Intersection guide line analysis method and system based on radar and vision fusion

Abstract

The application discloses a crossroad guide line analysis method and system based on radar and video fusion. Video image data and radar detection data are fused to obtain relative space-time information; absolute space-time information of all vehicles is obtained according to map information and the relative space-time information; actual vehicle track width of each vehicle is obtained according to vehicle model and vehicle width; the actual vehicle track width is projected to the map information according to the actual vehicle track width and the vehicle longitude and latitude to obtain the simulated vehicle track width of each vehicle; the line segment extension of the simulated vehicle track width is carried out with the vehicle longitude and latitude at each moment as the center point and in the direction perpendicular to the vehicle direction angle direction to obtain the vehicle track at each moment; the vehicle tracks are connected by the end points to form the wheel tracks; the wheel track heat map is obtained by heat map effect processing of the wheel tracks; the wheel track heat map is transferred to the map information according to the vehicle longitude and latitude to analyze whether the guide line drawing is correct.

Description

Technical Field

[0001] This application relates to the field of intelligent transportation technology, and in particular to a method and system for analyzing traffic flow at intersections based on radar-visual fusion. Background Technology

[0002] Optimizing traffic conditions at intersections is key to preventing and alleviating traffic congestion, and the proper drawing of guide lines at intersections is crucial for controlling vehicles. Guide lines mainly take the form of one or more white V-shaped lines or diagonal lines set according to the intersection terrain, indicating that vehicles must travel along the prescribed route and must not cross or drive over the lines. They are mainly used for intersections that are too wide, irregular, or have complex driving conditions.

[0003] However, traditional methods use water spraying to observe tire tracks and analyze the guide lines at intersections to determine their rationality and whether they impede traffic flow. This traditional method can only analyze a specific section of the road at a time, resulting in high manpower consumption and low efficiency. Summary of the Invention

[0004] The purpose of this application is to solve the technical problems of traditional methods being labor-intensive and inefficient. To achieve the above objective, this application provides a method and system for analyzing traffic flow at intersections based on radar-visual fusion.

[0005] This application provides a method for analyzing traffic flow lines at intersections based on radar-visual fusion, including:

[0006] Acquire video image data and radar detection data;

[0007] The video image data and the radar detection data are fused to obtain the relative spatiotemporal information of each vehicle relative to the intersection.

[0008] Obtain map information of the intersection, and obtain absolute spatiotemporal information of all vehicles at the intersection based on the map information and the relative spatiotemporal information;

[0009] Based on the vehicle model and vehicle width in the absolute spatiotemporal information of each vehicle, the actual rut width of each vehicle is obtained. Based on the actual rut width and the vehicle latitude and longitude in the absolute spatiotemporal information, the latitude and longitude of the actual rut width are obtained. The latitude and longitude of the actual rut width are then projected onto the map information to obtain the simulated rut width of each vehicle.

[0010] Using the vehicle's latitude and longitude at each time as the center point, extend a line segment with the simulated rut width in a direction perpendicular to the vehicle's direction angle to obtain the ruts of each vehicle at each time. Connect the endpoints of the ruts of each vehicle at each time to form the wheel track of each vehicle.

[0011] Heatmap processing is performed on the wheel trajectories of each vehicle at the intersection to obtain a heatmap of the wheel trajectories at the intersection.

[0012] Based on the vehicle's latitude and longitude at each time, the heat map of the wheel trajectory at the intersection is transferred to the map information for analysis to determine whether the guide lines of the intersection are drawn correctly.

[0013] In one embodiment, after obtaining the rut width of each vehicle at each time by extending a line segment perpendicular to the vehicle's direction angle in the absolute spatiotemporal information of each vehicle at each time as the center point, and connecting the endpoints of the ruts of each vehicle at each time to form the wheel track of each vehicle, before performing heat map processing on the wheel tracks of each vehicle at the intersection to obtain the wheel track heat map of the intersection, the method further includes:

[0014] Based on the inbound / outbound directions and lane information in the map information, the wheel trajectories of each vehicle are classified to form an inbound direction wheel trajectory set, an outbound direction wheel trajectory set, and a lane wheel trajectory set.

[0015] In one embodiment, the step of performing heatmap processing on the wheel trajectories of each vehicle at the intersection to obtain a wheel trajectory heatmap of the intersection includes:

[0016] The wheel tracks corresponding to the approach directions at the intersection are collected and processed into a heat map effect for each wheel track to obtain an approach wheel track heat map; or

[0017] The exit wheel tracks corresponding to the intersections are collected and processed into a heatmap effect for each wheel track to obtain an exit wheel track heatmap; or

[0018] Heatmap effects are applied to each wheel trajectory in the lane corresponding to the intersection to obtain a lane wheel trajectory heatmap.

