Indicates the first position of the vehicle to be detected in the map coordinate system, Indicates the calibration parameters, Indicates the second position of the target marker in the camera coordinate system.

[0053] See also Figure 3 , Figure 3 yes Figure 1 The process flow diagram before step S104 in the embodiment corresponds to an embodiment. In response to the target scene including multiple reference vehicles, the process before step S104 also includes:

[0054] S301: Obtaining deviation probability distributions of reference vehicles respectively matched.

[0055] In one embodiment, for each reference vehicle in the target scene, a deviation probability distribution of its matching is obtained.

[0056] Specifically, when the reference vehicle docks with the target workstation, the third position of the reference vehicle in the map coordinate system is obtained. And, the fourth position of the corresponding target identifier in the camera coordinate system is obtained using the camera of the reference vehicle. Based on the calibration parameters between the camera coordinate system corresponding to the camera on the reference vehicle and the map coordinate system, the third position and the fourth position, the collection position of the corresponding target identifier in the map coordinate system is obtained. Based on the collection position of the target identifier collected by the reference vehicle and the corresponding reference position, the corresponding position deviation is obtained. According to the position deviation obtained when the reference vehicle docks with multiple target workstations, the deviation probability distribution of the reference vehicle matching is obtained.

[0057] Among them, the process of obtaining the deviation probability distribution of the reference vehicle matching is consistent with the process of obtaining the deviation probability distribution of the to-be-detected vehicle matching in the above-mentioned corresponding implementation manner. The specific implementation process can refer to the above-mentioned corresponding implementation manner and will not be elaborated in detail here.

[0058] In another embodiment, in response to the target scene containing multiple candidate vehicles, some reference vehicles are selected from the multiple candidate vehicles based on the number of target workstations for docking.

[0059] In one implementation scenario, a reference number of target workstations that each candidate vehicle has docked with is obtained. Candidate vehicles whose corresponding reference number exceeds a quantity threshold are used as reference vehicles. Reference vehicles are obtained by screening in this way to avoid the situation where some reference vehicles have docked with a small number of target workstations, thereby causing accidental errors in the data to affect the detection results. The above quantity threshold can be set according to actual conditions.

[0060] In another implementation scenario, a reference number of target workstations that each candidate vehicle has docked with is obtained. A second number of reference vehicles is selected from the candidate vehicles whose corresponding reference number exceeds the quantity threshold. The second number of reference vehicles is obtained by screening to avoid excessive number of reference vehicles, which may lead to greater difficulty in subsequently determining the reference probability deviation. The second number may be obtained by estimation, or may be obtained by reverse calculation after multiple experiments.

[0061] S302: Based on the deviation probability distribution matched by the reference vehicle, a reference probability distribution is obtained; wherein the sum of distribution similarities between the reference probability distribution and the deviation probability distributions matched by each reference vehicle is minimum.

[0062] In one embodiment, a reference probability distribution including variable parameters is constructed. The sum of the distribution similarities between the constructed reference probability distribution and the deviation probability distributions matched to the reference vehicles is used as the objective function. By minimizing the calculated value of the objective function, the solution value corresponding to the variable parameter in the reference probability distribution is obtained. The corresponding variable parameter is replaced with the above solution value to obtain an updated reference probability distribution. The reference probability distribution is used to characterize the distribution of position deviations when the vehicle in the target scene docks with each target station under normal camera conditions. By obtaining the reference probability distribution, the deviation probability distribution corresponding to the vehicle to be detected is subsequently compared with the reference probability distribution for similarity, so as to efficiently perform camera abnormality detection.

[0063] In one implementation scenario, the KL divergence (Kullback-Leibler divergence) between the constructed initial reference probability distribution and the deviation probability distribution of any reference vehicle matching is calculated as the corresponding distribution similarity. Based on this, the specific calculation formula of the above reference probability distribution is as follows:

[0064]

[0065] f refer (q) = N(μ refer ,∑ refer )

[0066] Among them, argmin represents the objective function, f sel_j represents the deviation probability distribution of the jth reference vehicle match, M1 represents the number of reference vehicles in the target scene, and f refer represents the reference probability distribution; μ refer and∑ refer They represent the parameters corresponding to the reference probability distribution, namely the expected value and standard deviation.

