License plate state detection method and device, and electronic equipment

Abstract

The disclosure provides a license plate state detection method and device and electronic equipment, relates to the technical field of image processing, and particularly relates to the technical fields of artificial intelligence, computer vision, deep learning and intelligent parking. The specific implementation scheme is: obtaining a first image, the first image including image content of a target vehicle; performing image segmentation on the first image to obtain target information, the target information including target key point information, the target key point information being key point information of a position region of a license plate that can be installed in the target vehicle; performing license plate detection on the target vehicle based on the first image to obtain a detection result; and detecting a license plate state of the target vehicle based on the target information and the detection result.

Description

TECHNICAL FIELD

[0001] The present disclosure relates to the technical field of image processing, in particular to the technical fields of artificial intelligence, computer vision, deep learning and intelligent parking, and specifically relates to a license plate state detection method and device and electronic equipment. BACKGROUND

[0002] In a road parking space automatic management system, obtaining license plate information of a vehicle is very important. Only when the license plate information is captured, can an effective parking order be generated.

[0003] At present, a deep learning model is usually used for license plate detection and character recognition to obtain the license plate information of the vehicle. SUMMARY

[0004] The present disclosure provides a license plate state detection method, device and electronic equipment.

[0005] According to a first aspect of the present disclosure, a license plate state detection method is provided, comprising:

[0006] obtaining a first image, the first image comprising image content of a target vehicle;

[0007] performing image segmentation on the first image to obtain target information, the target information comprising target key point information, the target key point information being key point information of a position region of the target vehicle where a license plate can be installed;

[0008] performing license plate detection on the target vehicle based on the first image to obtain a detection result;

[0009] detecting a license plate state of the target vehicle based on the target information and the detection result.

[0010] According to a second aspect of the present disclosure, a license plate state detection device is provided, comprising:

[0011] an obtaining module configured to obtain a first image, the first image comprising image content of a target vehicle;

[0012] an image segmentation module configured to perform image segmentation on the first image to obtain target information, the target information comprising target key point information, the target key point information being key point information of a position region of the target vehicle where a license plate can be installed;

[0013] a first detection module configured to perform license plate detection on the target vehicle based on the first image to obtain a detection result;

[0014] a second detection module configured to detect a license plate state of the target vehicle based on the target information and the detection result.

[0015] According to a third aspect of the present disclosure, an electronic device is provided, comprising:

[0016] at least one processor; and

[0017] a memory connected with the at least one processor in communication; wherein

[0018] The memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform any one of the methods in the first aspect.

[0019] According to a fourth aspect of the present disclosure, a non-transitory computer readable storage medium storing computer instructions is provided, the computer instructions being used to cause a computer to perform any one of the methods in the first aspect.

[0020] According to a fifth aspect of the present disclosure, a computer program product is provided, comprising a computer program which, when executed by a processor, implements any one of the methods in the first aspect.

[0021] The technical solution according to the present disclosure solves the problem of poor accuracy of license plate state detection, and improves the accuracy of license plate state detection.

[0022] It should be understood that the contents described in this part are not intended to identify key or important features of the embodiments of the present disclosure, nor to limit the scope of the present disclosure. Other features of the present disclosure will become apparent from the following description. BRIEF DESCRIPTION OF DRAWINGS

[0023] The accompanying drawings are used to better understand the present scheme, and do not constitute a limitation on the present disclosure. Among them:

[0024] Figure 1 is a flowchart of a license plate state detection method according to the first embodiment of the present disclosure;

[0025] Figure 2 is a flowchart of determining whether license plate state detection is needed in an intelligent parking scenario;

[0026] Figure 3 is a flowchart of a license plate state detection method of a specific example;

[0027] Figure 4 is a structural schematic diagram of a license plate state detection device according to the second embodiment of the present disclosure;

[0028] Figure 5 is a schematic block diagram of an example electronic device used to implement the embodiments of the present disclosure. DETAILED DESCRIPTION

[0029] Exemplary embodiments of the present disclosure are described herein with reference to the accompanying drawings, which are cited as illustrative examples. Various details of the embodiments of the present disclosure are described herein in order to provide a thorough understanding of the present disclosure. It will be understood by those of ordinary skill in the art that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of the present disclosure. Also, in the following description, descriptions of well-known functions and constructions are omitted for clarity and conciseness.

[0030] First embodiment

[0031] As shown in the first embodiment, the present disclosure provides a license plate state detection method, comprising the following steps: Figure 1

[0032] Step S101: obtaining a first image, the first image comprising image content of a target vehicle.

[0033] In this embodiment, the license plate state detection method relates to the field of image processing, and in particular to the fields of artificial intelligence, computer vision, deep learning, and intelligent parking technology, and can be widely applied in parking and security scenarios. The license plate state detection method of the present disclosure can be executed by the license plate state detection device of the present disclosure. The license plate state detection device of the present disclosure can be configured in any electronic device to execute the license plate state detection method of the present disclosure.

[0034] The first image can be any image comprising image content of a target vehicle, and the target vehicle can refer to a vehicle that needs to be recognized for license plate state. For example, in a vehicle monitoring scenario, a new target appears, which is the vehicle that needs to be recognized for license plate state.

[0035] The first image can be obtained in various ways, such as by a camera to obtain the first image in real time, or by obtaining a pre-stored first image, or by receiving a first image sent by another electronic device.

[0036] In an application scenario such as a parking scenario, a camera can be used to capture a parking vehicle in real time to obtain a real-time monitoring video stream. Based on the video stream, target tracking is performed to determine a target vehicle that needs to be recognized for license plate state. In a case where the latest image frame of the video stream comprises image content of the target vehicle, the latest image frame of the video stream is determined as the first image frame, or the first image is obtained by cutting from the latest image frame of the video stream.

[0037] Step S102: performing image segmentation on the first image to obtain target information, the target information comprising target key point information, the target key point information being key point information of a position region of the target vehicle where a license plate can be installed.

[0038] ​In this step, the first image can be subjected to image segmentation by a deep learning model for image segmentation to obtain the target information.

[0039] The target information can include only the target key point information of the target vehicle, or can include both the target key point information of the target vehicle and the mask image. The target key point information can be key point information of a position region of the target vehicle where a license plate can be installed.