[0019] In one embodiment, the step of transferring the wheel trajectory heatmap of the intersection to the map information for analysis based on the vehicle latitude and longitude of each vehicle at each time moment, and determining whether the guide lines of the intersection are drawn correctly, includes:

[0020] Based on the vehicle's latitude and longitude at each time point, the heat map of the approaching wheel trajectory is transferred to the map information for analysis to determine whether the guide lines for the approaching direction of the intersection are drawn correctly; or

[0021] Based on the vehicle's latitude and longitude at each time point, the heat map of the exit wheel trajectory is transferred to the map information for analysis to determine whether the guide lines for the exit direction of the intersection are drawn correctly; or

[0022] Based on the vehicle's latitude and longitude at each time, the heat map of the lane wheel trajectory is transferred to the map information for analysis to determine whether the lane guide lines at the intersection are drawn correctly.

[0023] In one embodiment, this application provides a method for analyzing traffic flow lines at intersections based on radar-visual fusion, including:

[0024] Acquire video image data and radar detection data;

[0025] The video image data and the radar detection data are fused to obtain the relative spatiotemporal information of each vehicle relative to the intersection.

[0026] Obtain map information of the intersection, and obtain absolute spatiotemporal information of all vehicles at the intersection based on the map information and the relative spatiotemporal information;

[0027] Based on the vehicle model and vehicle width in the absolute spatiotemporal information of each vehicle, the actual rut width of each vehicle is obtained. Based on the actual rut width and the vehicle latitude and longitude in the absolute spatiotemporal information, the latitude and longitude of the actual rut width are obtained. The latitude and longitude of the actual rut width are then projected onto the map information to obtain the simulated rut width of each vehicle.

[0028] Connect the vehicle's latitude and longitude coordinates in the absolute spatiotemporal information of each vehicle at each moment as the center point to obtain the vehicle center point trajectory of each vehicle.

[0029] Using each center point in the vehicle's center point trajectory as a vertex and the trajectories of adjacent center points as edges, construct angle bisectors with a length equal to the width of the ruts to the inner and outer angles to obtain the ruts of each vehicle at each time moment, and connect the endpoints of the ruts of each vehicle at each time moment to form the wheel trajectory of each vehicle.

[0030] Heatmap processing is performed on the wheel trajectories of each vehicle at the intersection to obtain a heatmap of the wheel trajectories at the intersection.

[0031] Based on the vehicle's latitude and longitude at each time, the heat map of the wheel trajectory at the intersection is transferred to the map information for analysis to determine whether the guide lines of the intersection are drawn correctly.

[0032] In one embodiment, this application provides an intersection traffic flow analysis system based on radar-visual fusion, comprising:

[0033] The data acquisition module is used to acquire video image data and radar detection data;

[0034] The relative spatiotemporal information acquisition module is used to fuse the video image data with the radar detection data to obtain the relative spatiotemporal information of each vehicle relative to the intersection.

[0035] An absolute spatiotemporal information acquisition module is used to acquire map information of the intersection and obtain absolute spatiotemporal information of all vehicles at the intersection based on the map information and the relative spatiotemporal information.

[0036] The rut width acquisition module is used to obtain the actual rut width of each vehicle based on the vehicle model and vehicle width in the absolute spatiotemporal information of each vehicle, obtain the latitude and longitude of the actual rut width based on the actual rut width and the vehicle latitude and longitude in the absolute spatiotemporal information, and project the latitude and longitude of the actual rut width onto the map information to obtain the simulated rut width of each vehicle.

[0037] The first wheel trajectory acquisition module is used to extend the simulated rut width of a line segment in a direction perpendicular to the vehicle's direction angle, with the vehicle's latitude and longitude as the center point, to obtain the ruts of each vehicle at each time, and connect the endpoints of the ruts of each vehicle at each time to form the wheel trajectory of each vehicle.

[0038] The wheel trajectory heatmap acquisition module is used to process the wheel trajectory of each vehicle at the intersection to obtain a wheel trajectory heatmap of the intersection.

[0039] The analysis module is used to transfer the wheel trajectory heatmap of the intersection to the map information for analysis based on the vehicle latitude and longitude of each vehicle at each time, and to determine whether the guide lines of the intersection are drawn correctly.

[0040] In one embodiment, the intersection traffic flow analysis system based on radar-visual fusion further includes:

[0041] The wheel trajectory set acquisition module is used to classify the wheel trajectory of each vehicle according to the entrance / exit direction and lane information in the map information, forming an entrance direction wheel trajectory set, an exit direction wheel trajectory set, and a lane wheel trajectory set.