[0067] See also Figure 4 , Figure 4 yes Figure 1 Step S104 in the flowchart corresponds to another embodiment. Specifically, the implementation process of step S104 includes:

[0068] S401: Obtain the distribution similarity between the deviation probability distribution of each reference vehicle match and the reference probability distribution.

[0069] In one embodiment, the KL divergence between the deviation probability distribution of each reference vehicle match and the reference probability distribution is obtained and used as the distribution similarity between the corresponding deviation probability distribution and the reference probability distribution.

[0070] S402: Based on the distribution similarities corresponding to all reference vehicles, a verification parameter is obtained; wherein the verification parameter is determined based on the standard deviation corresponding to all the distribution similarities.

[0071] In one embodiment, after obtaining the distribution similarity corresponding to each reference vehicle, the distribution similarities corresponding to all reference vehicles are determined to obey the normal distribution, and the corresponding means and standard deviations are calculated. The product of the standard deviations corresponding to all distribution similarities and the preset parameters is used as the verification parameter.

[0072] In one implementation scenario, the preset parameter is 3. The verification parameter is used to determine a target value range, which is used to characterize the range of distribution similarity corresponding to the vehicle to be detected under normal camera conditions.

[0073] S403: Obtain a comparison result between the distribution similarity corresponding to the vehicle to be detected and the verification parameter, and obtain a camera detection result.

[0074] In one embodiment, the distribution similarity corresponding to the vehicle to be detected is compared with the verification parameter to obtain a comparison result, and the camera detection result is determined based on the comparison result.

[0075] Specifically, in response to the distribution similarity corresponding to the to-be-detected vehicle being greater than or equal to the verification parameter, the camera detection result is determined to be normal. In response to the distribution similarity corresponding to the to-be-detected vehicle being less than the verification parameter, the camera detection result is determined to be abnormal.

[0076] See also Figure 5 , Figure 5 yes Figure 4 Step S401 in the flowchart corresponds to another embodiment. Specifically, the implementation process of step S401 includes:

[0077] S501: Obtain a first similarity between a deviation probability distribution of a reference vehicle match and a reference probability distribution, and a second similarity between a reference probability distribution and a deviation probability distribution of a reference vehicle match.

[0078] In one embodiment, for any reference vehicle, the KL divergence between the deviation probability distribution matched by the reference vehicle and the reference probability distribution is calculated and used as the first similarity. Also, for any reference vehicle, the KL divergence between the deviation probability distribution matched by the reference vehicle and the reference probability distribution is calculated and used as the second similarity.

[0079] In one implementation scenario, the specific calculation formulas for the first similarity and the second similarity are as follows:

[0080]

[0081] Among them, k(f1||f refer ) represents the first similarity, k(f refer ||f1) represents the second similarity. f1 represents the deviation probability distribution of the reference vehicle matching, f refer represents the reference probability distribution.

[0082] S502: Based on the first similarity and the second similarity, obtain corresponding distribution similarity.

[0083] In one implementation, a weighted sum is performed on the first similarity and the second similarity to obtain a corresponding distribution similarity.

[0084] In a specific implementation scenario, the sum of the first similarity and the second similarity is used as the corresponding distribution similarity.

[0085] Different from using the first similarity or the second similarity as the distribution similarity, the above scheme combines the first similarity and the second similarity obtained from different directions to obtain the distribution similarity, thereby reducing the influence of a single direction and improving the stability of the distribution similarity.

[0086] See also Figure 6 , Figure 6 yes Figure 1 Step S102 in the flowchart corresponds to another embodiment. In response to the vehicle to be detected docking with the same target station multiple times, the implementation process of step S102 includes:

[0087] S601: For each time when the vehicle to be detected docks with the target workstation, a reference deviation is obtained based on the acquisition position and the reference position of the corresponding target identifier in the map coordinate system.