[0040] In an optional embodiment, the first image can be subjected to a segmentation operation to obtain a mask image and key point information of each target in the first image, wherein the mask image and key point information of each target can include a mask image and key point information of the target vehicle.

[0041] In the image segmentation, the number of key points of interest can be 22 in total, and the key point corresponding serial numbers have specific meanings. For example, the position represented by the key point with serial number 9 is the upper left corner of the rear window, the position represented by the key point with serial number 10 is the upper right corner of the rear window, the positions represented by the key points with serial numbers 5, 6, 7, and 8 are the positions of the four vertices of the front of the vehicle, and the position regions indicated by them are the front regions of the vehicle where license plates can be installed, which can be referred to as a front key point group. The positions represented by the key points with serial numbers 11, 12, 13, and 14 are the positions of the four vertices of the rear of the vehicle, and the position regions indicated by them are the rear regions of the vehicle where license plates can be installed, which can be referred to as a rear key point group.

[0042] The target key point information of the target vehicle can be key point information of a position region of the target vehicle where a license plate can be installed, i.e., the target key point information of the target vehicle can include key point information of the front key point group and / or key point information of the rear key point group.

[0043] In one scenario, if the camera is shooting from the front of the vehicle, it only captures the front of the target vehicle, and the target key point information of the target vehicle can only include key point information of the front key point group. In another scenario, if the camera is shooting from the rear of the vehicle, it only captures the rear of the target vehicle, and the target key point information of the target vehicle can only include key point information of the rear key point group. In yet another scenario, if the camera captures the full view of the target vehicle, the target key point information of the target vehicle can include both the key point information of the front key point group and the key point information of the rear key point group.

[0044] It should be noted that the number of key points of interest, the number of key points in the front key point group, and the number of key points in the rear key point group are all examples. In actual applications, a corresponding image segmentation algorithm can be selected for image segmentation according to actual conditions.

[0045] In addition, due to reasons such as occlusion, the head region of the target vehicle may not be completely exposed, for example, the head position of the vehicle is blocked. In this case, the number of key points in the head key point group of the target vehicle may be less than a preset number, for example, 4. When the tail region of the target vehicle is not completely exposed, the number of key points in the tail key point group of the target vehicle may also be less than the preset number.

[0046] Step S103: performing license plate detection on the target vehicle based on the first image to obtain a detection result.

[0047] In this step, a deep learning model for license plate detection can be used to perform license plate detection on the target vehicle based on the first image. Specifically, the first image can be input into the deep learning model for license plate detection, and the deep learning model can output a result indicating whether a license plate is detected.

[0048] In an optional implementation, the result output by the deep learning model can be directly used as the license plate detection result of the target vehicle. If a license plate is detected, it is determined that the license plate is the license plate of the target vehicle. If no license plate is detected, it is determined that no license plate of the target vehicle is detected.

[0049] Since the first image may include multiple vehicles, in order to improve the accuracy of license plate detection of the target vehicle, in another optional implementation, in the case where a license plate is detected, the ownership of the detected license plate can be determined by using the relationship between the position of the detected license plate and the position region indicated by the target key point information, to determine whether the detected license plate is the license plate of the target vehicle. If the ownership of the license plate indicates that it is the license plate of the target vehicle, it is determined that the license plate of the target vehicle is detected. If the ownership of the license plate indicates that there is no license plate of the target vehicle, it is determined that no license plate of the target vehicle is detected.

[0050] Step S104: detecting the license plate state of the target vehicle based on the target information and the detection result.

[0051] In this step, the license plate state can be defined from the dimensions of whether there is a license plate, whether a license plate is recognized this time, whether the license plate is occluded, and the clarity of the license plate. For example, from the dimension of whether there is a license plate, the license plate state can include no license plate and license plate. From the dimension of whether a license plate is recognized, the license plate state can include no license plate recognized this time and license plate recognized this time. From the dimension of the clarity of the license plate, the license plate state can include clear license plate and blurred license plate. From the dimension of whether the license plate is occluded, the license plate state can include environmental occlusion, human occlusion, and no occlusion.

[0052] In an optional implementation, the license plate state can include no license plate recognized this time, blurred license plate, license plate with environmental occlusion, license plate with human occlusion, no license plate, and clear license plate.

[0053] Different manners can be adopted to detect the license plate state of the target vehicle based on the target information according to different detection results, and the detected license plate state is different.

[0054] In a case where the detection result indicates that the license plate of the target vehicle is detected, the first license plate state of the target vehicle can be detected based on the target information, and in a case where the detection result indicates that the license plate of the target vehicle is not detected, the second license plate state of the target vehicle can be detected based on the target information.

[0055] The first license plate state can indicate the license plate attribute of the target vehicle, such as license plate clarity, whether the license plate is blocked, and the like, and the second license plate state indicates the existence of the license plate of the target vehicle, such as no license plate, no license plate identified this time, and the like.

[0056] In an optional embodiment, the license plate state of the target vehicle can be detected based only on the target key point information.

[0057] For example, in a case where the license plate of the target vehicle is detected, it is determined based on the target key point information whether there is a completely exposed front key point group or a completely exposed rear key point group, and if there is an incompletely exposed front key point group and an incompletely exposed rear key point group, it can be determined that the license plate state is that the license plate is blocked by the environment.

[0058] For another example, in a case where the license plate of the target vehicle is not detected, if there is a completely exposed front key point group or a completely exposed rear key point group, it can be determined that the license plate state is that there is no license plate, and if there is an incompletely exposed front key point group and an incompletely exposed rear key point group, it can be determined that the license plate state is that the license plate is not identified this time.

[0059] In another optional embodiment, the license plate state of the target vehicle can be detected based on the mask image of the target vehicle and the target key point information.

[0060] For example, in a case where it is determined that the license plate of the target vehicle is blocked, if there is an incompletely exposed front key point group and an incompletely exposed rear key point group, the position where the front region or the rear region of the target vehicle is blocked can be determined based on the mask image of the target license plate and the target key point information, and it is determined based on the position of the block whether the license plate state is environmental block or artificial block.

[0061] In this embodiment, by acquiring a first image, the first image includes image content of a target vehicle; image segmentation is performed on the first image to obtain target information, the target information includes target key point information, the target key point information is key point information of a position region of a vehicle license plate in the target vehicle; license plate detection is performed on the target vehicle based on the first image to obtain a detection result; and license plate state of the target vehicle is detected based on the target information and the detection result. In this way, the target key point information obtained by image segmentation can be used to detect the license plate state, thereby improving the detection accuracy of the vehicle license plate state and enhancing the vehicle management capability for various license plate states.