[0042] In one embodiment, the wheel trajectory heatmap acquisition module includes:

[0043] An import wheel trajectory heatmap acquisition module is used to collect the import wheel trajectories corresponding to the intersection in the import direction, process each wheel trajectory to obtain an import wheel trajectory heatmap; or

[0044] The exit wheel trajectory heatmap acquisition module is used to collect the wheel trajectories of the exit direction corresponding to the intersection, process each wheel trajectory to obtain an exit wheel trajectory heatmap; or

[0045] The lane wheel trajectory heatmap acquisition module is used to process each wheel trajectory in the lane wheel trajectory set corresponding to the intersection with a heatmap effect to obtain a lane wheel trajectory heatmap.

[0046] In one embodiment, the analysis module includes:

[0047] The first analysis and judgment module is used to transfer the heat map of the approaching wheel trajectory to the map information for analysis based on the vehicle's latitude and longitude at each time, and to determine whether the guide line drawn in the approach direction of the intersection is correct; or

[0048] The second analysis and judgment module is used to transfer the heat map of the exit wheel trajectory to the map information for analysis based on the vehicle latitude and longitude of each vehicle at each time, and to determine whether the guide line of the exit direction of the intersection is drawn correctly; or

[0049] The third analysis and judgment module is used to transfer the heat map of the lane wheel trajectory to the map information for analysis based on the vehicle latitude and longitude of each vehicle at each time, and to determine whether the lane guide lines of the intersection are drawn correctly.

[0050] In one embodiment, this application provides an intersection traffic flow analysis system based on radar-visual fusion, comprising:

[0051] The data acquisition module is used to acquire video image data and radar detection data;

[0052] The relative spatiotemporal information acquisition module is used to fuse the video image data with the radar detection data to obtain the relative spatiotemporal information of each vehicle relative to the intersection.

[0053] An absolute spatiotemporal information acquisition module is used to acquire map information of the intersection and obtain absolute spatiotemporal information of all vehicles at the intersection based on the map information and the relative spatiotemporal information.

[0054] The rut width acquisition module is used to obtain the actual rut width of each vehicle based on the vehicle model and vehicle width in the absolute spatiotemporal information of each vehicle, obtain the latitude and longitude of the actual rut width based on the actual rut width and the vehicle latitude and longitude in the absolute spatiotemporal information, and project the latitude and longitude of the actual rut width onto the map information to obtain the simulated rut width of each vehicle.

[0055] The vehicle center point trajectory acquisition module is used to connect the vehicle latitude and longitude of each vehicle in the absolute spatiotemporal information at each time moment with the vehicle latitude and longitude as the center point to obtain the vehicle center point trajectory of each vehicle.

[0056] The second wheel trajectory acquisition module is used to construct angle bisectors with a length equal to the width of the ruts to the inner and outer angles, with each center point in the vehicle center point trajectory as a vertex and the trajectories of adjacent center points as edges, to obtain the ruts of each vehicle at each time moment, and connect the ruts of each vehicle at each time moment to form the wheel trajectory of each vehicle.

[0057] The wheel trajectory heatmap acquisition module is used to process the wheel trajectory of each vehicle at the intersection to obtain a wheel trajectory heatmap of the intersection.

[0058] The analysis module is used to transfer the wheel trajectory heatmap of the intersection to the map information for analysis based on the vehicle latitude and longitude of each vehicle at each time, and to determine whether the guide lines of the intersection are drawn correctly.

[0059] The aforementioned intersection traffic flow analysis method and system based on radar-visual fusion performs real-time measurements of the intersection using video image data and radar detection data to obtain the absolute spatiotemporal information of each vehicle at the intersection. Based on this absolute spatiotemporal information, the actual rut width of each vehicle is obtained, and this width is projected onto a high-precision map using latitude and longitude information to simulate the rut width of each vehicle. Based on the simulated rut width of each vehicle, a line segment extending from the vehicle's latitude and longitude at each moment in a direction perpendicular to the vehicle's heading angle is formed, creating the rut width for each vehicle at that moment. Connecting the rut widths of each vehicle at each moment forms the wheel trajectory for each vehicle, and a wheel trajectory heatmap is generated based on this trajectory. By transferring the wheel trajectory heatmap to the high-precision map, the rationality of the intersection traffic flow drawing can be judged. This judgment determines the accuracy of the drawing, completing the analysis of the rationality of the intersection traffic flow drawing, and accurately reflecting the relationship between vehicles and road traffic flow.

