Motor vehicle passing warning method, device, electronic device and computer equipment
By combining video and radar data processing, identifying and determining whether the motor vehicle is in the lane and generating warning information, the problem of low vehicle information recognition accuracy is solved and traffic safety is improved.
Patent Information
- Application Number
- CN202110756947.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2021-07-05
- Publication Date
- 2024-12-06
- Estimated Expiration
- 2041-07-05
AI Technical Summary
Among the existing vehicle detection methods, the accuracy of motor vehicle information identification is low, especially when the vehicle is not accurate enough to detect safety threats from pedestrians or non-motor vehicles in the blind spot of the driver's perspective during driving.
By obtaining video and radar data in the target traffic scene, combining the motor vehicle attributes in the video data and driving information in the radar data, using target detection algorithms, license plate recognition algorithms and video tracking algorithms to determine whether the motor vehicle meets the vehicle warning conditions. And generate warning messages.
It improves the accuracy of vehicle information identification of motor vehicles in the lane, enhances timely warnings for motor vehicles that may have traffic safety hazards, and reduces the probability of traffic accidents.
Smart Images

Figure CN113688662B_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of image recognition, and in particular to a method, device, electronic device and computer equipment for warning a passing vehicle. Background Art
[0002] In order to meet the needs of urban infrastructure, logistics and other travel, the number of various types of motor vehicles on urban roads has gradually increased, and traffic safety hazards have also increased. For example, due to the driver's blind spot, the motor vehicle may pose a threat to the safety of pedestrians and non-motor vehicles in the driver's blind spot during driving. Therefore, it is necessary to detect the motor vehicle information in the traffic scene to determine the motor vehicles with traffic safety hazards and take corresponding measures in time. In the current vehicle detection method, video tracking technology is often used to monitor the target vehicle in the acquired video data. This method of monitoring the target vehicle based only on video images is not very stable, and the vehicle information of the motor vehicle, such as speed and distance, is acquired. The accuracy is low.
[0003] With respect to the problem of low accuracy in vehicle information recognition existing in the related art, no effective solution has been proposed yet. Summary of the invention
[0004] In this embodiment, a motor vehicle passing warning method, device, electronic device and computer equipment are provided to solve the problem of low accuracy of vehicle information recognition and lane judgment existing in the related art.
[0005] In a first aspect, a method for warning a passing vehicle is provided in this embodiment, comprising the following steps:
[0006] Acquire a target video in a target traffic scene, wherein the target video includes at least one motor vehicle and a target lane;
[0007] Acquire vehicle attributes of a target motor vehicle in the target video, wherein the target motor vehicle includes a motor vehicle among the at least one motor vehicle traveling in the target lane;
[0008] Acquiring radar data corresponding to the target motor vehicle, wherein the radar data includes driving information of the target motor vehicle;
[0009] Determine whether the target motor vehicle meets vehicle warning conditions based on the vehicle attributes and the driving information.
[0010] In some embodiments, obtaining the vehicle attributes of the target motor vehicle in the target video includes:
[0011] Based on the target detection algorithm, it is determined whether the motor vehicle in the target video is traveling in the target lane. If so, the motor vehicle is set as a target motor vehicle, and the vehicle attributes of the target motor vehicle are obtained.
[0012] In some embodiments, the target lane is determined based on a lane line preconfigured in the target video, and the step of determining whether the motor vehicle in the target video is traveling in the target lane based on a target detection algorithm includes:
[0013] Inputting the target video into a preset target detection model to obtain key point information of the motor vehicle;
[0014] Calculate the center point of the front projection of the motor vehicle according to the key point information;
[0015] Whether the motor vehicle is traveling in the target lane is determined based on the positional relationship between the center point of the vehicle head projection and the lane line.
[0016] In some embodiments, the acquiring radar data corresponding to the target motor vehicle includes:
[0017] Acquiring radar data under the target traffic scene;
[0018] The radar data in the target traffic scene is associated with the corresponding target motor vehicle in the target video through an image calibration algorithm to obtain the radar data corresponding to the target motor vehicle.
[0019] In some embodiments, associating the radar data to a corresponding target motor vehicle in the target video by an image calibration algorithm includes:
[0020] The image data in the target video is mapped into a bird's-eye view through an image calibration algorithm, and the radar data is associated with the corresponding target motor vehicle according to the bird's-eye view.
[0021] In some embodiments, the vehicle attributes of the target motor vehicle include license plate attributes and vehicle model attributes, and obtaining the vehicle attributes of the target motor vehicle in the target video includes:
[0022] Acquire a target motor vehicle image of the target motor vehicle, and process the target motor vehicle image using a license plate detection method to obtain a license plate image of the target motor vehicle;
[0023] Processing the license plate image using a license plate recognition algorithm to obtain the license plate attributes of the target motor vehicle;
[0024] The target motor vehicle image is processed using a vehicle type recognition algorithm to obtain the vehicle type attributes of the target motor vehicle.
[0025] In some embodiments, when it is determined that the target motor vehicle meets the vehicle warning condition according to the vehicle attributes and the driving information, the method further includes:
[0026] Determining identification information of the target motor vehicle based on a video tracking algorithm;
[0027] In the target traffic scenario, warning information is generated for each target motor vehicle according to the identification information.