[0062] Optionally, in a case where the detection result represents that the license plate of the target vehicle is detected, the license plate state detection of the target vehicle can be implemented in the following manner, and the step S104 specifically includes:

[0063] In a case where the detection result represents that the license plate of the target vehicle is detected, a license plate attribute of the target vehicle is identified;

[0064] In a case where the license plate attribute is identified as a first attribute, an occlusion reason of the target vehicle is identified based on the target information to determine the license plate state of the target vehicle, and the first attribute indicates that the license plate of the target vehicle is occluded;

[0065] In a case where the license plate attribute is identified as a second attribute, the license plate state of the target vehicle is determined based on the second attribute, and the second attribute indicates that the license plate of the target vehicle is not occluded and indicates license plate clarity of the target vehicle.

[0066] In this embodiment, the license plate attribute of the target vehicle can be identified based on a deep learning model for identifying the license plate attribute. Specifically, the first image can be input into the deep learning model, and the deep learning model can output the license plate attribute of the target vehicle. The license plate attribute can be whether the license plate is occluded or blurred.

[0067] If the license plate attribute output by the deep learning model is a first attribute (i.e., indicating that the license plate of the target vehicle is occluded), in this scenario, the reason for the occlusion can be further determined based on the target information to determine whether the license plate state of the target vehicle is environmental occlusion or artificial occlusion. Specifically, based on the mask image of the target license plate and the target key point information, a position of the occlusion in the head region or the tail region of the target vehicle can be determined, and based on the position of the occlusion, it can be determined whether the license plate state is environmental occlusion or artificial occlusion.

[0068] If the license plate attribute output by the deep learning model is a second attribute (i.e., indicating the license plate clarity of the target vehicle), in this scenario, the license plate state of the target vehicle can be determined based on the second attribute. Specifically, if the second attribute indicates that the license plate of the target vehicle is blurred, it is determined that the license plate state of the target vehicle is license plate blurred, and if the second attribute indicates that the license plate of the target vehicle is clear, it is determined that the license plate state of the target vehicle is license plate clear.

[0069] Further, in the vehicle real-time monitoring scenario, in the case where it is determined that the license plate state of the target vehicle is license plate clear, it can be compared whether the license plate of the target vehicle detected this time for the first image is optimized relative to the license plate of the target vehicle detected last time for the image frame captured before. For example, the license plate state detected last time is license plate blurred or no license plate is recognized, and the license plate state detected this time is license plate clear, which indicates that the license plate state detected this time is optimized.

[0070] If the license plate state detected this time is optimized, character recognition of the license plate can be performed for corresponding applications, such as generating a parking order based on license plate characters. If the license plate state detected this time is not optimized, the character detection of the license plate of the target vehicle is abandoned this time.

[0071] In the embodiment, in the case where the detection result represents that the license plate of the target vehicle is detected, the license plate state of the target vehicle can be detected based on the license plate attribute in combination with the target information, and the detection accuracy of the license plate state of the target vehicle is improved.

[0072] Optionally, if the license plate attribute output by the deep learning model indicates that the license plate of the target vehicle is blocked, the reason for the blocking can be further identified to accurately distinguish between license plate environmental blocking and human blocking, and the detection accuracy of the license plate state is improved. The identification of the license plate blocking reason of the target vehicle can be performed in the following manner, which comprises:

[0073] In the case where the number of positions corresponding to the target key point information is greater than or equal to a preset number, a first mask image is obtained, the first mask image comprises a position region mask indicated by the target key point information and intercepted from a second mask image of the target vehicle, and the target information further comprises the second mask image.

[0074] The first pixel point in the first mask image is set to a first value, and the second pixel point in the first mask image is set to a second value to obtain a binary image, the first pixel point is a pixel point belonging to the target vehicle, and the second pixel point is other pixel points in the first mask image except the first pixel point.

[0075] determining a width of each line break in the binary image, the line break representing a region between adjacent first values including the second value;

[0076] in a case where a number of lines of the target line break in the binary image is greater than a first preset threshold and less than a second preset threshold, determining that the license plate state is that the target vehicle exists artificial occlusion, the target line break being a line break with a width greater than a third preset threshold, the second preset threshold being greater than the first preset threshold;

[0077] in a case where the number of lines of the target line break in the binary image is less than or equal to the first preset threshold or the number of lines of the target line break in the binary image is greater than or equal to the second preset threshold, determining that the license plate state is that the target vehicle exists environmental occlusion.

[0078] In this embodiment, first, the exposure degree of the key points in the head region and / or the tail region can be detected, when the number of positions corresponding to the target key point information is greater than or equal to a preset number, it can be determined that the target vehicle exists a completely exposed head key point group or a completely exposed tail key point group, when the number of positions corresponding to the target key point information is less than the preset number, it can be determined that the target vehicle does not exist a completely exposed head key point group or a completely exposed tail key point group. The preset number can be 4.

[0079] For example, if the positions corresponding to the target key point information include positions represented by serial numbers 5, 6, 7 and 8, it can be determined that the target vehicle exists a completely exposed head key point group, or if the positions corresponding to the target key point information include positions represented by serial numbers (11, 12, 13 and 14), it can be determined that the target vehicle exists a completely exposed tail key point group.

[0080] In a case where it is determined that the target vehicle exists a completely exposed head key point group or a completely exposed tail key point group, it indicates that the license plate of the target vehicle exists intermediate occlusion, and the reason for the license plate occlusion can be further determined.

[0081] The position region mask indicated by the target key point information can be cut from the second mask image of the target vehicle obtained by image segmentation to obtain a first mask image, so as to determine the position of the occlusion in the license plate region of the target vehicle based on the first mask image.

[0082] The first pixel points in the first mask image, which are determined to belong to the target vehicle, can be set to 1 (i.e., a first value), and the remaining pixel points, i.e., second pixel points, can be set to 0 (i.e., a second value), to obtain a binary image.

[0083] Afterwards, the width of each row break in the binary image can be determined, which can be specifically counting the number of pixel points with pixel value of 0 between two adjacent pixel points with pixel value of 1 in each row to determine the width of each row break in the binary image. Moreover, before determining the width of each row break, a closing operation can be performed to eliminate small breaks in the binary image to reduce the amount of calculation.