[0060] Therefore, the intersection guide line analysis method based on radar-visual fusion provided in this application can analyze the rationality of the intersection guide lines on a large scale, accurately and extensively provide the intersection guide line drawing and trajectory distribution, reduce manual experimentation and statistics, save manpower and improve analysis efficiency, and solve the problem of low efficiency and high manpower caused by the traditional method of using water to look at vehicle ruts. Attached Figure Description

[0061] Figure 1 This is a flowchart illustrating the steps of a crossroads traffic flow analysis method based on radar-visual fusion in one embodiment of this application.

[0062] Figure 2 This application provides Figure 1 A schematic diagram of the formation of the wheel track in the illustrated embodiment.

[0063] Figure 3 This is a flowchart illustrating the steps of a crossroads traffic flow analysis method based on radar-visual fusion in another embodiment provided in this application.

[0064] Figure 4 This application provides Figure 3 A schematic diagram illustrating the formation of the vehicle center point trajectory in the illustrated embodiment.

[0065] Figure 5 This application provides Figure 3 A schematic diagram of the formation of the wheel track in the illustrated embodiment. Detailed Implementation

[0066] The technical solution of this application will be further described in detail below with reference to the accompanying drawings and embodiments.

[0067] Please see Figure 1 This application provides a method for analyzing traffic flow lines at intersections based on radar-visual fusion, including:

[0068] S110, acquires video image data and radar detection data;

[0069] S120 fuses video image data with radar detection data to obtain the relative spatiotemporal information of each vehicle relative to the intersection.

[0070] S130: Obtain map information of the intersection, and obtain absolute spatiotemporal information of all vehicles at the intersection based on the map information and relative spatiotemporal information;

[0071] S140: Based on the vehicle model and vehicle width in the absolute spatiotemporal information of each vehicle, obtain the actual rut width of each vehicle. Based on the actual rut width and the vehicle's latitude and longitude in the absolute spatiotemporal information, obtain the latitude and longitude of the actual rut width. Project the latitude and longitude of the actual rut width into the map information to obtain the simulated rut width of each vehicle.

[0072] S150, taking the vehicle's latitude and longitude at each moment as the center point, extends a line segment with simulated rut width in a direction perpendicular to the vehicle's heading angle to obtain the ruts of each vehicle at each moment, and connects the endpoints of the ruts of each vehicle at each moment to form the wheel track of each vehicle.

[0073] S160, perform heat map processing on the wheel trajectory of each vehicle at the intersection to obtain a heat map of the wheel trajectory of the intersection.

[0074] S170, based on the latitude and longitude of each vehicle at each time, transfers the heat map of the wheel trajectory at the intersection to the map information for analysis, and determines whether the guide lines of the intersection are drawn correctly.

[0075] In this embodiment, in S110, video image data and radar detection data can be obtained through a radar-visual integrated machine. The radar-visual integrated machine is installed at the intersection. In S120 and S130, video processing software extracts images frame by frame from the video images, and performs detection and recognition on each frame based on deep learning methods to obtain vehicle information captured by the camera, such as vehicle headway, vehicle headway, vehicle license plate number, vehicle color, vehicle license plate color, vehicle width, vehicle model, and direction angle. Radar detection can acquire vehicle information based on radar, such as vehicle distance, vehicle orientation, vehicle speed, vehicle acceleration, and vehicle spatial position, even in harsh detection environments and adverse conditions.

[0076] Radar detection data can be used to determine the relative distance of vehicles to the intersection. Video image data can provide vehicle attribute information (such as license plate number, license plate color, and vehicle color) and time information for each frame. By fusing the video image data and radar detection data, the relative spatiotemporal information of each vehicle relative to the intersection can be obtained. This relative spatiotemporal information is based on the intersection as a reference. It can be understood as the spatial position of a vehicle at different times, specifically including time information, vehicle position information, vehicle speed information, vehicle direction information, vehicle model information, and other multi-dimensional information. The relative spatiotemporal information of each vehicle relative to the intersection includes the spatial position of each vehicle relative to the intersection at different times.

[0077] High-precision maps are obtained through actual surveying and mapping, providing realistic twin information of the physical world environment, including reference lines, lane lines, center lines, curbs, corrugated guardrails, concrete guardrails, pedestrian overpasses, traffic signs, delineators, bridge piers, arrows, text, symbols, warning zones, traffic dividers, traffic lights, stop locations, pedestrian crossings, public transport stops, speed reduction zones, bicycle lanes within intersections, no-parking zones, manhole covers, parking spaces, traffic light poles, timing signs, traffic signals, traffic islands, overpasses, checkpoints, intersection center circles, tunnel walls, bus bays, central medians, intersection surfaces, and tollbooth elements. Map information includes, but is not limited to, lane areas, lane marking locations, channelization information, and ground markings.

