Method, device, electronic device and storage medium for detecting accident vehicle license plate number

Abstract

The present application provides a method, device, electronic device, and storage medium for detecting the license plate number of an accident vehicle. The method comprises: upon acquiring an image of an accident vehicle captured by a target inspection vehicle and containing a preset accident identifier, retrieving at least one image of the accident vehicle captured by the target inspection vehicle within a first preset time period before and after the capture of the image of the accident vehicle, thereby obtaining at least two images of the accident vehicle; performing license plate character recognition on the at least two images of the accident vehicle, thereby obtaining at least two first license plate character recognition results corresponding to the at least two images of the accident vehicle; and determining the license plate number of the accident vehicle based on the frequency of occurrence of characters in the at least two first license plate character recognition results. The present application can determine the license plate number of the accident vehicle by combining the character recognition results of multiple images of the accident vehicle, thereby improving the accuracy of detecting the license plate number of the accident vehicle.

Description

Technical Field

[0001] The present application relates to the field of image detection technology, and in particular to a method, device, electronic device and storage medium for detecting the license plate number of an accident vehicle. Background Art

[0002] Vehicle accidents, vehicle breakdowns, and pedestrian activity on highways are all extremely dangerous. If not addressed promptly, they can lead to even greater traffic accidents. Currently, these incidents are primarily monitored through public alerts, QR code reporting, and highway patrols. Active patrols are costly and incomplete, while public reports often fall short of timely coverage. Algorithms have also been designed to identify dangerous highway incidents and extract license plates from vehicles involved in accidents. However, existing technologies mostly rely on single-frame image recognition, which is difficult to accomplish with single-frame images and results in low accuracy. This makes it difficult to identify the license plate number in a timely manner, hindering timely contact with the driver of the vehicle involved and preventing timely action on traffic accidents or vehicle breakdowns.

[0003] That is, the accuracy of detecting the license plate number of the accident vehicle in the existing technology is low. Summary of the Invention

[0004] The present application aims to provide a method, device, electronic device and storage medium for detecting the license plate number of an accident vehicle, aiming to solve the problem of low accuracy in detecting the license plate number of an accident vehicle in the prior art.

[0005] On the one hand, the present application provides a method for detecting the license plate number of an accident vehicle, the method comprising:

[0006] When an image of an accident vehicle containing a preset accident identifier is acquired and taken by a target inspection vehicle, at least one image of the accident vehicle taken by the target inspection vehicle within a first preset time period before and after the image of the accident vehicle is acquired, to obtain at least two images of the accident vehicle;

[0007] Performing license plate character recognition on the at least two accident vehicle images to obtain at least two first license plate character recognition results corresponding to the at least two accident vehicle images;

[0008] The license plate number of the accident vehicle is determined based on the appearance frequencies of the characters in the at least two first license plate character recognition results.

[0009] Wherein, the license plate number of the accident vehicle includes a plurality of alphanumeric characters, and the alphanumeric characters include any one of letters and numbers;

[0010] The determining the license plate number of the accident vehicle based on the frequencies of occurrence of characters in the at least two first license plate character recognition results comprises:

[0011] Screening the at least two first license plate character recognition results to obtain at least two second license plate character recognition results, wherein the number of alphanumeric characters in the second license plate character recognition results is greater than a first preset number;

[0012] The license plate number of the accident vehicle is determined based on the appearance frequencies of the characters in the at least two second license plate character recognition results.

[0013] Wherein, determining the license plate number of the accident vehicle based on the frequencies of occurrence of characters in the at least two second license plate character recognition results includes:

[0014] determining whether a third license plate character recognition result exists in the at least two second license plate character recognition results, wherein the number of alphanumeric characters in the third license plate character recognition result is less than a second preset number;

[0015] If so, the plurality of second license plate character recognition results are complemented based on the preset characters to obtain a plurality of fourth license plate character recognition results, wherein the number of alphanumeric characters in the fourth license plate character recognition results is the second preset number.

[0016] The method of supplementing the plurality of second license plate character recognition results based on the preset characters to obtain a plurality of fourth license plate character recognition results includes:

[0017] Arrange and combine the alphanumeric characters in each of the fourth license plate character recognition results to obtain a plurality of fifth license plate character recognition results, wherein the relative positions of the alphanumeric characters in the fourth license plate character recognition results remain unchanged;

[0018] determining a weight coefficient of each of the fifth license plate character recognition results based on the number of the preset characters in each of the fifth license plate character recognition results;

[0019] calculating the frequency of occurrence of the alphanumeric characters in each alphanumeric position of the accident vehicle license plate number based on the plurality of fifth license plate character recognition results and the weight coefficient;

[0020] The alphanumeric character that appears most frequently at each alphanumeric position of the accident vehicle license plate number is determined as the alphanumeric character at each alphanumeric position of the accident vehicle license plate number.

[0021] Wherein, the license plate number of the accident vehicle also includes Chinese characters,

[0022] The determining the license plate number of the accident vehicle based on the frequencies of occurrence of characters in the at least two second license plate character recognition results comprises:

[0023] Counting the occurrence frequency of each Chinese character based on the second license plate character recognition result;

[0024] Obtaining a target Chinese character with the highest appearance frequency and an appearance frequency greater than a preset frequency from each of the Chinese characters;

[0025] The target Chinese character is determined to be the Chinese character of the license plate of the accident vehicle.

[0026] The preset accident sign includes at least one of a warning triangle, a lit double flash light, and an ordinary pedestrian;

[0027] When an image of an accident vehicle containing a preset accident identifier is acquired and taken by a target inspection vehicle, at least one image of the accident vehicle taken by the target inspection vehicle within a first preset time period before and after the image of the accident vehicle is retrieved to obtain at least two images of the accident vehicle, before which the method includes:

[0028] Acquire a first vehicle image captured by a target inspection vehicle;

[0029] Detecting whether the preset accident mark exists in the first vehicle image;

[0030] If the preset accident mark exists on the first vehicle image, the first vehicle image is determined as the accident vehicle image.

[0031] If the preset accident mark exists on the first vehicle image, determining the first vehicle image as the accident vehicle image includes:

[0032] If the preset accident mark exists on the first vehicle image, retrieving at least one second vehicle image taken by the target inspection vehicle within a second preset time period before and after the first vehicle image is taken;

[0033] performing preset accident sign detection on each of the at least one second vehicle images;

[0034] When the proportion of vehicle images in which a preset accident mark is detected in the at least one second vehicle image and the first vehicle image meets a preset proportion condition, the first vehicle image is determined as the accident vehicle image.