[0084] The number of rows N with a break width greater than a third preset threshold (i.e., a target break) can be counted, because the human shielding of the license plate has higher pertinence, i.e., the shielding is usually at the corresponding height position. If the number of rows N is greater than a first preset threshold and less than a second preset threshold, it is determined that the license plate state of the target vehicle is human shielding. If the number of rows N is less than or equal to the first preset threshold or the number of rows N is greater than or equal to the second preset threshold, it is determined that the license plate state of the target vehicle is environmental shielding.

[0085] The first preset threshold, the second preset threshold, and the third preset threshold can be set according to actual conditions. The first preset threshold and the second preset threshold can be pre-set fixed values, which can be determined based on the height of the license plate. For example, the first preset threshold can be set to 0.8*license plate height, and the second preset threshold can be set to 1.2*license plate height, i.e., in the case of 0.8*license plate height < number of rows N < 1.2*license plate height, it can be determined that the license plate state of the target vehicle is human shielding.

[0086] Optionally, the identifying of the shielding cause of the target vehicle based on the target information to determine the license plate state of the target vehicle further includes:

[0087] In the case that the number of positions corresponding to the target key point information is less than a preset number, it is determined that the license plate state of the target vehicle is environmental shielding.

[0088] In the embodiment, when the number of positions corresponding to the target key point information is less than a preset number, such as 4, it can be determined that the target vehicle does not have a completely exposed head key point group and a tail key point group, and at this time, it can be determined that the license plate state of the target vehicle is environmental shielding. In this way, the judgment of the license plate shielding of the target vehicle can be further implemented.

[0089] Optionally, in the case that the detection result represents that the license plate of the target vehicle is not detected, the license plate state of the target vehicle can be detected by using the following method, and the step S104 further includes:

[0090] In the case that the detection result represents that the license plate of the target vehicle is not detected, if the number of positions corresponding to the target key point information is greater than or equal to a preset number, it is determined that the license plate state of the target vehicle is that the target vehicle does not have a license plate.

[0091] If the number of positions corresponding to the target key point information is less than the preset number, it is determined that the license plate state is that the license plate of the target vehicle is not recognized.

[0092] In this embodiment, if the number of positions corresponding to the target key point information is greater than or equal to the preset number, such as 4, it can be determined that the target vehicle has a completely exposed front key point group or rear key point group, and at this time, it can be determined that the license plate state is that the target vehicle does not have a license plate. If the number of positions corresponding to the target key point information is less than the preset number, it can be determined that the target vehicle does not have a completely exposed front key point group and rear key point group, and at this time, it can be determined that the license plate state is that the license plate of the target vehicle is not recognized, and the image including the image content of the target vehicle is continuously captured for subsequent license plate state and license plate character recognition. In this way, the unlicensed vehicle can be recognized, and the detection accuracy of the license plate state can be improved.

[0093] Optionally, the step S103 specifically includes:

[0094] Performing license plate detection based on the first image;

[0095] In the case of detecting a license plate, determining the detection result based on the relationship between the detected license plate position and the position region indicated by the target key point information; wherein in the case that the license plate position is located in the position region indicated by the target key point information, the detection result is that the license plate of the target vehicle is detected, and in the case that the license plate position is not located in the position region indicated by the target key point information, the detection result is that the license plate of the target vehicle is not detected.

[0096] In the case of not detecting a license plate, determining that the detection result is that the license plate of the target vehicle is not detected.

[0097] In this embodiment, a deep learning model for performing license plate detection can be used to perform license plate detection of the target vehicle based on the first image. Specifically, the first image can be input to the deep learning model for license plate detection, and the deep learning model can output a result of whether a license plate is detected.

[0098] In the case of detecting a license plate, the relationship between the detected license plate position and the position region indicated by the target key point information can be used to determine the ownership of the detected license plate, so as to determine whether the detected license plate is the license plate of the target vehicle. If the license plate position is located in the position region indicated by the target key point information, the ownership of the license plate indicates that it is the license plate of the target vehicle, that is, it is determined that the license plate of the target vehicle is detected, and if the license plate position is not located in the position region indicated by the target key point information, the ownership of the license plate does not indicate that it is the license plate of the target vehicle, and it is determined that the license plate of the target vehicle is not detected.

[0099] In a case where no license plate is detected, it is determined that no license plate of the target vehicle is detected.

[0100] In this embodiment, by identifying the ownership of the vehicle license plate, it is determined whether the license plate belongs to the target vehicle, so that in a case where the first image includes multiple vehicles, the license plate detection accuracy of the target vehicle can be improved.

[0101] Optionally, the step S101 specifically includes:

[0102] Based on the video stream, target tracking is performed to obtain a tracking result, and the video stream includes vehicle images;

[0103] In a case where the tracking result meets a preset condition, it is determined that the latest image frame of the video stream includes image content of the target vehicle.

[0104] An image including the image content of the target vehicle is intercepted from the image frame to obtain the first image.

[0105] In this embodiment, it can be applied to intelligent parking or security scenes, and in the scene, the vehicles passing by can be monitored in real time by a camera, the video stream monitored in real time by the camera is taken as an input of license plate state detection, target detection and tracking matching are performed on the input monitored scene image by a deep learning model for target tracking, and a tracking result is obtained.

[0106] In a case where the tracking result meets a preset condition, it is determined that the latest image frame of the video stream includes image content of the target vehicle. When the tracking result meets the preset condition, it can be indicated that the target requiring license plate recognition is tracked based on the latest image frame of the video stream.

[0107] Then, according to the obtained target requiring license plate state recognition, the position of the target is externally expanded (for example, the length and width are externally expanded by 1.5 times), and a first image is obtained by interception. In this way, the first image can be obtained.

[0108] Optionally, the tracking result meeting the preset condition includes at least one of the following:

[0109] The target identified vehicle is tracked based on the latest image frame of the video stream, and the target identification is a vehicle identification that is not detected in an image frame before the latest image frame of the video stream;

[0110] The vehicle with a license plate state mark being a first license plate state is tracked based on the latest image frame of the video stream, and the vehicle with the first license plate state does not exist other target occlusion, and the first license plate state indicates that no license plate is recognized.