[0078] Based on the conversion relationship between relative positions, relative spatiotemporal information can be mapped to absolute spatiotemporal information. Absolute spatiotemporal information is independent of the intersection's location and the location of the radar-sensing integrated machine; it is data with independent temporal and spatial information. For example, in one embodiment, a high-precision map can provide the stop line outline location information, lane line outline location information, and latitude and longitude information of the intersection. Relative spatiotemporal information includes the relative spatial position of a vehicle relative to the intersection. Based on the vehicle's distance from the intersection, the relative spatial position can be converted into an absolute spatial position in a geographic coordinate system, obtaining the vehicle's corresponding latitude and longitude information. Furthermore, based on the vehicle's latitude and longitude information at various times, vehicles can be marked on the map and combined with the high-precision map generated by surveying to provide relevant geographic location information support. By fusing video image data and radar detection data to obtain the relative spatiotemporal information of each vehicle, the relative spatiotemporal information of all vehicles passing through the intersection can be further obtained. Combined with the vehicle's distance from the intersection, the absolute spatiotemporal information of all vehicles passing through the intersection can be obtained.

[0079] In S140, the vehicle model indicates the brand and type of vehicle. The rut width can be calculated by identifying the vehicle model and vehicle width. Ruts are the tire marks left by a vehicle on the road surface, providing a reference for road maintenance, repair, and resurfacing. The actual rut width can be understood as the distance between the tire marks left by the two wheels (left and right) on the actual road surface. In this step, the actual rut width is the width of the actual tire marks left by the vehicle while driving on the road. The vehicle position is represented by its latitude and longitude. The vehicle's latitude and longitude represent the location of its center point. Based on the vehicle's latitude and longitude representing its center point and the actual rut width, the latitude and longitude corresponding to the actual rut width can be calculated. Projecting these latitude and longitude onto a high-precision map yields the simulated rut width projected onto the high-precision map. The simulated rut width is the width projected onto the high-precision map for each vehicle's actual rut width.

[0080] Please see Figure 2 , Figure 2 The solid black dots represent the latitude and longitude of each vehicle, the arrows indicate the direction of the vehicle's heading angle, the dashed line segment represents the simulated rut width, and the two solid curves represent the wheel tracks of each vehicle. These two solid curves can be understood as the tracks of the left and right wheels, thus forming the vehicle's wheel trajectory. In S150, extending a line segment of rut width perpendicular to the heading angle from the vehicle's latitude and longitude as the center point yields the ruts at each moment centered on the vehicle's latitude and longitude (which can also be understood as the vehicle's center point). Connecting the ruts from multiple moments sequentially forms the wheel track of each vehicle.

[0081] In S160, the heatmap effect processing can be understood as displaying the data in different brightness levels, reflecting the depth of color and representing the degree of overlap in the trajectories of various vehicles. In this step, the wheel trajectories of each vehicle are overlaid as semi-transparent curves with varying widths. As the wheel trajectories are continuously overlaid, darker colors represent higher frequency of occurrence, while lighter colors represent lower frequency, indicating the degree of overlap in the driving positions of various vehicles, thus forming a wheel trajectory heatmap.

[0082] In S170, the wheel trajectory heatmap is formed from the wheel trajectories of multiple vehicles, with the vehicle's latitude and longitude as the center point. Based on the vehicle's latitude and longitude, the wheel trajectory heatmap can be transferred to a high-precision map for analysis. At intersections, vehicles must follow the direction of the guide lines. Using the high-precision map as a reference, the wheel trajectory heatmap is displayed at the corresponding intersection location on the map. Combined with the guide line information for each direction, entrance / exit, and lane at the intersection on the map, the positional relationship between the wheel trajectories of each vehicle in the heatmap and the guide lines on the map can be accurately determined. This allows for assessment of any boundary conflicts between the wheel trajectory heatmap and the guide lines on the map, as well as any unreasonable situations such as line crossings. This helps determine if the guide lines at the intersection are drawn correctly. If the guide lines at the intersection are drawn incorrectly, they can be redrawn with reference to the wheel trajectory heatmap or the original guide lines can be corrected to ensure that the guide lines at the intersection are applicable to most vehicle routes. If the traffic flow lines at the intersection are drawn correctly, then the traffic flow lines at the intersection will be retained.