[0035] On the one hand, the present application provides a device for detecting the license plate number of an accident vehicle, the device comprising:

[0036] a retrieval unit configured to retrieve, when an image of an accident vehicle containing a preset accident identifier is acquired and taken by a target inspection vehicle, at least one image of the accident vehicle taken by the target inspection vehicle within a first preset time period before and after the image of the accident vehicle is taken, thereby obtaining at least two images of the accident vehicle;

[0037] a license plate character recognition unit, configured to perform license plate character recognition on the at least two accident vehicle images to obtain at least two first license plate character recognition results corresponding to the at least two accident vehicle images;

[0038] A determination unit is used to determine the license plate number of the accident vehicle based on the appearance frequency of characters in the at least two first license plate character recognition results.

[0039] Wherein, the license plate number of the accident vehicle includes a plurality of alphanumeric characters, and the alphanumeric characters include any one of letters and numbers;

[0040] The determining unit is further configured to screen the at least two first license plate character recognition results to obtain at least two second license plate character recognition results, wherein the number of alphanumeric characters in the second license plate character recognition results is greater than a first preset number;

[0041] The license plate number of the accident vehicle is determined based on the appearance frequencies of the characters in the at least two second license plate character recognition results.

[0042] The determining unit is further configured to determine whether a third license plate character recognition result exists in the at least two second license plate character recognition results, wherein the number of alphanumeric characters in the third license plate character recognition result is less than a second preset number;

[0043] If so, the plurality of second license plate character recognition results are complemented based on the preset characters to obtain a plurality of fourth license plate character recognition results, wherein the number of alphanumeric characters in the fourth license plate character recognition results is the second preset number.

[0044] The determining unit is further configured to arrange and combine the alphanumeric characters in each of the fourth license plate character recognition results to obtain a plurality of fifth license plate character recognition results, wherein the relative positions of the alphanumeric characters in the fourth license plate character recognition results remain unchanged;

[0045] determining a weight coefficient of each of the fifth license plate character recognition results based on the number of the preset characters in each of the fifth license plate character recognition results;

[0046] calculating the frequency of occurrence of the alphanumeric characters in each alphanumeric position of the accident vehicle license plate number based on the plurality of fifth license plate character recognition results and the weight coefficient;

[0047] The alphanumeric character that appears most frequently at each alphanumeric position of the accident vehicle license plate number is determined as the alphanumeric character at each alphanumeric position of the accident vehicle license plate number.

[0048] Wherein, the license plate number of the accident vehicle also includes Chinese characters,

[0049] The determining unit is further configured to count the occurrence frequencies of each Chinese character based on the second license plate character recognition result;

[0050] Obtaining a target Chinese character with the highest appearance frequency and an appearance frequency greater than a preset frequency from each of the Chinese characters;

[0051] The target Chinese character is determined to be the Chinese character of the license plate of the accident vehicle.

[0052] The preset accident sign includes at least one of a warning triangle, a lit double flash light, and an ordinary pedestrian;

[0053] The retrieving unit is further configured to obtain a first vehicle image captured by a target inspection vehicle;

[0054] Detecting whether the preset accident mark exists in the first vehicle image;

[0055] If the preset accident mark exists on the first vehicle image, the first vehicle image is determined as the accident vehicle image.

[0056] The retrieval unit is further configured to retrieve, if the preset accident mark exists on the first vehicle image, at least one second vehicle image captured by the target inspection vehicle within a second preset time period before and after the first vehicle image is captured;

[0057] performing preset accident sign detection on each of the at least one second vehicle images;

[0058] When the proportion of vehicle images in which a preset accident mark is detected in the at least one second vehicle image and the first vehicle image meets a preset proportion condition, the first vehicle image is determined as the accident vehicle image.

[0059] In one aspect, the present application further provides an electronic device, comprising:

[0060] one or more processors;

[0061] Memory; and

[0062] One or more applications, wherein the one or more applications are stored in the memory and configured to be executed by the processor to implement the method for detecting the license plate number of the accident vehicle according to any one of the first aspects.

[0063] On the one hand, the present application also provides a computer-readable storage medium on which a computer program is stored, and the computer program is loaded by a processor to execute the steps in the method for detecting the license plate number of an accident vehicle as described in any one of the first aspects.

[0064] The present application provides a method for detecting the license plate number of an accident vehicle. The method comprises: when an image of an accident vehicle captured by a target inspection vehicle and containing a preset accident logo is acquired, retrieving at least one image of the accident vehicle captured by the target inspection vehicle within a first preset time period before and after the capture of the image of the accident vehicle, thereby obtaining at least two images of the accident vehicle; performing license plate character recognition on the at least two images of the accident vehicle, thereby obtaining at least two first license plate character recognition results corresponding to the at least two images of the accident vehicle; and determining the license plate number of the accident vehicle based on the frequency of occurrence of the characters in the at least two first license plate character recognition results. After acquiring the image of the accident vehicle, the present application retrieves at least one image of the accident vehicle related to the image of the accident vehicle, thereby obtaining at least two images of the accident vehicle, and determines the vehicle license plate number based on the frequency of occurrence of each character in the character recognition results of the at least two images of the accident vehicle. The license plate number of the accident vehicle can be determined by combining the character recognition results of multiple images of the accident vehicle, thereby improving the accuracy of the detection of the license plate number of the accident vehicle. BRIEF DESCRIPTION OF THE DRAWINGS

[0065] 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 those skilled in the art, other drawings can be obtained based on these drawings without creative work.

[0066] Figure 1 A schematic diagram of a scene of a detection system for the license plate number of an accident vehicle provided in an embodiment of the present application;

[0067] Figure 2 This is a flow chart of an embodiment of a method for detecting the license plate number of an accident vehicle provided in an embodiment of the present application;

[0068] Figure 3 This is a schematic structural diagram of an embodiment of a device for detecting the license plate number of an accident vehicle provided in an embodiment of the present application;

[0069] Figure 4 It is a schematic diagram of the structure of an embodiment of an electronic device provided in the embodiments of the present application. DETAILED DESCRIPTION

[0070] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without making creative efforts are within the scope of protection of this application.