[0028] In a second aspect, a motor vehicle passing warning device is provided in this embodiment, including: an acquisition module, a judgment module, a radar-visual correlation module and a warning module, wherein:
[0029] The acquisition module is used to acquire a target video in a target traffic scene, wherein the target video contains at least one motor vehicle and a target lane;
[0030] The judgment module is used to obtain vehicle attributes of a target motor vehicle in the target video, wherein the target motor vehicle includes a motor vehicle among the at least one motor vehicle traveling in the target lane;
[0031] The radar-visual correlation module is used to obtain radar data corresponding to the target motor vehicle, wherein the radar data includes driving information of the target motor vehicle;
[0032] The warning module is used to determine whether the target motor vehicle meets the vehicle warning condition according to the vehicle attributes and the driving information.
[0033] According to a third aspect, an electronic device is provided in this embodiment, comprising a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the steps of the method according to the first aspect are implemented.
[0034] In a fourth aspect, a computer device is provided in this embodiment, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the method of the first aspect when executing the computer program.
[0035] The above-mentioned motor vehicle passing warning method, device, electronic device and computer equipment obtain a target video in a target traffic scene, wherein the target video contains at least one motor vehicle and a target lane, obtain the vehicle attributes of the target motor vehicle in the target video, wherein the target motor vehicle includes at least one motor vehicle traveling in the target lane, obtain radar data corresponding to the target motor vehicle, wherein the radar data contains the driving information of the target motor vehicle, and determine whether the target motor vehicle meets the vehicle warning conditions according to the vehicle attributes and driving information. Through the correlation processing of radar data and video data, it can improve the accuracy of the vehicle information of the motor vehicle in the target lane obtained, thereby improving the accuracy of the detection of the vehicle information to be warned, thereby realizing the timely warning of the motor vehicle information in the target lane that meets the warning conditions.
[0036] Details of one or more embodiments of the present application are set forth in the following drawings and description to make other features, objects, and advantages of the present application more readily apparent. BRIEF DESCRIPTION OF THE DRAWINGS
[0037] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:
[0038] Figure 1 This is an application scenario diagram of a motor vehicle passing warning method according to an embodiment of the present application;
[0039] Figure 2 is a flow chart of a method for warning a passing vehicle of a motor vehicle according to an embodiment of the present application;
[0040] Figure 3 is a flow chart of another motor vehicle passing warning method according to an embodiment of the present application;
[0041] Figure 4 is a schematic structural diagram of a motor vehicle passing warning device according to an embodiment of the present application;
[0042] Figure 5 is a schematic structural diagram of an electronic device according to an embodiment of the present application;
[0043] Figure 6 It is a schematic diagram of the structure of a computer device according to an embodiment of the present application. DETAILED DESCRIPTION
[0044] In order to make the purpose, technical solutions and advantages of the present application clearer, the present application is described and illustrated below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not intended to limit the present application. Based on the embodiments provided in the present application, all other embodiments obtained by ordinary technicians in the field without making creative work are within the scope of protection of the present application.
[0045] Obviously, the drawings described below are only some examples or embodiments of the present application. For ordinary technicians in this field, the present application can also be applied to other similar scenarios based on these drawings without creative work. In addition, it can also be understood that although the efforts made in this development process may be complicated and lengthy, for ordinary technicians in this field related to the content disclosed in this application, some changes in design, manufacturing or production based on the technical content disclosed in this application are just conventional technical means, and should not be understood as insufficient content disclosed in this application.
[0046] Reference to "embodiments" in this application means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase in various locations in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those of ordinary skill in the art that the embodiments described in this application may be combined with other embodiments without conflict.
[0047] Unless otherwise defined, the technical terms or scientific terms involved in this application should be understood by people with ordinary skills in the technical field to which this application belongs. The words "one", "a", "a", "the" and the like involved in this application do not indicate a quantitative limitation, and may represent the singular or plural. The terms "include", "comprise", "have" and any of their variations involved in this application are intended to cover non-exclusive inclusions; for example, a process, method, system, product or device that includes a series of steps or modules (units) is not limited to the listed steps or units, but may also include steps or units that are not listed, or may also include other steps or units inherent to these processes, methods, products or devices. The words "connect", "connected", "coupled" and the like involved in this application are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. The "multiple" involved in this application refers to two or more. "And / or" describes the association relationship of associated objects, indicating that there may be three relationships, for example, "A and / or B" can represent: A exists alone, A and B exist at the same time, and B exists alone. The character " / " generally indicates that the objects before and after are in an "or" relationship. The terms "first", "second", "third", etc. involved in this application are only used to distinguish similar objects and do not represent a specific ordering of the objects.
[0048] Figure 1 FIG. 1 is an application scenario diagram of a method for warning a vehicle passing by in one embodiment. Figure 1 As shown, it includes a monitoring device 101, a server 102, a radar 103 and a display screen 104. Among them, the monitoring device 101 is used to shoot the target traffic scene to obtain a target video of the vehicle driving in the target traffic scene, and the target video contains at least one motor vehicle and a target lane. The server 102 is used to receive the target video obtained by the monitoring device 101, and obtain the vehicle attributes of the target motor vehicle driving in the target lane of the target traffic scene in the target video. In addition, the radar 103 is used to obtain radar data in the target traffic scene, and send the radar data to the server 102, and the server 102 obtains the radar data corresponding to the target vehicle. Finally, the server 102 determines whether the target motor vehicle meets the vehicle warning condition based on the vehicle attributes and the driving information in the radar data. If so, a warning message is generated for the target motor vehicle, and the warning message is displayed on the display screen 104 to prompt other vehicles and pedestrians in the target traffic scene to pay attention to avoid. Among them, the monitoring device 101 and the server 102, the server 102 and the radar 103, and the server 102 and the display screen 104 can all be connected through a network.