[0088] In one embodiment, for the same target station, the acquisition position obtained when the vehicle to be detected docks with the target station each time is obtained, and the reference deviation is obtained according to the acquisition position and the reference position matched with the target station. The process of obtaining the reference deviation in this embodiment can refer to the process of obtaining the position deviation in the above corresponding embodiment, and will not be elaborated in detail here.

[0089] S602: Taking the average value of all reference deviations obtained when the vehicle to be inspected docks with the same target workstation as the position deviation.

[0090] In one implementation, for the same target workstation, an average value of all reference deviations obtained when the vehicle to be detected docks with the target workstation for multiple times is used as the position deviation.

[0091] The above scheme obtains the position deviation of the corresponding target identification collected by the vehicle to be detected by taking the average value when the vehicle to be detected docks with the same target station multiple times, thereby reducing the impact of accidental errors on the obtained position deviation, thereby helping to improve the accuracy of camera detection.

[0092] In another embodiment, in response to the vehicle to be detected docking with the same target station multiple times, in order to improve the detection efficiency, the implementation process of step S102 can also be: for each time the vehicle to be detected docks with the target station, based on the collection position and the reference position of the corresponding target mark in the map coordinate system, obtain a reference deviation. From all the reference deviations obtained when the vehicle to be detected docks with the same target station, any reference deviation is randomly selected as the position deviation of the vehicle to be detected to collect the corresponding target mark.

[0093] In yet another embodiment, Figure 1 After step S104, the method further includes: when the camera detection result corresponding to the vehicle to be detected is normal, the vehicle to be detected is used as a reference vehicle, and a new vehicle to be detected is selected from the target scene. By executing the camera detection method corresponding to any of the above embodiments, the camera detection result corresponding to the currently selected vehicle to be detected is obtained.

[0094] Alternatively, when the camera detection result corresponding to the vehicle to be inspected is abnormal, the camera of the vehicle to be inspected is replaced or recalibrated, and the vehicle to be inspected continues to run until the number of docking target workstations exceeds a quantity threshold, and then the camera detection method corresponding to any of the above-mentioned embodiments is re-executed to obtain the camera detection result until the camera detection result is normal.

[0095] See also Figure 7 , Figure 71 is a schematic diagram of the structure of an embodiment of an electronic device of the present application. The electronic device includes: a memory 10 and a processor 20 coupled to each other. The memory 10 stores program instructions, and the processor 20 is used to execute the program instructions to implement the camera anomaly detection method mentioned in any of the above embodiments. Specifically, the electronic device includes but is not limited to: a desktop computer, a laptop computer, a tablet computer, a server, etc., which are not limited here. In addition, the processor 20 can also be called a CPU (Center Processing Unit). The processor 20 may be an integrated circuit chip with signal processing capabilities. The processor 20 may also be a general-purpose processor, a digital signal processor (Digital Signal Processor, DSP), an application-specific integrated circuit (Application Specific Integrated Circuit, ASIC), a field-programmable gate array (Field-Programmable Gate Array, FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc. In addition, the processor 20 may be implemented by an integrated circuit chip.

[0096] See also Figure 8 , Figure 8 The storage device 30 stores program instructions 40 that can be executed by a processor, and when the program instructions 40 are executed by the processor, the camera abnormality detection method mentioned in any of the above embodiments is implemented.

[0097] In the several embodiments provided in the present application, it should be understood that the disclosed methods and devices can be implemented in other ways. For example, the device implementation described above is only schematic. For example, the division of modules or units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, and the indirect coupling or communication connection of devices or units can be electrical, mechanical or other forms.

[0098] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

[0099] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0100] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or all or 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, server, or network device, etc.) or a processor (processor) to perform all or part of the steps of each implementation method of the present application. 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 and other media that can store program code.

[0101] The above description is only an implementation method of the present application, and does not limit the patent scope of the present application. Any equivalent structure or equivalent process transformation made using the contents of the present application specification and drawings, or directly or indirectly used in other related technical fields, are also included in the patent protection scope of the present application.