[0111] tracking a vehicle with a license plate state marked as the first license plate state based on the latest image frame of the video stream, and a difference between a distance of the vehicle with the first license plate state and a first target position is greater than a fourth preset threshold, the first target position being a position of the vehicle with the first license plate state in an image frame before the latest image frame of the video stream;

[0112] tracking a vehicle with a license plate state marked as the second license plate state based on the latest image frame of the video stream, the second license plate state indicating that a license plate is blurred or blocked, and the vehicle with the second license plate state not being blocked by other targets;

[0113] tracking a vehicle with a license plate state marked as the second license plate state based on the latest image frame of the video stream, and a difference between a distance of the vehicle with the second license plate state and a second target position is greater than a fifth preset threshold, the second target position being a position of the vehicle with the second license plate state in an image frame before the latest image frame of the video stream.

[0114] In the embodiment, the condition that there is a target requiring license plate state recognition is determined, i.e., the tracking result meets a preset condition, which includes:

[0115] a. a new target appears in the latest image frame of the video stream, i.e., a vehicle identification that has not been detected before is tracked based on the latest image frame of the video stream.

[0116] b. there is a tracking target with an unrecognized license plate in the latest image frame of the video stream, and there is no other target blocking around the target at present, i.e., a vehicle with a license plate state of an unrecognized license plate is tracked based on the latest image frame of the video stream, and a vehicle attribute of the target at present is that there is no blocking of the license plate.

[0117] c. there is a tracking target with an unrecognized license plate in the latest image frame of the video stream, and there is a large movement of the position of the target compared with a last license plate recognition position.

[0118] d. there is a target with a license plate attribute marked as blurred or blocked in the latest image frame of the video stream, and there is no other target blocking around the target at present.

[0119] e. there is a target with a license plate attribute marked as blurred or blocked in the latest image frame of the video stream, and there is a large movement of the position of the target compared with a last license plate recognition position.

[0120] Correspondingly, when the tracking result meets any of the preset conditions, it can be determined that there is a target requiring license plate state detection in the latest image frame of the video stream.

[0121] The license plate state detection method of the embodiment is described in detail below with a specific example.

[0122] Figure 2 is a flowchart of a process for determining whether license plate state detection is needed in an intelligent parking scenario, as shown in Figure 2 , including the following processes:

[0123] Step 201: Obtain a real-time monitoring video stream;

[0124] Step 202: Perform target detection and tracking based on the video stream;

[0125] Step 203: Determine whether license plate state detection is needed; if so, perform step 204; if not, perform step 205;

[0126] Step 204: Perform license plate state detection;

[0127] Step 205: Perform in and out position analysis on the vehicle.

[0128] Figure 3 is a flowchart of a license plate state detection method of a specific example, as shown in Figure 3 , including the following processes:

[0129] Step 301: Take a screenshot of the image area of the target vehicle to obtain a first image;

[0130] Step 302: Perform image segmentation on the first image to obtain a mask image of the target image and target key point information;

[0131] Step 303: License plate detection and attribution determination to determine whether the license plate of the target vehicle is detected; if so, perform step 304; if not, perform step 308;

[0132] Step 304: License plate attribute recognition;

[0133] Step 305: Determine whether the license plate is obstructed; if not, perform step 306; if so, perform step 3010;

[0134] Step 306: Determine whether the license plate state is optimized; if so, perform step 307;

[0135] Step 307: Update the license plate information;

[0136] Step 308: Determine whether the target has a complete and exposed vehicle head key point group or vehicle tail key point group; if so, perform step 309:

[0137] Step 309: Determine that the license plate state is no license plate;

[0138] Step 3010: Determine whether the target has a complete and exposed vehicle head key point group or vehicle tail key point group; if so, perform step 3011; if not, perform step 3013;

[0139] Step 3011: judging the relationship between the height of the shield and the height of the license plate to determine whether it is a human shield; if yes, step 3012 is executed; if no, step 3013 is executed.

[0140] Step 3012: determining that the license plate state is a human shield.

[0141] Step 3013: determining that the license plate state is an environmental shield.

[0142] Second embodiment

[0143] As shown in the first embodiment, Figure 4 The present disclosure provides a license plate state detection device 400, comprising:

[0144] An acquisition module 401 is configured to acquire a first image, wherein the first image comprises image content of a target vehicle.

[0145] An image segmentation module 402 is configured to perform image segmentation on the first image to obtain target information, wherein the target information comprises target key point information, and the target key point information is key point information of a position region of the target vehicle where a license plate can be installed.

[0146] A first detection module 403 is configured to perform license plate detection on the target vehicle based on the first image to obtain a detection result.

[0147] A second detection module 404 is configured to detect a license plate state of the target vehicle based on the target information and the detection result.

[0148] Optionally, the second detection module 404 comprises:

[0149] A first identification unit is configured to identify a license plate attribute of the target vehicle in a case where the detection result indicates that the license plate of the target vehicle is detected.

[0150] A second identification unit is configured to identify a shield reason of the target vehicle based on the target information in a case where the license plate attribute is identified as a first attribute, to determine the license plate state of the target vehicle, wherein the first attribute indicates that the license plate of the target vehicle is shielded.

[0151] A first determination unit is configured to determine the license plate state of the target vehicle based on a second attribute in a case where the license plate attribute is identified as the second attribute, wherein the second attribute indicates that the license plate of the target vehicle is not shielded, and indicates a license plate clarity of the target vehicle.

[0152] Optionally, the second identification unit is specifically configured to:

[0153] In a case where the number of positions corresponding to the target key point information is greater than or equal to a preset number, a first mask image is obtained, the first mask image including a position region mask of the position indicated by the target key point information, which is intercepted from a second mask image of the target vehicle, and the target information further includes the second mask image.

[0154] A first pixel point in the first mask image is set as a first value, and a second pixel point in the first mask image is set as a second value, to obtain a binary image, the first pixel point being a pixel point belonging to the target vehicle, and the second pixel point being any pixel point in the first mask image other than the first pixel point.

[0155] The width of each row break in the binary image is determined, the break representing that the second value is included between adjacent first values.

[0156] In a case where the number of rows of target breaks in the binary image is greater than a first preset threshold and less than a second preset threshold, the license plate state is determined to be that the target vehicle is subjected to artificial shielding, the target break being a break with a width greater than a third preset threshold, and the second preset threshold being greater than the first preset threshold.