[0049] In this embodiment, a method for warning a passing vehicle is provided. Figure 2: is a flow chart of the motor vehicle passing warning method of this embodiment, such as Figure 2 As shown, the process includes the following steps:
[0050] Step S210, obtaining a target video in a target traffic scene, wherein the target video includes at least one motor vehicle and a target lane.
[0051] The target traffic scene may specifically be a scene at a traffic intersection. The target video may be video data obtained by a monitoring device installed at the intersection shooting the driving conditions of vehicles at the intersection.
[0052] Step S220, obtaining vehicle attributes of a target motor vehicle in a target video, where the target motor vehicle includes a motor vehicle among at least one motor vehicle traveling in a target lane.
[0053] Specifically, the vehicle attributes may include, but are not limited to, information such as the model and license plate of the target motor vehicle. Among them, the model information of the target motor vehicle is mainly the vehicle type, which can be any one of a multi-purpose vehicle MPV (Multi-Purpose Vehicles), a sports utility vehicle SUV (Sport Utility Vehicle), a taxi, a small truck, a medium truck, a large truck, a small passenger car, a medium bus, a large bus, a bus, a sedan, a pickup truck, and other special vehicles. In addition, the license plate information of the target motor vehicle specifically includes the license plate type, the license plate color, and the license plate number. Specifically, the vehicle attributes can be obtained by performing image processing on the vehicle image of the target motor vehicle. Further, the anchor-base method or the anchor-free method can be used to segment the license plate area of the vehicle image of the target motor vehicle, and the segmented license plate image can be detected using the traditional support vector machine SVM (Support Vector Machine) or the deep learning method to obtain the license plate number, the license plate type, the license plate color, and other information. The vehicle image of the target motor vehicle can also be processed using a convolutional neural network (CNN) network classification method to obtain the vehicle model information of the target motor vehicle. The vehicle image of the target motor vehicle can be a frame of image data in the target video, or a license plate image obtained by real-time shooting of the target motor vehicle passing through a specific location in the target traffic scene.
[0054] In addition, the target video may contain multiple lanes and multiple motor vehicles. Before obtaining the vehicle attributes of the target motor vehicle in the target lane, it is necessary to perform target detection on the motor vehicle appearing in the target video by using the anchor-base or anchor-free method to determine the motor vehicle traveling in the target lane, and use the motor vehicle as the target motor vehicle. Among them, the target lane can be a lane that needs to be monitored for traffic safety in actual situations. For example, a truck traveling in a right-turn lane may not be able to take into account nearby pedestrians or other vehicles when turning right due to the driver's blind spot, which poses a threat to traffic safety. Therefore, the right-turn lane in the target video can be set as the target lane. Specifically, after obtaining the target video under the target traffic scene, the lane line in the target video can be configured in advance to determine the target lane in the target video. Among them, it can be configured manually or in real time using any lane line detection algorithm. Further, it can be determined whether the motor vehicle is traveling in the target lane based on the positional relationship between the motor vehicle in the target video and the target lane. For example, after obtaining a number of key point information of the motor vehicle, it is determined whether the motor vehicle is a target motor vehicle traveling in a target lane based on the positional relationship between the key point information and the configured lane lines.
[0055] In addition, the target video can be processed using circulant structure kernels (CSK), kernel correlation filters (KCF), or a series of deep learning methods such as CenterTrack and SiamRPN to achieve video tracking of the motor vehicle therein, so that the identification information of the same motor vehicle in the target video is consistent between two adjacent frames. Specifically, CenterNet can be used to detect the center point of the vehicle, and the displacement characteristics of the excited vehicle can be obtained using the center point of the vehicle between different frames to achieve tracking of the motor vehicle. By tracking the motor vehicle in the target video, the same motor vehicle in the actual scene can be bound to the same identification information from appearance to disappearance in the target video, thereby achieving the uniqueness of the motor vehicle in the target video and avoiding subsequent repeated processing of the motor vehicle.
[0056] Step S230, obtaining radar data corresponding to the target motor vehicle, wherein the radar data includes driving information of the target motor vehicle.
[0057] Among them, the radar data corresponding to the target motor vehicle can be obtained by a radar device installed in the target traffic scene, which emits electromagnetic waves to the motor vehicle in the target traffic scene and receives its echo. The radar device can obtain information such as the distance, direction, and speed of the motor vehicle in the target traffic scene relative to the radar device. The above information can be used as the driving information of the motor vehicle in the target traffic scene. In addition, in order to obtain the driving information of the target motor vehicle from the radar data of the radar device, it is also necessary to associate the radar data and video data in the target traffic scene to associate the radar data with the motor vehicle in the target video respectively. Specifically, each frame of image data in the target video can be mapped into a bird's-eye view using traditional affine transformation or deep learning methods, and then the radar data can be associated with the corresponding motor vehicle in the bird's-eye view. In order to obtain the radar data corresponding to the target motor vehicle, the driving information such as the speed and distance of the target motor vehicle can be obtained.