[0083] This application provides a method for analyzing traffic flow lines at intersections based on radar-visual fusion. It performs real-time measurements of intersections using video image data and radar detection data to obtain the absolute spatiotemporal information of each vehicle at the intersection. Based on this information, the actual rut width of each vehicle is obtained, and the actual rut width on the road surface is projected onto a high-precision map using latitude and longitude information to simulate the rut width of each vehicle. Based on the simulated rut width of each vehicle, a line segment extending from the vehicle's latitude and longitude at each moment in a direction perpendicular to the vehicle's heading angle is formed, creating the rut width for each vehicle at each moment. Connecting the rut widths of each vehicle at each moment forms the wheel trajectory for each vehicle, and a wheel trajectory heatmap is generated based on these trajectories. By transferring the wheel trajectory heatmap to a high-precision map, the rationality of the intersection traffic flow line drawing can be judged. This judgment determines whether the drawing is accurate, completing the analysis of the rationality of the intersection traffic flow line drawing and accurately reflecting the relationship between vehicles and road traffic flow lines. Therefore, the intersection guide line analysis method based on radar-visual fusion provided in this application can analyze the rationality of the intersection guide lines on a large scale, accurately and extensively provide the intersection guide line drawing and trajectory distribution, reduce manual experimentation and statistics, save manpower and improve analysis efficiency, and solve the problem of low efficiency and high manpower caused by the traditional method of using water to look at vehicle ruts.

[0084] Please see Figure 3 In one embodiment, this application provides a method for analyzing traffic flow lines at intersections based on radar-visual fusion, comprising:

[0085] S210, acquires video image data and radar detection data;

[0086] S220 fuses video image data with radar detection data to obtain the relative spatiotemporal information of each vehicle relative to the intersection.

[0087] S230: Obtain map information of the intersection, and obtain absolute spatiotemporal information of all vehicles at the intersection based on the map information and relative spatiotemporal information;

[0088] S240: Based on the vehicle model and vehicle width in the absolute spatiotemporal information of each vehicle, the actual rut width of each vehicle is obtained. Based on the actual rut width and the vehicle's latitude and longitude in the absolute spatiotemporal information, the latitude and longitude of the actual rut width are obtained. The latitude and longitude of the actual rut width are then projected onto the map information to obtain the simulated rut width of each vehicle.

[0089] S250 connects the latitude and longitude of each vehicle at each moment as the center point to obtain the trajectory of the center point of each vehicle.

[0090] S260: Using each center point in the vehicle's center point trajectory as a vertex and the trajectories of adjacent center points as edges, construct angle bisectors with a length equal to the width of the ruts to the inner and outer angles to obtain the ruts of each vehicle at each time moment, and connect the endpoints of the ruts of each vehicle at each time moment to form the wheel trajectory of each vehicle.

[0091] S270, perform heat map processing on the wheel trajectory of each vehicle at the intersection to obtain a heat map of the wheel trajectory of the intersection.

[0092] S280, based on the latitude and longitude of each vehicle at each time, transfers the heat map of the wheel trajectory at the intersection to the map information for analysis, and determines whether the guide lines of the intersection are drawn correctly.

[0093] In this embodiment, the descriptions of each step from S210 to S240 can be found in the descriptions of each step from S110 to S140 in the above embodiment. Please refer to [link to previous document]. Figure 4 and Figure 5 In S250, the latitude and longitude of each vehicle at each moment are used as the center point. Multiple center points are connected one by one to form the vehicle center point trajectory. The vehicle latitude and longitude represent the position of the vehicle's center point, and the center point trajectory also represents the driving route of each vehicle.

[0094] In S260, each center point in the vehicle's center point trajectory is considered a vertex, and the trajectories between adjacent center points are considered edges. Each center point corresponds to two edges, resulting in interior and exterior angles, as shown in the figure. Using half the width of the rut as the angle bisector of the interior angle and half the width of the rut as the angle bisector of the exterior angle, angle bisectors with a length equal to the width of the rut are constructed for both the interior and exterior angles, as shown in the figure. Figure 5 As shown by the dashed line segment, the endpoints of the angle bisector form the tire tracks of the vehicle at each moment. Connecting the endpoints of the tire tracks of each vehicle at each moment forms the wheel track of each vehicle, as shown in the image. Figure 5 As shown by the solid curve lines.

[0095] In this embodiment, the description of step S270 can be referred to the description of step S160 in the above embodiment. The description of step S280 can be referred to the description of step S170 in the above embodiment.