[0071] In the description of the present application, it should be understood that the terms "center", "longitudinal", "lateral", "length", "width", "thickness", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside" and the like indicate orientations or positional relationships based on the orientations or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present application. In addition, the terms "first" and "second" are used for descriptive purposes only and cannot be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Therefore, the features defined as "first" and "second" may explicitly or implicitly include one or more features. In the description of the present application, "multiple" means two or more, unless otherwise clearly and specifically defined.

[0072] In this application, the word "exemplary" is used to mean "serving as an example, illustration, or illustration." Any embodiment described in this application as "exemplary" is not necessarily to be construed as preferred or advantageous over other embodiments. The following description is given to enable any person skilled in the art to implement and use the present application. In the following description, details are listed for the purpose of explanation. It should be understood that one of ordinary skill in the art can recognize that the present application can be implemented without using these specific details. In other instances, well-known structures and processes are not elaborated in detail to avoid obscuring the description of the present application with unnecessary details. Therefore, the present application is not intended to be limited to the embodiments shown, but is consistent with the widest scope consistent with the principles and features disclosed in this application.

[0073] It should be noted that since the method of the embodiment of the present application is executed in an electronic device, the processing objects of each electronic device exist in the form of data or information. For example, time is actually time information. It can be understood that if size, quantity, position, etc. are mentioned in subsequent embodiments, the corresponding data exist for the electronic device to process. The details will not be repeated here.

[0074] The embodiments of the present application provide a method, device, electronic device and storage medium for detecting the license plate number of an accident vehicle, which are described in detail below.

[0075] See also Figure 1 , Figure 1 The scene diagram of the accident vehicle license plate number detection system provided in the embodiment of the present application is as follows. The accident vehicle license plate number detection system may include an electronic device 100, in which an accident vehicle license plate number detection device is integrated, such as Figure 1 electronic devices in the.

[0076] Furthermore, the accident vehicle license plate detection system is connected to multiple patrol vehicle networks, thereby receiving information sent by these networks. Information such as video, images, GPS coordinates, camera heading angles, and timestamps collected by the patrol vehicles are transmitted back to the accident vehicle license plate detection system in the form of messages. Each patrol vehicle has an inspection number, and the accident vehicle license plate detection system tracks the target patrol vehicle based on the patrol vehicle number.

[0077] Specifically, patrol vehicles are equipped with video acquisition devices, such as cameras. The video acquisition device is responsible for collecting data and performing preliminary processing of the perception data. Considering that the relative speed between the patrol vehicle and the accident vehicle on the highway is approximately 80 km / h, and that intercity highways lack auxiliary lighting at night, identifying dangerous highway behaviors and the license plates of the vehicles involved in this situation places extremely high demands on the perception equipment. These parameters translate to global exposure, a pixel size of 3μm or greater, a maximum aperture of F1.2 or greater, and an exposure time of less than 1 / 200s. Of course, the appropriate video acquisition device can be selected based on the specific situation.

[0078] In the embodiments of the present application, the electronic device 100 may be an independent server, or a server network or server cluster composed of servers. For example, the electronic device 100 described in the embodiments of the present application includes but is not limited to a computer, a network host, a single network server, a set of multiple network servers, or a cloud server composed of multiple servers. The cloud server is composed of a large number of computers or network servers based on cloud computing.

[0079] Those skilled in the art will understand that Figure 1 The application environment shown in the figure is only one application scenario of the present application solution and does not constitute a limitation on the application scenario of the present application solution. Other application environments may also include Figure 1 More or fewer electronic devices as shown in, e.g. Figure 1 Only one electronic device is shown. It can be understood that the accident vehicle license plate number detection system can also include one or more other servers, which are not limited here.

[0080] In addition, if Figure 1As shown, the accident vehicle license plate number detection system may further include a memory 200 for storing data.

[0081] It should be noted that Figure 1 The scenario diagram of the accident vehicle license plate number detection system shown is only an example. The accident vehicle license plate number detection system and scenario described in the embodiment of the present application are intended to more clearly illustrate the technical solution of the embodiment of the present application, and do not constitute a limitation on the technical solution provided by the embodiment of the present application. Ordinary technicians in this field can know that with the evolution of the accident vehicle license plate number detection system and the emergence of new business scenarios, the technical solution provided by the embodiment of the present application is also applicable to similar technical problems.

[0082] First, an embodiment of the present application provides a method for detecting the license plate number of an accident vehicle. The execution subject of the method is a detection device for the license plate number of the accident vehicle. The detection device for the license plate number of the accident vehicle is applied to an electronic device. The method for detecting the license plate number of the accident vehicle includes:

[0083] When an image of an accident vehicle containing a preset accident identifier is acquired by a target inspection vehicle, at least one image of the accident vehicle taken by the target inspection vehicle within a first preset time period before and after the image of the accident vehicle is acquired, to obtain at least two images of the accident vehicle;

[0084] Performing license plate character recognition on at least two accident vehicle images to obtain at least two first license plate character recognition results corresponding to the at least two accident vehicle images;

[0085] The license plate number of the accident vehicle is determined based on the appearance frequencies of the characters in at least two first license plate character recognition results.

[0086] See Figure 2 , Figure 2 This is a flow chart of an embodiment of the method for detecting the license plate number of an accident vehicle provided in the embodiment of the present application. Figure 2 As shown, the detection method of the accident vehicle license plate number includes:

[0087] S201. When an image of an accident vehicle containing a preset accident mark taken by a target inspection vehicle is obtained, at least one image of the accident vehicle taken by the target inspection vehicle within a first preset time period before and after the image of the accident vehicle is taken is retrieved to obtain at least two images of the accident vehicle.

[0088] Since the detection device for the accident vehicle license plate number is connected to multiple patrol vehicles, when the detection device for the accident vehicle license plate number obtains the accident vehicle images sent by multiple patrol vehicles simultaneously or successively, the accident vehicle images sent by multiple patrol vehicles can be processed one by one, simultaneously or in a predetermined order. This application does not limit this.

[0089] In the embodiment of the present application, the preset sign can be at least one of a warning triangle, illuminated hazard lights, and an ordinary pedestrian. A vehicle's warning triangle is a passive reflector made of plastic reflective material. When a driver stops for repairs or an accident occurs on the road, the warning triangle's reflective properties can be used to alert other vehicles to avoid a secondary accident. The hazard lights refer to the pair of taillights at the rear of the vehicle that flash when the hazard lights in the cab are activated. Generally, the hazard lights are two flashing yellow lights.