[0157] In a case where the number of rows of the target break in the binary image is less than or equal to the first preset threshold or greater than or equal to the second preset threshold, the license plate state is determined to be that the target vehicle is subjected to environmental shielding.

[0158] Optionally, the second identification unit is further configured to:

[0159] In a case where the number of positions corresponding to the target key point information is less than the preset number, the license plate state is determined to be that the target vehicle is subjected to environmental shielding.

[0160] Optionally, the second detection module 404 includes:

[0161] The second determination unit is configured to, in a case where the detection result indicates that the license plate of the target vehicle is not detected, determine the license plate state to be that the target vehicle does not have a license plate, if the number of positions corresponding to the target key point information is greater than or equal to a preset number.

[0162] The third determination unit is configured to, in a case where the number of positions corresponding to the target key point information is less than the preset number, determine the license plate state to be that the license plate of the target vehicle is not identified.

[0163] Optionally, the first detection module 403 is specifically configured to:

[0164] Perform license plate detection based on the first image.

[0165] In a case where the license plate is detected, the detection result is determined based on a relationship between a detected license plate position and a position region indicated by the target key point information, wherein in a case where the license plate position is located in the position region indicated by the target key point information, the detection result is that the license plate of the target vehicle is detected, and in a case where the license plate position is not located in the position region indicated by the target key point information, the detection result is that the license plate of the target vehicle is not detected.

[0166] In a case where the license plate is not detected, the detection result is determined as that the license plate of the target vehicle is not detected.

[0167] Optionally, the acquisition module 401 is specifically configured to:

[0168] perform target tracking based on the video stream to obtain a tracking result, the video stream comprising vehicle images;

[0169] in a case where the tracking result meets a preset condition, determine that the latest image frame of the video stream comprises image content of the target vehicle;

[0170] cut an image comprising the image content of the target vehicle from the image frame to obtain the first image.

[0171] Optionally, the tracking result meeting the preset condition comprises at least one of the following:

[0172] the target vehicle tracked based on the latest image frame of the video stream has a target identification that is not detected in an image frame before the latest image frame of the video stream;

[0173] the target vehicle tracked based on the latest image frame of the video stream has a license plate state marker of a first license plate state, and the target vehicle of the first license plate state does not have other target occlusions, the first license plate state indicating that no license plate is recognized;

[0174] the target vehicle tracked based on the latest image frame of the video stream has a license plate state marker of the first license plate state, and a distance difference between a position of the target vehicle of the first license plate state and a first target position is greater than a fourth preset threshold, the first target position being a position of the target vehicle of the first license plate state in an image frame before the latest image frame of the video stream;

[0175] the target vehicle tracked based on the latest image frame of the video stream has a license plate state marker of a second license plate state, and the target vehicle of the second license plate state does not have other target occlusions, the second license plate state indicating that the license plate is blurred or the license plate has occlusions;

[0176] The latest image frame based on the video stream tracks a vehicle in the second license plate state, and a difference between a distance of the vehicle in the second license plate state and a second target position is greater than a fifth preset threshold, the second target position being a position of the vehicle in the second license plate state in an image frame before the latest image frame of the video stream.

[0177] The license plate state detection device 400 provided by the present disclosure can realize the various processes of the license plate state detection method embodiment, and achieve the same beneficial effects. To avoid repetition, details are not repeated here.

[0178] In the technical solution of the present disclosure, the collection, storage, use, processing, transmission, provision and disclosure of user personal information comply with relevant laws and regulations and do not violate public order and good customs.

[0179] According to embodiments of the present disclosure, the present disclosure further provides an electronic device, a readable storage medium and a computer program product.

[0180] Figure 5 A schematic block diagram of an example electronic device that can be used to implement embodiments of the present disclosure is shown. The electronic device is intended to represent various forms of digital computers, such as laptops, desktops, tablets, personal digital assistants, servers, blade servers, mainframes, and other appropriate computers. The electronic device can also represent various forms of mobile devices, such as personal digital processors, cellular telephones, smart phones, wearable devices, and other similar computing devices. The components shown here, their connections and relationships, and their functions, are meant to be examples only, and are not intended to limit the implementations of the present disclosure described and / or claimed in this document.

[0181] As shown in Figure 5 The device 500 includes a computing unit 501 that can perform various appropriate actions and processes in accordance with a computer program stored in a read-only memory (ROM) 502 or a computer program loaded from a storage unit 508 into a random access memory (RAM) 503. In the RAM 503, various programs and data required for the operation of the device 500 can also be stored. The computing unit 501, the ROM 502, and the RAM 503 are connected to each other through a bus 504. An input / output (I / O) interface 505 is also connected to the bus 504.

[0182] A plurality of components in the device 500 are connected to the I / O interface 505, including: an input unit 506, such as a keyboard, a mouse, etc.; an output unit 507, such as various types of displays, speakers, etc.; a storage unit 508, such as a magnetic disk, an optical disk, etc.; and a communication unit 509, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 509 allows the device 500 to exchange information / data with other devices through a computer network, such as the Internet, and / or various telecommunication networks.

[0183] The computing unit 501 can be various general and / or special-purpose processing components with processing and computing capabilities. Some examples of the computing unit 501 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, a digital signal processor (DSP), and any appropriate processor, controller, microcontroller, etc. The computing unit 501 performs various methods and processes described above, such as the license plate state detection method. For example, in some embodiments, the license plate state detection method can be implemented as a computer software program, which is tangibly embodied in a machine-readable medium, such as the storage unit 508. In some embodiments, part or all of the computer program can be loaded and / or installed onto the device 500 via the ROM 502 and / or the communication unit 509. When the computer program is loaded onto the RAM 503 and executed by the computing unit 501, one or more steps of the license plate state detection method described above can be performed. Alternatively, in other embodiments, the computing unit 501 can be configured to perform the license plate state detection method by any other appropriate means, such as by means of firmware.

[0184] Various implementations of the systems and techniques described above herein can be realized in digital electronic circuitry, integrated circuitry, a field programmable gate array (FPGA), an application specific integrated circuit (ASIC), a system on a chip system (SOC), a programmable logic device (CPLD), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include implementation in one or more computer programs that are executable and / or interpretable on a programmable system including at least one programmable processor, which can be special or general purpose, coupled to receive data and instructions from, and to transmit data and instructions to, a storage system, at least one input device, and at least one output device.