[0096] This application provides a method for analyzing traffic flow lines at intersections based on radar-visual fusion. It performs real-time measurements of intersections using video image data and radar detection data to obtain the absolute spatiotemporal information of each vehicle at the intersection. Based on this information, the actual rut width of each vehicle is obtained, and the actual rut width on the road surface is projected onto a high-precision map using latitude and longitude information to simulate the rut width of each vehicle. Based on the simulated rut width of each vehicle, angle bisectors are constructed from each center point of the vehicle's trajectory as vertices and the trajectories of adjacent center points as edges, forming the ruts of each vehicle at each moment. Connecting the ruts of each vehicle at each moment forms the wheel trajectory of each vehicle, and a wheel trajectory heatmap is generated based on these trajectories. By transferring the wheel trajectory heatmap to a high-precision map, the rationality of the intersection traffic flow line drawing can be judged. This allows for the analysis of the rationality of the intersection traffic flow line drawing, accurately reflecting the relationship between vehicles and road traffic flow lines. Therefore, the intersection guide line analysis method based on radar-visual fusion provided in this application can analyze the rationality of the intersection guide lines on a large scale, accurately and extensively provide the intersection guide line drawing and trajectory distribution, reduce manual experimentation and statistics, save manpower and improve analysis efficiency, and solve the problem of low efficiency and high manpower caused by the traditional method of using water to look at vehicle ruts.

[0097] In one embodiment, S150, taking the latitude and longitude of each vehicle at each time as the center point, a line segment with the simulated rut width is extended in a direction perpendicular to the vehicle's heading angle to obtain the ruts of each vehicle at each time. After connecting the endpoints of the ruts of each vehicle at each time to form the wheel trajectory of each vehicle, S160, heatmap processing is performed on the wheel trajectory of each vehicle at the intersection to obtain the heatmap of the wheel trajectory of the intersection. Before this, the intersection guide line analysis method based on radar-visual fusion also includes:

[0098] S151. Based on the import / export directions and lane information in the map information, the wheel trajectories of each vehicle are classified to form an import direction wheel trajectory set, an export direction wheel trajectory set, and a lane wheel trajectory set.

[0099] In this embodiment, by classifying the actual lanes, approach directions, and exit directions of the intersection, a set of wheel trajectories for the approach direction, a set of wheel trajectories for the exit direction, and a set of wheel trajectories for each lane can be obtained. By classifying the wheel trajectory sets, different judgments can be made on the correctness of different guide lines. Using the wheel trajectory sets for the approach direction, the wheel trajectory sets for the exit direction, and the lane wheel trajectory sets, the rationality judgment and analysis of the guide lines for the approach direction, the guide lines for the exit direction, and the guide lines for each lane can be performed separately, which is beneficial for judging and analyzing a specific guide line at a specific intersection.

[0100] In one embodiment, S160, heatmap processing is performed on the wheel trajectories of each vehicle at the intersection to obtain a heatmap of the wheel trajectories at the intersection, including:

[0101] S161, collect the wheel tracks of the approaching vehicles at the intersection, process each wheel track into a heatmap, and obtain an approaching vehicle wheel track heatmap; or

[0102] S162, collect the wheel tracks corresponding to the exit directions of the intersection, process each wheel track into a heatmap, and obtain an exit wheel track heatmap; or

[0103] S163: Collect the wheel trajectories of the lanes corresponding to the intersection and process each wheel trajectory into a heat map to obtain a lane wheel trajectory heat map.

[0104] In this embodiment, heatmap effects are processed based on the wheel trajectory sets for the inbound direction, the outbound direction, and the lane wheel trajectory sets, respectively, to obtain corresponding inbound wheel trajectory heatmaps, outbound wheel trajectory heatmaps, and lane wheel trajectory heatmaps. The wheel trajectories in the inbound direction wheel trajectory set are overlaid with semi-transparent curves with width to highlight the high-frequency positions of vehicles in the inbound direction, showing the overlap of vehicle positions and thus determining the rationality of the guide lines drawn in the inbound direction. Similarly, the wheel trajectories in the outbound direction wheel trajectory set are overlaid with semi-transparent curves with width to highlight the high-frequency positions of vehicles in the outbound direction, showing the overlap of vehicle positions and thus determining the rationality of the guide lines drawn in the outbound direction. Likewise, the wheel trajectories in the lane wheel trajectory set are overlaid with semi-transparent curves with width to highlight the high-frequency positions of vehicles in each lane, showing the overlap of vehicle positions and thus determining the rationality of the guide lines drawn in each lane.