[0058] Since the radar setting is relatively stable in long-distance tracking and target detection, by correlating the radar data with the target video, the driving information of the target motor vehicle can be obtained from the radar data. Compared with using image processing methods to obtain the driving information of the target motor vehicle, it has higher accuracy and stability.
[0059] Step S240, determining whether the target motor vehicle meets the vehicle warning conditions based on the vehicle attributes and driving information.
[0060] Among them, the vehicle warning condition is used to limit which motor vehicle is warned, and can specifically include limiting the vehicle model, speed, distance between vehicles, license plate type and other contents. Among them, the vehicle model and license plate type can be limited to one or more, and the speed and distance between vehicles can be limited to a certain range. The vehicle warning condition can be pre-configured according to the actual traffic scene requirements and traffic rules, and the vehicle warning condition is not specifically limited in this embodiment. When the vehicle attributes and driving information of the target motor vehicle meet the vehicle warning conditions, a warning message can be generated for the target motor vehicle. The warning message can be specifically determined by the requirements of the actual traffic scene, and is used to warn other motor vehicles, non-motor vehicles and pedestrians in the target traffic scene to avoid the target motor vehicle and avoid traffic accidents. Specifically, the warning message can be information such as the distance between vehicles, speed, vehicle model, license plate type and license plate number of the target motor vehicle. In addition, after the warning message is generated, a display screen set in the target traffic scene can be used for a picture warning, an audio output device can be used for a voice warning, or a combination of pictures and voice can be used for a warning, which is not specifically limited in this embodiment.
[0061] For example, a motor vehicle A is detected in the target video, and the key point information of the motor vehicle A is obtained by the centerNet method. After the key point information is processed to obtain the center point of the front projection of the motor vehicle A, the positional relationship between the center point of the front projection and the pre-configured lane line is compared. After determining that the center point of the front projection is between two pre-configured lane lines, the motor vehicle A is set as the target motor vehicle, and the driving information of the target motor vehicle A is obtained through the radar visual association. In addition, after the target motor vehicle A passes through the detection line set somewhere in the target lane, the target motor vehicle A is captured in real time to obtain the vehicle image of the target motor vehicle. The vehicle attributes of the target motor vehicle are obtained by performing license plate and vehicle model detection on the vehicle image. When the distance between the target motor vehicle is less than the preset warning distance, the speed is greater than the preset warning speed, the vehicle model is a designated large truck model, and the license plate type is a yellow plate or a double-layer yellow plate, the distance, speed, vehicle model, license plate type and license plate number of the target motor vehicle A are output to the warning device in the target traffic scene to remind surrounding vehicles and pedestrians to pay attention to safety.
[0062] In the above steps S210 to S240, a target video under a target traffic scene is obtained, wherein the target video contains at least one motor vehicle and a target lane, the vehicle attributes of the target motor vehicle in the target video are obtained, and the target motor vehicle includes at least one motor vehicle traveling in the target lane, and radar data corresponding to the target motor vehicle is obtained, wherein the radar data contains the driving information of the target motor vehicle, and whether the target motor vehicle meets the vehicle warning conditions is determined according to the vehicle attributes and driving information. Through the correlation processing of radar data and video data, the accuracy of the vehicle information of the motor vehicle in the target lane obtained can be improved, thereby improving the accuracy of the detection of the vehicle information to be warned, thereby realizing the timely warning of the motor vehicle information in the target lane that meets the warning conditions.
[0063] Further, based on the above step S220, the vehicle attributes of the target motor vehicle in the target video are obtained, which specifically includes the following steps:
[0064] Step S221, based on the target detection algorithm, determine whether the motor vehicle in the target video is traveling in the target lane. If so, set the motor vehicle as the target motor vehicle and obtain the vehicle attributes of the target motor vehicle.
[0065] You can use anchor-base methods, such as the unified real-time object detection algorithm YOLO (You Only Look Once: Unified, Real-Time Object Detection), single shot detector SSD (Single Shot MultiBox Detector), regional convolutional neural network RCNN (Regions with CNN features), etc., or anchor-free methods, such as CenterNet, CornerNet, etc., to perform target detection on the target video, so as to locate the motor vehicle in the target video, determine the positional relationship between the motor vehicle and the lane line, and thus determine whether the motor vehicle is traveling in the target lane.
[0066] Further, in one embodiment, based on the above step S221, the target lane is determined based on the lane line pre-configured in the target video, and based on the target detection algorithm, judging whether the motor vehicle in the target video is traveling in the target lane includes the following steps:
[0067] Step S2211, input the target video into a preset target detection model to obtain key point information of the motor vehicle.
[0068] Specifically, the target detection model may be a CenterNet model. By processing the target video using CenterNet based on key point detection, key point information of the motor vehicle in the target video is obtained. Specifically, the key point information may be the center point T of the motor vehicle. c , left front headlight point L l , right front headlight point L r , Left tire pressure point T l , right tire pressure point T r 、Wide T of the body frame w , high T of the body frame h 、License plate center point P c , Left rearview mirror point M l And right rearview mirror point M r The key point information can be used to characterize the position information of the motor vehicle in the target video.