[0090] The first preset time period can be 1s, 2s, 3s, etc., and can be set according to the specific situation. Specifically, the accident vehicle video within the first preset time period before and after the target inspection vehicle takes the accident vehicle image is retrieved; the accident vehicle video is decoded to obtain at least two accident vehicle images. For example, the target inspection vehicle takes the accident vehicle image at 10:09:09, and the first preset time period is 2s, then the accident vehicle video from 10:09:07 to 10:09:11 is extracted. Of course, key frames in the accident vehicle video can also be extracted as at least one accident vehicle image, which can improve the picture effect of the accident vehicle image and improve the detection effect.

[0091] Furthermore, when an image of an accident vehicle containing a preset accident mark taken by a target inspection vehicle is obtained, the shooting position of the target inspection vehicle when taking the image of the accident vehicle is obtained, and it is determined whether the shooting position is within a preset area. When the shooting position is within the preset area, the shooting position is determined as the location where the accident occurred. For example, the preset area is a highway area. Specifically, the shooting time of the accident vehicle image is obtained, and the shooting position of the target inspection vehicle when taking the image of the accident vehicle is determined based on the shooting time of the accident vehicle image, and the shooting position is determined as the location where the accident occurred. Detection is only performed when the shooting position is within the preset area, which can reduce unnecessary detection and avoid increasing the system load. At the same time, the road conditions in the highway area are relatively simple, and the solution of this application has higher accuracy in this scenario.

[0092] Furthermore, when the shooting area is within a preset area, the shooting location of the target patrol vehicle when it captured the image of the accident vehicle is obtained, and a determination is made as to whether there are historical images of the accident vehicle sent from the shooting location by other patrol vehicles within a preset historical time period. The preset historical time period can be one hour, two hours, or the like, depending on the specific situation. If there are historical images of the accident vehicle sent from the shooting location by other patrol vehicles within the preset historical time period, the shooting heading angle when the target patrol vehicle captured the image of the accident vehicle and the historical heading angles when the other patrol vehicles captured the historical images of the accident vehicle are obtained. If the historical heading angle differs from the shooting heading angle, a determination is made as to whether the shooting location is within the preset area. If the shooting area is within the preset area, the shooting location is determined as the location of the accident. If the historical heading angle and the shooting heading angle are the same, the process ends. For two-way lanes, since images of the accident vehicle at the same location may have been captured by patrol vehicles driving in the same direction, determining the specific lane where the accident occurred based on the patrol vehicle's heading angle can prevent patrol vehicles driving in the same direction from reporting the same accident repeatedly, identify accidents at the same location but in different driving directions, improve the accuracy of accident vehicle detection, and reduce system load.

[0093] In an embodiment of the present application, when an image of an accident vehicle containing a preset accident identifier is acquired by a target inspection vehicle, at least one image of the accident vehicle taken by the target inspection vehicle within a first preset time period before and after the image of the accident vehicle is acquired is retrieved to obtain at least two images of the accident vehicle, before which:

[0094] (1) Obtain a first vehicle image taken by a target inspection vehicle.

[0095] In this embodiment of the present application, the target inspection vehicle is a vehicle connected to the accident vehicle license plate detection system. The target inspection vehicle can capture an image of the first vehicle on the road at a predetermined frequency and upload it to the accident vehicle license plate detection system. The predetermined frequency can be 1 Hz or 2 Hz, etc., depending on the specific situation. The accident vehicle license plate detection system can then acquire the first vehicle image captured by the target inspection vehicle.

[0096] (2) Detect whether there is a preset accident mark in the first vehicle image.

[0097] In an embodiment of the present application, a warning triangle detection network model is used to detect whether a warning triangle exists in the first vehicle image, a hazard light detection network model is used to detect whether hazard lights exist in the first vehicle image, and a pedestrian detection network model is used to detect whether pedestrians exist in the first vehicle image. When a warning triangle, hazard lights, or pedestrians exist in the first vehicle image, it is determined that a preset accident marker exists in the first vehicle image. Of course, it is also possible to determine that a preset accident marker exists in the first vehicle image when a warning triangle, hazard lights, or pedestrians exist in the first vehicle image. This can be determined based on the specific circumstances, and this application does not limit this.

[0098] Specifically, a large number of images of illuminated hazard lights are obtained as basic training data for hazard lights, and the YOLOv4 model is trained using this data to obtain a hazard light detection model. A large number of images of warning triangles are obtained as basic training data for warning triangles, and the YOLOv4 model is trained using this data to obtain a network model for detecting warning triangles. Images of ordinary pedestrians, dummies, on-duty personnel, and construction workers are obtained as basic training data for pedestrians, and the YOLOv4 model and the ResNet model are trained using this data to obtain a network model for detecting ordinary pedestrians. Of course, YOLOv4 can also be replaced with models such as YOLOv3, Faster R-CNN, and SSD.

[0099] YOLO-V4 uses the head of yolov3, selects CSPDarknet53 as the backbone network, SPP and modified-PAN as the neck, and adds an attention module to the network through various necks for extracting and combining features, as well as the NMS (non-maximum suppression) post-processing method, which can significantly improve the accuracy of target detection.

[0100] In low-light conditions, yellow and red lights may be confused, making it difficult to distinguish between the hazard lights and brake lights. Furthermore, the accuracy of the network model used to detect whether the hazard lights are activated in an image is limited. Therefore, after using the hazard light detection network model to detect the presence of hazard lights in the first vehicle image, if the hazard light detection network model determines that hazard lights are present in the first vehicle image, the hazard light area in the first vehicle image is obtained, the color channels of the hazard light area are obtained, and the color of the hazard light area is determined based on the color channels. If the color of the hazard light area is a preset color, the presence of hazard lights in the first vehicle image is determined. Specifically, the preset color is yellow. Every image has one or more color channels, and the default number of color channels in an image depends on its color mode. That is, the color mode of an image determines the number of color channels. For example, a CMYK image has four channels by default: cyan, magenta, yellow, and black. By default, bitmap, grayscale, duotone, and indexed color images have only one channel. RGB and Lab images have three channels, while CMYK images have four channels. You can choose the color channel that best suits your specific situation.