[0185] Program code for carrying out methods of the present disclosure can be written in any combination of one or more programming languages. The program code can be provided to a processor or controller of a general purpose computer, special purpose computer, or other programmable data processing apparatus to produce a machine, such that the program code, when executed by the processor or controller, produces the functions / operations specified in the flowcharts and / or block diagrams. The program code can be executed entirely on a machine, partially on a machine, partially on a machine as a standalone software package, or entirely on a remote machine or server.

[0186] In the context of the present disclosure, a machine-readable medium can be a tangible medium that contains or stores a program for use by or in connection with an instruction execution system, apparatus, or device. The machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can include but is not limited to an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples of the machine-readable storage medium will include one or more lines of electrical connections, portable computer disks, hard disk drives, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or Flash memory), optical fibers, portable compact disc read-only memories (CD-ROMs), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0187] To provide for interaction with a user, the systems and techniques described here can be implemented on a computer having a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user and a keyboard and a pointing device (e.g., a mouse or a trackball) by which the user can provide input to the computer. Other kinds of devices can be used to provide for interaction with a user as well; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form, including acoustic, speech, or tactile input.

[0188] The systems and techniques described here can be implemented in a computing system that includes a back end component (e.g., as a data server), or that includes a middleware component (e.g., an application server), or that includes a front end component (e.g., a user computer having a graphical user interface or a Web browser through which a user can interact with an implementation of the systems and techniques described here), or any combination of such back end, middleware, or front end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include a local area network (LAN), a wide area network (WAN), and the Internet.

[0189] The computer system can include clients and servers. This relationship can be. The servers are typically remote from the clients with the interactions between them occurring over a communication network. The relationship between client and server arises by virtue of computer programs running on the respective computers and having a client-server relationship to each other. The servers can be cloud servers, servers of a distributed system, or servers incorporating blockchain.

[0190] It should be understood that the steps shown in the various forms above can be reordered, added to, or removed. For example, the steps described in the present disclosure can be performed in parallel, in series, or in a different order, without limitation herein, so long as the desired results of the technology disclosed in the present disclosure are achieved.

[0191] The specific embodiments described above are not intended to be limiting, and persons skilled in the art will appreciate that various modifications, combinations, sub-combinations and alternatives can be made to the specific embodiments without departing from the spirit and principles of the disclosure. Any further modifications, equivalents or alternatives within the spirit and principles of the disclosure are to be considered as falling within the scope of the disclosure.

Claims

1. A method for detecting license plate status, comprising: Acquire a first image, which includes image content of the target vehicle; The first image is segmented to obtain target information, which includes target key point information and a second mask image of the target vehicle. The target key point information is the key point information of the location area in the target vehicle where the license plate can be installed. Based on the first image, the license plate of the target vehicle is detected to obtain the detection result; If the detection result indicates that the license plate of the target vehicle has been detected, the license plate attributes of the target vehicle are identified. If the license plate attribute is identified as a first attribute indicating that the license plate of the target vehicle is obscured, and if the number of positions corresponding to the target key point information is greater than or equal to a preset number, a first mask image is obtained. The first mask image includes a mask of the position area indicated by the target key point information, which is extracted from the second mask image. The first pixel in the first mask image is set to a first value, and the second pixel in the first mask image is set to a second value to obtain a binary image. The first pixel is the pixel belonging to the target vehicle, and the second pixel is the other pixels in the first mask image except for the first pixel. Determine the width of the break in each row of the binary graph, wherein the break represents the second value between adjacent first values; If the number of rows at the target fracture point in the binary image is greater than a first preset threshold and less than a second preset threshold, the license plate status is determined to be that the target vehicle is artificially obscured, and the target fracture is a fracture with a width greater than a third preset threshold, where the second preset threshold is greater than the first preset threshold.

2. The method according to claim 1, wherein, The method further includes: If the license plate attribute is identified as the second attribute, the license plate status of the target vehicle is determined based on the second attribute. The second attribute indicates that the license plate of the target vehicle is not obstructed and indicates the clarity of the license plate of the target vehicle.

3. The method according to claim 1, wherein, The method further includes: If the number of rows of the target break in the binary image is less than or equal to the first preset threshold, or if the number of rows of the target break in the binary image is greater than or equal to the second preset threshold, the license plate status is determined to be that the target vehicle is subject to environmental occlusion.

4. The method according to claim 1, wherein, The method further includes: If the number of locations corresponding to the target key point information is less than a preset number, the license plate status is determined to be that the target vehicle is obscured by the environment.

5. The method according to claim 1, wherein, The method further includes: If the detection result indicates that the license plate of the target vehicle was not detected, and if the number of positions corresponding to the target key point information is greater than or equal to a preset number, the license plate status is determined to be that the target vehicle does not have a license plate. If the number of locations corresponding to the target key point information is less than the preset number, the license plate status is determined to be that the license plate of the target vehicle has not been identified.

6. The method according to claim 1, wherein, The step of detecting the license plate of the target vehicle based on the first image to obtain the detection result includes: License plate detection is performed based on the first image; When a license plate is detected, the detection result is determined based on the relationship between the detected license plate position and the location area indicated by the target key point information; wherein, if the license plate position is located within the location area indicated by the target key point information, the detection result is that the license plate of the target vehicle has been detected, and if the license plate position is not located within the location area indicated by the target key point information, the detection result is that the license plate of the target vehicle has not been detected. If no license plate is detected, the detection result is determined to be that no license plate of the target vehicle was detected.

7. The method according to claim 1, wherein, The acquisition of the first image includes: Target tracking is performed based on a video stream to obtain tracking results, wherein the video stream includes vehicle images; If the tracking result meets the preset conditions, it is determined that the latest image frame of the video stream includes the image content of the target vehicle; The first image is obtained by extracting an image containing the image content of the target vehicle from the image frame.