[0069] Step S2212, calculating the center point of the front projection of the motor vehicle based on the key point information.
[0070] Specifically, in the detected key point information, if the left tire pressure point T l and right tire pressure point T r If both exist, then the center point of the front projection P-lane is the left tire pressure point T l and right tire pressure point T rIf the left tire pressure point T is not detected at the same time l and right tire pressure point T r , then determine the left front headlight point L l And the right front headlight dot L r Do they exist at the same time? If so, the calculation method of the center point of the front projection P-lane is:
[0071]
[0072] Otherwise, if the left rearview mirror point M l and right rearview mirror point M r If both exist, the calculation method of the center point of the front projection P-lane is:
[0073]
[0074] Step S2213, determining whether the motor vehicle is traveling in the target lane based on the positional relationship between the center point of the vehicle head projection and the lane line.
[0075] Specifically, if the center point P-lane of the vehicle head projection falls between two pre-configured lane lines, it is confirmed that the motor vehicle is traveling in the target lane.
[0076] Additionally, in one embodiment, based on the above step S230, obtaining radar data corresponding to the target motor vehicle further includes the following steps:
[0077] Step S231, obtaining radar data in a target traffic scene.
[0078] Specifically, the radar data in the target traffic scene can be provided by a radar device installed in the target traffic scene by transmitting electromagnetic waves to the motor vehicles in the target traffic scene and receiving their echoes. The radar device can be installed adjacent to the monitoring device in the target traffic scene, or can be installed separately from the monitoring device in the target traffic scene. The installation method is not specifically limited in this embodiment.
[0079] Step S232, associating the radar data in the target traffic scene with the corresponding target motor vehicle in the target video through an image calibration algorithm to obtain radar data corresponding to the target motor vehicle.
[0080] Additionally, in one embodiment, based on the above step S232, the radar data is associated with the corresponding target motor vehicle in the target video through an image calibration algorithm, and the following steps are also included:
[0081] Step S2321, mapping the image data in the target video into a bird's-eye view through an image calibration algorithm, and associating the radar data with the corresponding target motor vehicle according to the bird's-eye view.
[0082] Specifically, the video image is calibrated using affine transformation or deep learning methods, and a single frame in the video image is mapped into a bird's-eye view. The coordinates in the bird's-eye view are associated with the radar data to obtain the radar data corresponding to the target motor vehicle.
[0083] Additionally, in one embodiment, based on the above step S220, the vehicle attributes of the target motor vehicle include license plate attributes and vehicle model attributes, and obtaining the vehicle attributes of the target motor vehicle in the target video further includes the following steps:
[0084] Step S222, acquiring a target motor vehicle image of the target motor vehicle, and processing the target motor vehicle image using a license plate detection method to obtain a license plate image of the target motor vehicle.
[0085] Specifically, the target motor vehicle image can be obtained by performing image segmentation on the target motor vehicle located in the target lane in the target video, or by real-time capturing the target motor vehicle in the target lane by a camera set at a specific location. The license plate detection method is not limited to any one of the methods such as YOLO, SSD, RCNN, CenterNet or CornerNet. The target motor vehicle image is processed by using the license plate detection method to segment the license plate area, thereby obtaining the license plate image of the target motor vehicle.
[0086] Step S223, using a license plate recognition algorithm to process the license plate image to obtain the license plate attributes of the target motor vehicle.
[0087] Specifically, the license plate image is recognized by using morphology combined with SVM, or by using deep learning methods such as recurrent neural network (RNN), spatial transformation network (STN) or CNN, so as to obtain the license plate type and license plate number of the license plate image. Among them, the license plate types may include: yellow plate, double-layer yellow plate, black plate, yellow-green double plate, gradient green plate, blue plate and white plate, etc.
[0088] Step S224, using a vehicle type recognition algorithm to process the target motor vehicle image to obtain the vehicle type attributes of the target motor vehicle.
[0089] In addition, the target motor vehicle image can also be input into the CNN network to output the model information of the target motor vehicle.
[0090] In one embodiment, based on the above step S240, when it is determined according to the vehicle attributes and driving information that the target motor vehicle meets the vehicle warning condition, the following steps are also included:
[0091] Step S241, determining identification information of the target motor vehicle based on a video tracking algorithm.
[0092] Specifically, by using a video tracking algorithm to track a target motor vehicle in a target video, the target motor vehicle can correspond to the same motor vehicle in the actual traffic scene from its appearance to its disappearance in the target video, thereby achieving the uniqueness of the target motor vehicle in the target video and avoiding subsequent repeated processing of the target motor vehicle.
[0093] Step S242: in the target traffic scenario, generate warning information for each target motor vehicle according to the identification information.
[0094] Specifically, a warning message can be generated for each target motor vehicle, so that in actual traffic scenarios, each target motor vehicle that meets the vehicle warning conditions can be warned of vehicle information once or a preset number of times, avoiding repeated prompts that may cause interference to surrounding residents.