[0101] After using the double flash light detection network model to detect whether the first vehicle image has double flash lights, adding RGB channel values for yellow and red verification can further improve the accuracy of double flash light detection.

[0102] (3) If a preset accident mark exists on the first vehicle image, the first vehicle image is determined to be an accident vehicle image.

[0103] In a specific embodiment, if a preset accident marker is present on a first vehicle image, at least one second vehicle image captured by the target inspection vehicle within a second preset time period before and after the first vehicle image is captured is retrieved, and the preset accident marker is detected for each of the at least one second vehicle image. When the proportion of vehicle images with the preset accident marker detected in the at least one second vehicle image and the first vehicle image satisfies a preset proportion condition, the first vehicle image is determined to be an accident vehicle image. For example, if the preset marker is a hazard light, the preset proportion condition may be greater than the first preset proportion and less than the second preset proportion. For example, the first preset proportion is 30%, and the second preset proportion is 80%. The second preset time period may be 1s, 2s, 3s, etc., and may be set according to the specific situation.

[0104] Specifically, after retrieving a video of suspected double-flash flashes within a second preset time period before and after the vehicle image was captured, the double-flash lights are detected frame by frame in the suspected double-flash flashes video. If the number of frames in which double-flash lights are detected is greater than 30% and less than 80%, the event is considered a valid double-flash flash event, and the first vehicle image is determined to be an image of the accident vehicle.

[0105] S202. Perform license plate character recognition on at least two accident vehicle images to obtain at least two first license plate character recognition results corresponding to the at least two accident vehicle images.

[0106] In the embodiment of this application, license plate character recognition is performed on at least two accident vehicle images to obtain at least two first license plate character recognition results corresponding to the at least two accident vehicle images.

[0107] Specifically, use a license plate detection model to detect the license plate in the accident vehicle image to obtain multiple license plate candidate regions, and select the license plate candidate region with the largest IOU with the sample from the multiple license plate candidate regions as the accident vehicle license plate number region. The conventional non-maximum suppression algorithm uses IoU as the distance metric. IoU is what we call the intersection over union, which is the most commonly used metric in object detection and can reflect the detection effect of the predicted detection box and the true detection box. The license plate detection model can be a model obtained by training the YOLOv3 model directly using license plate number images. Perform character segmentation on the accident vehicle license plate number region to obtain multiple characters. Character segmentation generally uses the vertical projection method. Since the projection of characters in the vertical direction must obtain a local minimum value near the gap between or within characters, and this position should meet the character writing format, characters, size limitations, and some other conditions of the license plate. Using the vertical projection method has a good effect on character segmentation in automotive images in complex environments. After obtaining multiple characters, perform recognition on each of the multiple characters to obtain the first license plate character recognition result. Character recognition mainly includes template matching algorithms and artificial neural network algorithms. For the template matching algorithm, first binarize the segmented characters and scale their size to the size of the templates in the character database, and then match them with all the templates to select the best match as the result. There are two algorithms based on artificial neural networks: one is to first extract the features of the characters and then use the obtained features to train the neural network allocator; the other method is to directly input the image into the network, and the network automatically realizes feature extraction until the result is recognized. The first license plate character recognition result includes the positions of each character in the first license plate character recognition result.

[0108] S203. Determine the accident vehicle license plate number based on the occurrence frequency of characters in at least two first license plate character recognition results.

[0109] The accident vehicle license plate number generally includes Chinese characters and alphanumeric characters. Alphanumeric characters include either letters or numbers. Taking Chinese license plates as an example, the license plate number generally includes one Chinese character and 6 alphanumeric characters. For example, the accident vehicle license plate number can be E ABC123, Yue BBC123, etc. Of course, the content of the accident vehicle license plate number can be determined according to the traffic rules of different countries or regions, and this application does not limit this. The following only takes Chinese license plates as an example for illustration.

[0110] In an embodiment of the present application, determining the license plate number of the accident vehicle based on the frequency of occurrence of characters in at least two first license plate character recognition results can include: screening at least two first license plate character recognition results to obtain at least two second license plate character recognition results, wherein the number of alphanumeric characters in the second license plate character recognition results is greater than a first preset number; determining the license plate number of the accident vehicle based on the frequency of occurrence of characters in at least two second license plate character recognition results.

[0111] Among them, the first preset number can be set according to the specific situation, for example, the preset number is 3. When the number of alphanumeric characters in the first license plate character recognition result does not exceed 3, it indicates that the recognition effect of the license plate character recognition result is poor, and the license plate character recognition result is discarded, and only the second license plate character recognition result is retained. For example, at least two first license plate character recognition results are 4 first license plate character recognition results, namely Hubei ABC123, Hubei AB123, Hubei A23 and Guangdong BBC123. The number of alphanumeric characters in Hubei A23 does not exceed 3 and needs to be discarded. Therefore, at least two second license plate character recognition results are Hubei ABC123, Hubei AB123 and Guangdong BBC123.

[0112] In a specific embodiment, determining the license plate number of the accident vehicle based on the frequency of occurrence of characters in at least two second license plate character recognition results can include: judging whether there is a third license plate character recognition result in at least two second license plate character recognition results, wherein the number of alphanumeric characters in the third license plate character recognition result is less than a second preset number; if so, completing the multiple second license plate character recognition results based on the preset characters to obtain multiple fourth license plate character recognition results, wherein the number of alphanumeric characters in the fourth license plate character recognition result is the second preset number.

[0113] Among them, the second preset number and the preset character can be set according to the specific situation. For example, the second preset number is 6, and the preset character is *. For example, at least two second license plate character recognition results are Hubei ABC123, Hubei AB123, and Guangdong BBC123. Among them, the number of alphanumeric characters of Hubei AB123 is less than the second preset number, and it is determined to be the third license plate character recognition result. The third license plate character recognition result is completed to obtain *AB123. The number of alphanumeric characters of the remaining second license plate character recognition results other than the third license plate character recognition result is not less than the second preset number, and no completion is required. Therefore, the multiple fourth license plate character recognition results are Hubei ABC123, *AB123, and Guangdong BBC123.

[0114] In a specific embodiment, the plurality of second license plate character recognition results are supplemented based on the preset characters to obtain the plurality of fourth license plate character recognition results, which may include:

[0115] (1) Arranging and combining the alphanumeric characters in each fourth license plate character recognition result to obtain a plurality of fifth license plate character recognition results, wherein the relative positions of the alphanumeric characters in the fourth license plate character recognition results remain unchanged.