8. The method according to claim 7, wherein, The tracking result meets at least one of the following preset conditions: The vehicle is tracked based on the latest image frame of the video stream, the target identifier being a vehicle identifier that was not detected in image frames preceding the latest image frame of the video stream; Based on the latest image frame of the video stream, a vehicle with a license plate status marked as the first license plate status is tracked, and there are no other targets obstructing the vehicle with the first license plate status. The first license plate status indicates that no license plate has been identified. Based on the latest image frame of the video stream, the vehicle with license plate status marked as the first license plate status is tracked, and the difference between the distance between the position of the vehicle with the first license plate status and the first target position is greater than a fourth preset threshold. The first target position is the position of the vehicle with the first license plate status in the image frame before the latest image frame of the video stream. Based on the latest image frame of the video stream, a vehicle with a license plate status marked as the second license plate status is tracked, and the vehicle with the second license plate status is not obstructed by other targets. The second license plate status indicates that the license plate is blurred or that the license plate is obstructed. The vehicle with the license plate status marked as the second license plate status is tracked based on the latest image frame of the video stream, and the difference between the distance between the location of the vehicle with the second license plate status and the second target location is greater than a fifth preset threshold. The second target location is the location of the vehicle with the second license plate status in the image frame before the latest image frame of the video stream.

9. A license plate status detection device, comprising: The acquisition module is used to acquire a first image, wherein the first image includes the image content of the target vehicle; The image segmentation module is used to segment the first image to obtain target information, the target information including target key point information, the key point information being the key point information of the location area where the license plate can be installed in the target vehicle; The first detection module is used to detect the license plate of the target vehicle based on the first image and obtain the detection result; The second detection module is used to detect the license plate status of the target vehicle based on the target information and the detection results; The second detection module includes: The first identification unit is used to identify the license plate attributes of the target vehicle when the detection result indicates that the license plate of the target vehicle has been detected. The second identification unit is used to identify the cause of the occlusion of the target vehicle based on the target information when the license plate attribute is identified as the first attribute, so as to determine the license plate status of the target vehicle, wherein the first attribute indicates that the license plate of the target vehicle is occluded. The second identification unit is specifically used for: If the number of locations corresponding to the target key point information is greater than or equal to a preset number, a first mask image is obtained. The first mask image includes a mask of the location area indicated by the target key point information, which is extracted from the second mask image of the target vehicle. The target information also includes the second mask image. The first pixel in the first mask image is set to a first value, and the second pixel in the first mask image is set to a second value to obtain a binary image. The first pixel is the pixel belonging to the target vehicle, and the second pixel is the other pixels in the first mask image except for the first pixel. Determine the width of the break in each row of the binary graph, wherein the break represents the second value between adjacent first values; If the number of rows at the target fracture point in the binary image is greater than a first preset threshold and less than a second preset threshold, the license plate status is determined to be that the target vehicle is artificially obscured, and the target fracture is a fracture with a width greater than a third preset threshold, where the second preset threshold is greater than the first preset threshold.

10. The apparatus according to claim 9, wherein, The second detection module also includes: The first determining unit is configured to determine the license plate status of the target vehicle based on the second attribute when the license plate attribute is identified as the second attribute. The second attribute indicates that the license plate of the target vehicle is not obstructed and indicates the clarity of the license plate of the target vehicle.

11. The apparatus according to claim 9, wherein, The second identification unit is further configured to: If the number of rows of the target break in the binary image is less than or equal to the first preset threshold, or if the number of rows of the target break in the binary image is greater than or equal to the second preset threshold, the license plate status is determined to be that the target vehicle is subject to environmental occlusion.

12. The apparatus according to claim 9, wherein, The second identification unit is further configured to: If the number of locations corresponding to the target key point information is less than a preset number, the license plate status is determined to be that the target vehicle is obscured by the environment.

13. The apparatus according to claim 9, wherein, The second detection module includes: The second determining unit is configured to determine that the license plate status is that the target vehicle does not have a license plate when the detection result indicates that the license plate of the target vehicle has not been detected, and if the number of positions corresponding to the target key point information is greater than or equal to a preset number. The third determining unit is used to determine that the license plate status is "the license plate of the target vehicle has not been identified" if the number of positions corresponding to the target key point information is less than the preset number.

14. The apparatus according to claim 9, wherein, The first detection module is specifically used for: License plate detection is performed based on the first image; When a license plate is detected, the detection result is determined based on the relationship between the detected license plate position and the location area indicated by the target key point information; wherein, if the license plate position is located within the location area indicated by the target key point information, the detection result is that the license plate of the target vehicle has been detected, and if the license plate position is not located within the location area indicated by the target key point information, the detection result is that the license plate of the target vehicle has not been detected. If no license plate is detected, the detection result is determined to be that no license plate of the target vehicle was detected.

15. The apparatus according to claim 9, wherein, The acquisition module is specifically used for: Target tracking is performed based on a video stream to obtain tracking results, wherein the video stream includes vehicle images; If the tracking result meets the preset conditions, it is determined that the latest image frame of the video stream includes the image content of the target vehicle; The first image is obtained by extracting an image containing the image content of the target vehicle from the image frame.

16. The apparatus according to claim 15, wherein, The tracking result meets at least one of the following preset conditions: The vehicle is tracked based on the latest image frame of the video stream, the target identifier being a vehicle identifier that was not detected in image frames preceding the latest image frame of the video stream; Based on the latest image frame of the video stream, a vehicle with a license plate status marked as the first license plate status is tracked, and there are no other targets obstructing the vehicle with the first license plate status. The first license plate status indicates that no license plate has been identified. Based on the latest image frame of the video stream, the vehicle with license plate status marked as the first license plate status is tracked, and the difference between the distance between the position of the vehicle with the first license plate status and the first target position is greater than a fourth preset threshold. The first target position is the position of the vehicle with the first license plate status in the image frame before the latest image frame of the video stream. Based on the latest image frame of the video stream, a vehicle with a license plate status marked as the second license plate status is tracked, and the vehicle with the second license plate status is not obstructed by other targets. The second license plate status indicates that the license plate is blurred or that the license plate is obstructed. The vehicle with the license plate status marked as the second license plate status is tracked based on the latest image frame of the video stream, and the difference between the distance between the location of the vehicle with the second license plate status and the second target location is greater than a fifth preset threshold. The second target location is the location of the vehicle with the second license plate status in the image frame before the latest image frame of the video stream.

17. An electronic device comprising: At least one processor; as well as A memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor to enable the at least one processor to perform the method of any one of claims 1-8.

18. A non-transitory computer-readable storage medium storing computer instructions, wherein, The computer instructions are used to cause the computer to perform the method according to any one of claims 1-8.

19. A computer program product comprising a computer program that, when executed by a processor, implements the method according to any one of claims 1-8.

Images