[0095] In one embodiment, Figure 3 As shown, a method for warning a passing vehicle of a motor vehicle is provided, the method comprising:
[0096] Step S310, inputting a video of a target traffic scene;
[0097] Step S320, configuring lane lines for the video;
[0098] Step S330, performing vehicle detection, tracking and key point detection on the video;
[0099] Step S340, performing radar-visual correlation between the video data and the radar data;
[0100] Step S350, identifying vehicle attributes in the target traffic scene;
[0101] Step S360: Based on the radar data and vehicle attributes, a right-turning large truck warning is issued to the large truck in the right-turn lane.
[0102] In the above steps S210 to S240, based on the target detection algorithm, it is determined whether the motor vehicle in the target video is traveling in the target lane, the vehicle attributes of the target motor vehicle traveling in the target lane are identified to obtain the motor vehicle information in the specific lane, the key point information of the motor vehicle is obtained by using the target detection model, the center point of the front projection of the motor vehicle is calculated according to the key point information, and whether the motor vehicle is traveling in the target lane is determined according to the positional relationship between the center point of the front projection and the lane line, thereby improving the accuracy of judging the positional relationship between the motor vehicle and the target lane, obtaining radar data under the target traffic scene, and associating the radar data under the target traffic scene with the corresponding target motor vehicle in the target video through the image calibration algorithm to obtain the radar data corresponding to the target motor vehicle, thereby improving the accuracy of obtaining the driving information of the target motor vehicle, obtaining the license plate image of the target motor vehicle by using the license plate detection method, obtaining the license plate attributes of the license plate image by using the license plate recognition algorithm, and obtaining the model attributes of the target motor vehicle by using the model recognition algorithm, so that warning information can be generated for the target motor vehicle that meets the vehicle warning information according to the actual traffic scene, thereby realizing timely warning of the motor vehicle information that meets the vehicle warning conditions in the right turn lane.
[0103] In this embodiment, a motor vehicle passing warning device is also provided, which is used to implement the above-mentioned embodiments and preferred implementation modes, and the descriptions that have been made will not be repeated. The terms "module", "unit", "subunit", etc. used below can implement a combination of software and / or hardware of predetermined functions. Although the devices described in the following embodiments are preferably implemented in software, the implementation of hardware, or a combination of software and hardware, is also possible and conceivable.
[0104] Figure 4 4 is a schematic diagram of the structure of the motor vehicle passing warning device 40 in the implementation of the present application. Figure 4 As shown, the motor vehicle passing warning device 40 includes: an acquisition module 42, a judgment module 44, a radar-visual correlation module 46 and a warning module 48, wherein:
[0105] An acquisition module 42 is used to acquire a target video in a target traffic scene, wherein the target video includes at least one motor vehicle and a target lane;
[0106] The judgment module 44 is used to obtain the vehicle attributes of the target motor vehicle in the target video, where the target motor vehicle includes at least one motor vehicle traveling in the target lane;
[0107] A radar-visual correlation module 46 is used to obtain radar data corresponding to the target motor vehicle, wherein the radar data includes driving information of the target motor vehicle;
[0108] The warning module 48 is used to determine whether the target motor vehicle meets the vehicle warning condition according to the vehicle attributes and driving information.
[0109] The above-mentioned motor vehicle passing warning device 40 obtains a target video under a target traffic scene, wherein the target video includes at least one motor vehicle and a target lane, obtains the vehicle attributes of the target motor vehicle in the target video, wherein the target motor vehicle includes at least one motor vehicle traveling in the target lane, obtains radar data corresponding to the target motor vehicle, wherein the radar data includes the driving information of the target motor vehicle, and determines whether the target motor vehicle meets the vehicle warning conditions according to the vehicle attributes and driving information. Through the correlation processing of radar data and video data, it can improve the accuracy of the vehicle information of the motor vehicle in the target lane obtained, thereby improving the accuracy of the detection of the vehicle information to be warned, thereby realizing the timely warning of the motor vehicle information in the target lane that meets the warning conditions.
[0110] In one embodiment, the judgment module 44 is further used to judge whether the motor vehicle in the target video is traveling in the target lane based on the target detection algorithm. If so, the motor vehicle is set as the target motor vehicle and the vehicle attributes of the target motor vehicle are obtained.
[0111] In one of the embodiments, the judgment module 44 is also used to input the target video into a preset target detection model to obtain key point information of the motor vehicle, calculate the center point of the front projection of the motor vehicle based on the key point information, and determine whether the motor vehicle is traveling in the target lane based on the positional relationship between the center point of the front projection and the lane line.
[0112] In one embodiment, the radar-visual association module 46 is also used to obtain radar data in a target traffic scene, and associate the radar data in the target traffic scene with the corresponding target motor vehicle in the target video through an image calibration algorithm to obtain radar data corresponding to the target motor vehicle.
[0113] In one of the embodiments, the radar-visual association module 46 is further configured to map the image data in the target video into a bird's-eye view through an image calibration algorithm, and associate the radar data with the corresponding target motor vehicle according to the bird's-eye view.
[0114] In one of the embodiments, the judgment module 44 is also used to obtain a target motor vehicle image of the target motor vehicle, process the target motor vehicle image using a license plate detection method to obtain a license plate image of the target motor vehicle, process the license plate image using a license plate recognition algorithm to obtain the license plate attributes of the target motor vehicle, and process the target motor vehicle image using a vehicle type recognition algorithm to obtain the vehicle type attributes of the target motor vehicle.