[0116] For example, the multiple fourth license plate character recognition results are EABC123, *AB123, and YBBC123. EABC123 and YBBC123 have only one possible combination, assuming the relative positions of the alphanumeric characters remain unchanged. *AB123 has six possible combinations, assuming the relative positions of the alphanumeric characters remain unchanged: EAB123*, EA12*3, EAB1*23, EAB*123, EA*B123, and E*AB123. Therefore, the multiple fifth license plate character recognition results are eight.

[0117] (2) Determining a weight coefficient for each fifth license plate character recognition result based on the number of preset characters in each fifth license plate character recognition result.

[0118] Specifically, the weight coefficient of each fifth license plate character recognition result is determined based on the number of preset characters and the second preset number in each fifth license plate character recognition result. The larger the number of preset characters, the smaller the weight coefficient. For example, if the number of preset characters is M and the second preset number is N, the weight coefficient is

[0119] (3) Calculating the frequency of occurrence of the alphanumeric characters in each alphanumeric position of the accident vehicle license plate number based on the plurality of fifth license plate character recognition results and the weight coefficient.

[0120] For example, the following are the character recognition results and weight coefficients for the fifth license plate: EABC123, weight coefficient 1; YBBC123, weight coefficient 1; EAB123*, weight coefficient 1 / 6; EA12*3, weight coefficient 1 / 6; EAB1*23, weight coefficient 1 / 6; EAB*123, weight coefficient 1 / 6; EA*B123, weight coefficient 1 / 6; and E*AB123, weight coefficient 1 / 6. The characters in the first alphanumeric position are A, B, and *. The frequencies of A, B, and * are 11 / 18, 6 / 18, and 6 / 18, respectively.

[0121] (4) Determine the alphanumeric character that appears most frequently in each alphanumeric position of the accident vehicle's license plate number as the alphanumeric character in each alphanumeric position of the accident vehicle's license plate number.

[0122] Since A has the highest frequency of occurrence, the first alphanumeric character position of the license plate number of the accident vehicle is A. Based on the same method, multiple alphanumeric characters at each alphanumeric character position of the license plate number of the accident vehicle can be obtained as ABC123.

[0123] Furthermore, the license plate number of the accident vehicle also includes Chinese characters. Determining the license plate number of the accident vehicle based on the frequency of occurrence of characters in at least two second license plate character recognition results includes: counting the frequency of occurrence of each Chinese character based on the second license plate character recognition results; obtaining the target Chinese character with the highest frequency of occurrence and a frequency of occurrence greater than the preset frequency from each Chinese character; and determining the target Chinese character as the Chinese character of the license plate of the accident vehicle. The preset frequency can be set according to specific circumstances. For example, the preset frequency is 50%. For example, if at least two second license plate character recognition results are E ABC123, E AB123, and Yue BBC123 respectively, the frequencies of occurrence of the Chinese characters "E" and "Yue" are 2 / 3 and 1 / 3 respectively, then the target Chinese character is determined as "E".

[0124] Obtain the target Chinese character with the highest frequency of occurrence and a frequency of occurrence greater than the preset frequency from each Chinese character, and determine the target Chinese character as the first character of the license plate number of the accident vehicle, obtaining the license plate number of the accident vehicle as E ABC123.

[0125] Furthermore, display the accident vehicle image, the license plate number of the accident vehicle, the accident location, etc. on the real-scene display platform.

[0126] To better implement the method for detecting the license plate number of an accident vehicle in the embodiments of the present application, based on the method for detecting the license plate number of an accident vehicle, an apparatus for detecting the license plate number of an accident vehicle is further provided in the embodiments of the present application, as Figure 3 shown, Figure 3 is a schematic structural diagram of an embodiment of an apparatus for detecting the license plate number of an accident vehicle provided in the embodiments of the present application. The apparatus for detecting the license plate number of an accident vehicle includes:

[0127] Retrieving unit 301, configured to retrieve at least one accident vehicle image within a first preset time period before and after capturing the accident vehicle image when obtaining an accident vehicle image containing a preset accident identifier captured by a target patrol vehicle, so as to obtain at least two accident vehicle images;

[0128] License plate character recognition unit 302, configured to perform license plate character recognition on at least two accident vehicle images to obtain at least two first license plate character recognition results corresponding to the at least two accident vehicle images;

[0129] Determining unit 303, configured to determine the license plate number of the accident vehicle based on the frequency of occurrence of characters in at least two first license plate character recognition results.

[0130] The license plate number of the accident vehicle includes a plurality of alphanumeric characters, and the alphanumeric characters include any one of letters and numbers;

[0131] The determining unit 303 is further configured to screen the at least two first license plate character recognition results to obtain at least two second license plate character recognition results, wherein the number of alphanumeric characters in the second license plate character recognition results is greater than a first preset number;

[0132] The license plate number of the accident vehicle is determined based on the appearance frequencies of the characters in the at least two second license plate character recognition results.

[0133] The determining unit 303 is further configured to determine whether there is a third license plate character recognition result in the at least two second license plate character recognition results, wherein the number of alphanumeric characters in the third license plate character recognition result is less than a second preset number;

[0134] If so, the plurality of second license plate character recognition results are supplemented based on the preset characters to obtain a plurality of fourth license plate character recognition results, wherein the number of alphanumeric characters in the fourth license plate character recognition results is the second preset number.

[0135] The determining unit 303 is further configured to arrange and combine the alphanumeric characters in each of the fourth license plate character recognition results to obtain a plurality of fifth license plate character recognition results, wherein the relative positions of the alphanumeric characters in the fourth license plate character recognition results remain unchanged;

[0136] determining a weight coefficient for each fifth license plate character recognition result based on the number of preset characters in each fifth license plate character recognition result;

[0137] calculating the frequency of occurrence of the alphanumeric character at each alphanumeric position of the accident vehicle's license plate number based on the plurality of fifth license plate character recognition results and the weight coefficient;

[0138] The alphanumeric character that appears most frequently in each alphanumeric position of the accident vehicle license plate number is determined as the alphanumeric character in each alphanumeric position of the accident vehicle license plate number.

[0139] Among them, the license plate number of the accident vehicle also includes Chinese characters.