Claims

1. A camera anomaly detection method, characterized in that: include: Using the camera of the vehicle to be detected in the target scene, target identifiers matching multiple target workstations are collected respectively to obtain the collection positions of the target identifiers in the map coordinate system; Based on the acquisition position and a reference position corresponding to the target identifier in the map coordinate system, obtaining a position deviation matching the target identifier; Based on the position deviation, obtaining a deviation probability distribution of the matching of the to-be-detected vehicle; Based on the deviation probability distribution and the reference probability distribution, a camera detection result of the to-be-detected vehicle matching is obtained; wherein the reference probability distribution is determined based on the deviation probability distribution of multiple reference vehicle matchings in the target scene.

2. The method according to claim 1, characterized in that The method of using the camera of the vehicle to be detected in the target scene to respectively collect target identifiers matched with multiple target workstations to obtain the collection positions of the target identifiers in the map coordinate system includes: In response to the to-be-detected vehicle docking with the target workstation, acquiring a first position of the to-be-detected vehicle in the map coordinate system; and, Acquire a second position of the target identifier in a camera coordinate system using a camera of the vehicle to be detected; Based on the calibration parameters between the camera coordinate system and the map coordinate system, the first position and the second position, the collection position corresponding to the target identifier is obtained.

3. The method according to claim 1, characterized in that The acquiring a position deviation matching the target identifier based on the acquisition position and a reference position corresponding to the target identifier in the map coordinate system includes: Based on the position of each target identifier in the target scene, obtaining a reference position of the target identifier in the map coordinate system; For each target identifier, the coordinate difference between the corresponding acquisition position and the corresponding reference position is used as the position deviation.

4. The method according to claim 1, characterized in that: In response to the target scene including a plurality of reference vehicles, the vehicle to be detected is selected from the plurality of reference vehicles, and before obtaining the camera detection result matching the vehicle to be detected based on the deviation probability distribution and the reference probability distribution, the method includes: Obtaining deviation probability distributions of the reference vehicles respectively matched; Based on the deviation probability distribution matched by the reference vehicle, the reference probability distribution is obtained; wherein the sum of distribution similarities between the reference probability distribution and the deviation probability distributions matched by each of the reference vehicles is minimum.

5. The method according to claim 4, characterized in that The obtaining, based on the deviation probability distribution and the reference probability distribution, a camera detection result matching the vehicle to be detected includes: Obtaining distribution similarity between the deviation probability distribution matched by each reference vehicle and the reference probability distribution; Based on the distribution similarities corresponding to all the reference vehicles, obtaining a verification parameter; wherein the verification parameter is determined based on the standard deviation corresponding to all the distribution similarities; Obtain a comparison result between the distribution similarity corresponding to the vehicle to be detected and the verification parameter, and obtain the camera detection result.

6. The method according to claim 5, characterized in that The obtaining of the distribution similarity between the deviation probability distribution of each reference vehicle match and the reference probability distribution includes: Obtaining a first similarity between the deviation probability distribution matched by the reference vehicle and the reference probability distribution, and a second similarity between the reference probability distribution and the deviation probability distribution matched by the reference vehicle; Based on the first similarity and the second similarity, the corresponding distribution similarity is acquired.

7. The method according to claim 4, characterized in that Also includes: In response to the target scene including a plurality of candidate vehicles, obtaining a reference number of the target workstations that each of the candidate vehicles has docked with; The candidate vehicles whose corresponding reference quantity exceeds a quantity threshold are used as the reference vehicles.

8. The method according to claim 1, characterized in that In response to the to-be-detected vehicle docking with the same target workstation multiple times, obtaining a position deviation matching the target identifier based on the acquisition position and a reference position of the corresponding target identifier in the map coordinate system includes: For each time when the vehicle to be detected docks with the target workstation, a reference deviation is obtained based on the acquisition position and a reference position of the corresponding target identifier in the map coordinate system; The average value of all the reference deviations obtained when the vehicle to be detected docks with the same target workstation is used as the position deviation.

9. An electronic device, characterized in that: It comprises a memory and a processor coupled to each other, wherein the memory stores program instructions, and the processor is used to execute the program instructions to implement the camera abnormality detection method according to any one of claims 1 to 8.

10. A storage device, characterized in that: Program instructions that can be executed by a processor are stored, and the program instructions are used to implement the camera abnormality detection method according to any one of claims 1 to 8.

Images