[0115] In one of the embodiments, the warning module 48 is further configured to determine identification information of the target motor vehicle based on a video tracking algorithm, and generate warning information for each target motor vehicle according to the identification information in the target traffic scene.
[0116] It should be noted that the above modules can be functional modules or program modules, and can be implemented by software or hardware. For modules implemented by hardware, the above modules can be located in the same processor; or the above modules can be located in different processors in any combination.
[0117] In one embodiment, Figure 5 As shown, an electronic device is provided, including a memory and a processor. The memory stores a computer program, and the processor of the electronic device is used to provide computing and control capabilities. The memory of the electronic device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium.
[0118] Optionally, the electronic device may further include a transmission device and an input / output device, wherein the transmission device is connected to the processor, and the input / output device is connected to the processor.
[0119] Optionally, in this embodiment, the processor implements the following steps when executing the computer program:
[0120] Acquire a target video in a target traffic scene, wherein the target video includes at least one motor vehicle and a target lane;
[0121] Acquire vehicle attributes of a target motor vehicle in a target video, where the target motor vehicle includes at least one motor vehicle traveling in a target lane;
[0122] Acquiring radar data corresponding to the target motor vehicle, wherein the radar data includes driving information of the target motor vehicle;
[0123] Determine whether the target motor vehicle meets the vehicle warning conditions based on vehicle attributes and driving information.
[0124] In one embodiment, when the processor executes the computer program, the processor further implements the following steps:
[0125] Based on the target detection algorithm, it is determined whether the motor vehicle in the target video is traveling in the target lane. If so, the motor vehicle is set as the target motor vehicle and the vehicle attributes of the target motor vehicle are obtained.
[0126] In one embodiment, when the processor executes the computer program, the processor further implements the following steps:
[0127] Input the target video into a preset target detection model to obtain key point information of the motor vehicle;
[0128] Calculate the center point of the front projection of the motor vehicle according to the key point information;
[0129] According to the positional relationship between the center point of the vehicle head projection and the lane line, it is determined whether the motor vehicle is traveling in the target lane.
[0130] In one embodiment, when the processor executes the computer program, the processor further implements the following steps:
[0131] Acquire radar data in target traffic scenarios;
[0132] The radar data in the target traffic scene is associated with the corresponding target motor vehicle in the target video through the image calibration algorithm to obtain the radar data corresponding to the target motor vehicle.
[0133] In one embodiment, when the processor executes the computer program, the processor further implements the following steps:
[0134] The image data in the target video is mapped into a bird's-eye view through an image calibration algorithm, and the radar data is associated with the corresponding target motor vehicle based on the bird's-eye view.
[0135] In one embodiment, when the processor executes the computer program, the processor further implements the following steps:
[0136] Acquire a target motor vehicle image of the target motor vehicle, and process the target motor vehicle image using a license plate detection method to obtain a license plate image of the target motor vehicle;
[0137] The license plate image is processed using a license plate recognition algorithm to obtain the license plate attributes of the target motor vehicle;
[0138] The target motor vehicle image is processed using a vehicle type recognition algorithm to obtain the vehicle type attributes of the target motor vehicle.
[0139] In one embodiment, when the processor executes the computer program, the processor further implements the following steps:
[0140] Determine identification information of the target motor vehicle based on a video tracking algorithm;
[0141] In the target traffic scenario, warning information is generated for each target motor vehicle according to the identification information.
[0142] The electronic device obtains a target video in a target traffic scene, wherein the target video includes at least one motor vehicle and a target lane, obtains the vehicle attributes of the target motor vehicle in the target video, wherein the target motor vehicle includes at least one motor vehicle traveling in the target lane, obtains radar data corresponding to the target motor vehicle, wherein the radar data includes the driving information of the target motor vehicle, and determines whether the target motor vehicle meets the vehicle warning conditions according to the vehicle attributes and driving information. Through the correlation processing of radar data and video data, the accuracy of the vehicle information of the motor vehicle in the target lane obtained can be improved, thereby improving the accuracy of the detection of the vehicle information to be warned, thereby realizing the timely warning of the motor vehicle information in the target lane that meets the warning conditions.
[0143] In one embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as follows: Figure 6 As shown. The computer device includes a processor, a memory, a network interface and a database connected via a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store a set of preset configuration information. The network interface of the computer device is used to communicate with an external terminal via a network connection. When the computer program is executed by the processor, the above-mentioned motor vehicle passing warning method is implemented.
[0144] In one embodiment, a computer device is provided, which may be a terminal. The computer device includes a processor, a memory, a network interface, a display screen, and an input device connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The network interface of the computer device is used to communicate with an external terminal via a network connection. When the computer program is executed by the processor, a method for warning a passing vehicle of a motor vehicle is implemented. The display screen of the computer device may be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device may be a touch layer covering the display screen, or a key, a trackball, or a touchpad provided on the housing of the computer device, or an external keyboard, touchpad, or mouse, etc.
[0145] Those skilled in the art will understand that Figure 6The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.
[0146] It should be understood that the specific embodiments described herein are only used to explain the application, rather than to limit it. Based on the embodiments provided in this application, all other embodiments obtained by ordinary technicians in this field without creative work are within the protection scope of this application.