[0140] The determining unit 303 is further configured to count the occurrence frequencies of each Chinese character based on the second license plate character recognition result;

[0141] Obtain the target Chinese character with the highest appearance frequency and a frequency greater than a preset frequency from each Chinese character;

[0142] The target Chinese characters are determined to be the Chinese characters of the license plate of the accident vehicle.

[0143] The preset accident signs include at least one of a warning triangle, a lit double flash light, and an ordinary pedestrian;

[0144] The retrieving unit 301 is further configured to obtain a first vehicle image captured by the target inspection vehicle;

[0145] Detecting whether a preset accident mark exists in the first vehicle image;

[0146] If a preset accident mark exists on the first vehicle image, the first vehicle image is determined to be an accident vehicle image.

[0147] The retrieving unit 301 is further configured to retrieve at least one second vehicle image captured by the target inspection vehicle within a second preset time period before or after the first vehicle image is captured, if a preset accident mark is present on the first vehicle image;

[0148] performing preset accident sign detection on at least one second vehicle image;

[0149] When the proportion of vehicle images in which a preset accident mark is detected in at least one second vehicle image and the first vehicle image meets a preset proportion condition, the first vehicle image is determined to be an accident vehicle image.

[0150] The embodiment of the present application also provides an electronic device that integrates any of the detection devices for the license plate number of an accident vehicle provided in the embodiment of the present application. Figure 4 , which shows a schematic diagram of the structure of the electronic device involved in the embodiment of the present application, specifically:

[0151] The electronic device may include components such as a processor 401 with one or more processing cores, a memory 402 with one or more computer-readable storage media, a power supply 403, and an input unit 404. Those skilled in the art will appreciate that the electronic device structure shown in the figure does not limit the electronic device and may include more or fewer components than shown, or combine certain components, or arrange the components differently.

[0152] Processor 401 is the control center of the electronic device, connecting the various parts of the entire electronic device using various interfaces and lines. By running or executing software programs and / or modules stored in memory 402 and accessing data stored in memory 402, it performs various functions of the electronic device and processes data, thereby monitoring the electronic device as a whole. Optionally, processor 401 may include one or more processing cores; preferably, processor 401 may integrate an application processor and a modem processor, wherein the application processor primarily processes the operating system, user interface, and application programs, while the modem processor primarily handles wireless communications. It is understood that the modem processor may not be integrated into processor 401.

[0153] The memory 402 can be used to store software programs and modules. The processor 401 executes various functional applications and data processing by running the software programs and modules stored in the memory 402. The memory 402 may mainly include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application required for at least one function (such as a sound playback function, an image playback function, etc.), etc.; the data storage area may store data created according to the use of the electronic device, etc. In addition, the memory 402 may include a high-speed random access memory, and may also include a non-volatile memory, such as at least one disk storage device, a flash memory device, or other volatile solid-state storage device. Accordingly, the memory 402 may also include a memory controller to provide the processor 401 with access to the memory 402.

[0154] The electronic device also includes a power supply 403 for supplying power to various components. Preferably, the power supply 403 can be logically connected to the processor 401 via a power management system, thereby enabling the power management system to manage charging, discharging, and power consumption. The power supply 403 can also include one or more DC or AC power supplies, a recharging system, a power failure detection circuit, a power converter or inverter, a power status indicator, and other arbitrary components.

[0155] The electronic device may further include an input unit 404, which may be configured to receive input digital or character information and generate keyboard, mouse, joystick, optical or trackball signal inputs related to user settings and function control.

[0156] Although not shown, the electronic device may further include a display unit, etc., which will not be described in detail here. Specifically, in this embodiment, the processor 401 in the electronic device will load the executable files corresponding to the processes of one or more application programs into the memory 402 according to the following instructions, and the processor 401 will run the application programs stored in the memory 402 to implement various functions as follows:

[0157] When an image of an accident vehicle containing a preset accident identifier is acquired by a target inspection vehicle, at least one image of the accident vehicle taken by the target inspection vehicle within a first preset time period before and after the image of the accident vehicle is acquired, to obtain at least two images of the accident vehicle;

[0158] Performing license plate character recognition on at least two accident vehicle images to obtain at least two first license plate character recognition results corresponding to the at least two accident vehicle images;

[0159] The license plate number of the accident vehicle is determined based on the appearance frequencies of the characters in at least two first license plate character recognition results.

[0160] Those skilled in the art will appreciate that all or part of the steps in the various methods of the above embodiments may be accomplished by instructions, or by controlling related hardware through instructions. The instructions may be stored in a computer-readable storage medium and loaded and executed by a processor.

[0161] To this end, embodiments of the present application provide a computer-readable storage medium, which may include a read-only memory (ROM), random access memory (RAM), a disk, or an optical disk. A computer program is stored thereon, and the computer program is loaded by a processor to execute the steps of any of the accident vehicle license plate detection methods provided in embodiments of the present application. For example, the computer program loaded by the processor may execute the following steps:

[0162] When an image of an accident vehicle containing a preset accident identifier is acquired by a target inspection vehicle, at least one image of the accident vehicle taken by the target inspection vehicle within a first preset time period before and after the image of the accident vehicle is acquired, to obtain at least two images of the accident vehicle;

[0163] Performing license plate character recognition on at least two accident vehicle images to obtain at least two first license plate character recognition results corresponding to the at least two accident vehicle images;

[0164] The license plate number of the accident vehicle is determined based on the appearance frequencies of the characters in at least two first license plate character recognition results.

[0165] In the above embodiments, the description of each embodiment has its own focus. For parts that are not described in detail in a certain embodiment, please refer to the detailed description of other embodiments above and will not be repeated here.

[0166] In specific implementation, the above units or structures can be implemented as independent entities, or can be arbitrarily combined to implement as the same or several entities. The specific implementation of the above units or structures can refer to the previous method embodiments and will not be repeated here.

[0167] The specific implementation of the above operations can be found in the previous embodiments and will not be repeated here.

[0168] The above is a detailed introduction to the method, device, electronic device and storage medium for detecting the license plate number of an accident vehicle provided in the embodiments of the present application. Specific examples are used in this article to illustrate the principles and implementation methods of the present application. The description of the above embodiments is only used to help understand the method of the present application and its core idea; at the same time, for technical personnel in this field, based on the ideas of the present application, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as a limitation on the present application.

Claims

1. A method for detecting the license plate number of an accident vehicle, characterized in that: The detection method of the license plate number of the accident vehicle includes: When an image of an accident vehicle containing a preset accident identifier is acquired and taken by a target inspection vehicle, at least one image of the accident vehicle taken by the target inspection vehicle within a first preset time period before and after the image of the accident vehicle is acquired, to obtain at least two images of the accident vehicle; Performing license plate character recognition on the at least two accident vehicle images to obtain at least two first license plate character recognition results corresponding to the at least two accident vehicle images; The license plate number of the accident vehicle is determined based on the frequency of occurrence of characters in the at least two first license plate character recognition results; the license plate number of the accident vehicle includes a plurality of alphanumeric characters, and the alphanumeric characters include any one of letters and numbers, including: Screening the at least two first license plate character recognition results to obtain at least two second license plate character recognition results, wherein the number of alphanumeric characters in the second license plate character recognition results is greater than a first preset number; determining whether a third license plate character recognition result exists in the at least two second license plate character recognition results, wherein the number of alphanumeric characters in the third license plate character recognition result is less than a second preset number; If so, completing the plurality of second license plate character recognition results based on the preset characters to obtain a plurality of fourth license plate character recognition results, wherein the number of alphanumeric characters in the fourth license plate character recognition results is the second preset number; Arrange and combine the alphanumeric characters in each of the fourth license plate character recognition results to obtain a plurality of fifth license plate character recognition results, wherein the relative positions of the alphanumeric characters in the fourth license plate character recognition results remain unchanged; determining a weight coefficient of each of the fifth license plate character recognition results based on the number of the preset characters in each of the fifth license plate character recognition results; calculating the frequency of occurrence of the alphanumeric characters in each alphanumeric position of the accident vehicle license plate number based on the plurality of fifth license plate character recognition results and the weight coefficient; The alphanumeric character that appears most frequently at each alphanumeric position of the accident vehicle license plate number is determined as the alphanumeric character at each alphanumeric position of the accident vehicle license plate number.

2. The method for detecting the license plate number of an accident vehicle according to claim 1, wherein: The license plate number of the accident vehicle also includes Chinese characters, The determining the license plate number of the accident vehicle based on the frequencies of occurrence of characters in the at least two second license plate character recognition results comprises: Counting the frequency of occurrence of each Chinese character based on the second license plate character recognition result; Obtaining a target Chinese character with the highest appearance frequency and an appearance frequency greater than a preset frequency from each of the Chinese characters; The target Chinese character is determined to be the Chinese character of the license plate of the accident vehicle.

3. The method for detecting the license plate number of an accident vehicle according to any one of claims 1 to 2, characterized in that: The preset accident sign includes at least one of a warning triangle, a lit double flash light, and an ordinary pedestrian; When an image of an accident vehicle containing a preset accident identifier is acquired and taken by a target inspection vehicle, at least one image of the accident vehicle taken by the target inspection vehicle within a first preset time period before and after the image of the accident vehicle is retrieved to obtain at least two images of the accident vehicle, before which the method includes: Acquire a first vehicle image captured by a target inspection vehicle; Detecting whether the preset accident mark exists in the first vehicle image; If the preset accident mark exists on the first vehicle image, the first vehicle image is determined as the accident vehicle image.

4. The method for detecting the license plate number of an accident vehicle according to claim 3, wherein: If the preset accident mark exists on the first vehicle image, determining the first vehicle image as the accident vehicle image includes: If the preset accident mark exists on the first vehicle image, retrieving at least one second vehicle image taken by the target inspection vehicle within a second preset time period before and after the first vehicle image is taken; performing preset accident sign detection on each of the at least one second vehicle images; When the proportion of vehicle images in which a preset accident mark is detected in the at least one second vehicle image and the first vehicle image meets a preset proportion condition, the first vehicle image is determined as the accident vehicle image.

5. A device for detecting the license plate number of an accident vehicle, characterized in that: The detection device for the license plate number of the accident vehicle includes: a retrieval unit configured to retrieve, when an image of an accident vehicle containing a preset accident identifier is acquired and taken by a target inspection vehicle, at least one image of the accident vehicle taken by the target inspection vehicle within a first preset time period before and after the image of the accident vehicle is taken, thereby obtaining at least two images of the accident vehicle; a license plate character recognition unit, configured to perform license plate character recognition on the at least two accident vehicle images to obtain at least two first license plate character recognition results corresponding to the at least two accident vehicle images; A determination unit is configured to determine the license plate number of the accident vehicle based on the frequency of occurrence of characters in the at least two first license plate character recognition results; the license plate number of the accident vehicle includes a plurality of alphanumeric characters, and the alphanumeric characters include any one of letters and numbers, including: Screening the at least two first license plate character recognition results to obtain at least two second license plate character recognition results, wherein the number of alphanumeric characters in the second license plate character recognition results is greater than a first preset number; determining whether a third license plate character recognition result exists in the at least two second license plate character recognition results, wherein the number of alphanumeric characters in the third license plate character recognition result is less than a second preset number; If so, completing the plurality of second license plate character recognition results based on the preset characters to obtain a plurality of fourth license plate character recognition results, wherein the number of alphanumeric characters in the fourth license plate character recognition results is the second preset number; Arrange and combine the alphanumeric characters in each of the fourth license plate character recognition results to obtain a plurality of fifth license plate character recognition results, wherein the relative positions of the alphanumeric characters in the fourth license plate character recognition results remain unchanged; determining a weight coefficient of each of the fifth license plate character recognition results based on the number of the preset characters in each of the fifth license plate character recognition results; calculating the frequency of occurrence of the alphanumeric characters in each alphanumeric position of the accident vehicle license plate number based on the plurality of fifth license plate character recognition results and the weight coefficient; The alphanumeric character that appears most frequently at each alphanumeric position of the accident vehicle license plate number is determined as the alphanumeric character at each alphanumeric position of the accident vehicle license plate number.

6. An electronic device, characterized in that: The electronic device comprises: one or more processors; Memory; and One or more applications, wherein the one or more applications are stored in the memory and configured to be executed by the processor to implement the method for detecting the license plate number of an accident vehicle according to any one of claims 1 to 5.

7. A computer-readable storage medium, characterized in that A computer program is stored thereon, and the computer program is loaded by a processor to execute the steps in the method for detecting the license plate number of an accident vehicle according to any one of claims 1 to 5.

Images