[0147] Obviously, the drawings are only some examples or embodiments of the present application. For ordinary technicians in the field, the present application can also be applied to other similar situations based on these drawings without creative work. In addition, it is understandable that although the work done in this development process may be complicated and lengthy, for ordinary technicians in the field, certain changes in design, manufacturing or production based on the technical content disclosed in this application are only conventional technical means and should not be regarded as insufficient content disclosed in this application.
[0148] The term "embodiment" in this application refers to a specific feature, structure or characteristic described in conjunction with the embodiment that can be included in at least one embodiment of the present application. The appearance of this phrase in various locations in the specification does not necessarily mean the same embodiment, nor does it mean that it is mutually exclusive with other embodiments and is independent or optional. It is clearly or implicitly understood by those of ordinary skill in the art that the embodiments described in this application can be combined with other embodiments without conflict.
[0149] The above-mentioned embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of patent protection. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the scope of protection of the present application. Therefore, the scope of protection of the present application shall be subject to the attached claims.
Claims
1. A method for warning a motor vehicle passing by, It is characterized in that The method comprises: Acquire a target video in a target traffic scene, wherein the target video includes at least one motor vehicle and a target lane; Acquire vehicle attributes of a target motor vehicle in the target video, wherein the target motor vehicle includes a motor vehicle among the at least one motor vehicle traveling in the target lane; Acquiring radar data corresponding to the target motor vehicle, wherein the radar data includes driving information of the target motor vehicle; Determine whether the target motor vehicle meets the vehicle warning condition according to the vehicle attributes and the driving information; wherein: the target lane is determined based on the lane line pre-configured in the target video; The step of obtaining vehicle attributes of a target motor vehicle in the target video includes: Inputting the target video into a preset target detection model to obtain key point information of the motor vehicle; Calculate the center point of the front projection of the motor vehicle according to the key point information; According to the positional relationship between the center point of the vehicle head projection and the lane line, it is determined whether the motor vehicle is traveling in the target lane. If so, the motor vehicle is set as a target motor vehicle and the vehicle attributes of the target motor vehicle are obtained.
2. According to the motor vehicle passing warning method according to claim 1, It is characterized in that The obtaining of radar data corresponding to the target motor vehicle includes: Acquiring radar data under the target traffic scene; The radar data in the target traffic scene is associated with the corresponding target motor vehicle in the target video through an image calibration algorithm to obtain the radar data corresponding to the target motor vehicle.
3. The motor vehicle passing warning method according to claim 2, It is characterized in that The step of associating the radar data in the target traffic scene with the corresponding target motor vehicle in the target video by using an image calibration algorithm includes: The image data in the target video is mapped into a bird's-eye view through the image calibration algorithm, and the radar data is associated with the corresponding target motor vehicle according to the bird's-eye view.
4. The motor vehicle passing warning method according to claim 1, It is characterized in that The vehicle attributes of the target motor vehicle include license plate attributes and vehicle model attributes, and the step of obtaining the vehicle attributes of the target motor vehicle in the target video includes: Acquire a target motor vehicle image of the target motor vehicle, and process the target motor vehicle image using a license plate detection method to obtain a license plate image of the target motor vehicle; Processing the license plate image using a license plate recognition algorithm to obtain the license plate attributes of the target motor vehicle; The target motor vehicle image is processed using a vehicle type recognition algorithm to obtain the vehicle type attributes of the target motor vehicle.
5. The motor vehicle passing warning method according to claim 1, It is characterized in that In the case where it is determined, based on the vehicle attributes and the driving information, that the target motor vehicle meets the vehicle warning condition, the method further includes: Determining identification information of the target motor vehicle based on a video tracking algorithm; In the target traffic scenario, warning information is generated for each target motor vehicle according to the identification information.
6. A vehicle passing warning device, It is characterized in that include: Acquisition module, judgment module, radar and visual correlation module and warning module, among which: The acquisition module is used to acquire a target video in a target traffic scene, wherein the target video contains at least one motor vehicle and a target lane; The judgment module is used to obtain the vehicle attributes of the target motor vehicle in the target video, and the target motor vehicle includes the motor vehicle traveling in the target lane among the at least one motor vehicle; wherein: the target lane is determined based on the lane line pre-configured in the target video; the obtaining of the vehicle attributes of the target motor vehicle in the target video includes: inputting the target video into a preset target detection model to obtain key point information of the motor vehicle; calculating the center point of the front projection of the motor vehicle according to the key point information; determining whether the motor vehicle is traveling in the target lane according to the positional relationship between the center point of the front projection and the lane line, and if so, setting the motor vehicle as a target motor vehicle and obtaining the vehicle attributes of the target motor vehicle; The radar-visual correlation module is used to obtain radar data corresponding to the target motor vehicle, wherein the radar data includes driving information of the target motor vehicle; The warning module is used to determine whether the target motor vehicle meets the vehicle warning condition according to the vehicle attributes and the driving information.
7. An electronic device comprising a memory and a processor, It is characterized in that The memory stores a computer program, and the processor implements the steps of the method according to any one of claims 1 to 5 when executing the computer program.
8. A computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, It is characterized in that When the processor executes the computer program, the steps of the method according to any one of claims 1 to 5 are implemented.
9. A computer-readable storage medium, It is characterized in that The storage medium stores a computer program, wherein the computer program is configured to execute the steps of the method according to any one of claims 1 to 5 when executed.