[0105] In one embodiment, S170, based on the latitude and longitude of each vehicle at each time moment, the heat map of the wheel trajectory at the intersection is transferred to map information for analysis to determine whether the guide lines of the intersection are drawn correctly, including:

[0106] S171, based on the latitude and longitude of each vehicle at each time moment, the heat map of the approach wheel trajectory is transferred to the map information for analysis to determine whether the guide lines of the approach direction of the intersection are drawn correctly; or

[0107] S172, based on the latitude and longitude of each vehicle at each time moment, transfer the heat map of the exit wheel trajectory to the map information for analysis to determine whether the guide lines for the exit direction of the intersection are drawn correctly; or

[0108] S173, based on the vehicle latitude and longitude of each vehicle at each time, transfers the heat map of the lane wheel trajectory to the map information for analysis to determine whether the lane guide lines of the intersection are drawn correctly.

[0109] In this embodiment, the import wheel trajectory heatmap is formed using the wheel trajectories of multiple vehicles, with the vehicle's latitude and longitude as the center point. The high-precision map information includes the latitude and longitude information of each intersection, lane, and entrance / exit. The import wheel trajectory heatmap carries the latitude and longitude information of each location. Based on this information, the import wheel trajectory heatmap is transferred to the high-precision map for analysis, and the heatmap is displayed on the map at the corresponding import direction location. Therefore, by comparing the guide lines in the import direction on the high-precision map with the import wheel trajectory heatmap, the accuracy of the guide lines in the import direction can be analyzed. Similarly, the export wheel trajectory heatmap carries the latitude and longitude information of each location. Based on this information, the export wheel trajectory heatmap is transferred to the high-precision map for analysis, and the heatmap is displayed on the map at the corresponding export direction location. Therefore, by comparing the guide lines in the export direction on the high-precision map with the export wheel trajectory heatmap, the accuracy of the guide lines in the export direction can be analyzed. Similarly, the lane wheel trajectory heatmap carries the latitude and longitude information of each location. Based on this information, the lane wheel trajectory heatmap is transferred to a high-precision map for analysis, and the heatmap is then reflected in the corresponding lane location on the map. Therefore, by comparing the lane guide lines in the high-precision map with the lane wheel trajectory heatmap, it is possible to analyze whether the lane guide lines are accurately drawn.

[0110] In one embodiment, this application provides a traffic flow analysis system for intersections based on radar-visual fusion. The system includes a data acquisition module, a relative spatiotemporal information acquisition module, an absolute spatiotemporal information acquisition module, a rut width acquisition module, a first wheel trajectory acquisition module, a wheel trajectory heatmap acquisition module, and an analysis module. The data acquisition module acquires video image data and radar detection data. The relative spatiotemporal information acquisition module fuses the video image data and radar detection data to obtain the relative spatiotemporal information of each vehicle relative to the intersection. The absolute spatiotemporal information acquisition module acquires map information of the intersection and obtains the absolute spatiotemporal information of all vehicles at the intersection based on the map information and the relative spatiotemporal information.

[0111] The rut width acquisition module obtains the actual rut width of each vehicle based on its model and width in the absolute spatiotemporal information. It then obtains the latitude and longitude of the actual rut width using the vehicle's latitude and longitude in the absolute spatiotemporal information, and projects this latitude and longitude onto the map information to obtain the simulated rut width for each vehicle. The first wheel trajectory acquisition module extends a line segment centered on the vehicle's latitude and longitude at each moment, perpendicular to the vehicle's heading angle, to obtain the ruts for each vehicle at each moment. It then connects the endpoints of each vehicle's ruts at each moment to form the wheel trajectory for each vehicle. The wheel trajectory heatmap acquisition module processes the wheel trajectories of each vehicle at the intersection using a heatmap effect to obtain the wheel trajectory heatmap of the intersection. The analysis module transfers the wheel trajectory heatmap of the intersection to the map information for analysis based on the vehicle's latitude and longitude at each moment, determining whether the guide lines at the intersection are drawn correctly.

[0112] In this embodiment, the description of the data acquisition module can be found in the description of S110 in the above embodiment. The description of the relative spatiotemporal information acquisition module can be found in the description of S120 in the above embodiment. The description of the absolute spatiotemporal information acquisition module can be found in the description of S130 in the above embodiment. The description of the rut width acquisition module can be found in the description of S140 in the above embodiment. The description of the first wheel trajectory acquisition module can be found in the description of S150 in the above embodiment. The description of the wheel trajectory heatmap acquisition module can be found in the description of S160 in the above embodiment. The description of the analysis module can be found in the description of S170 in the above embodiment.

[0113] In one embodiment, the intersection traffic flow analysis system 100 based on radar-visual fusion further includes a wheel trajectory set acquisition module. This module categorizes the wheel trajectories of each vehicle based on the inbound / outbound directions and lane information in the map information, forming inbound wheel trajectory sets, outbound wheel trajectory sets, and lane wheel trajectory sets.

[0114] In this embodiment, the description of the wheel trajectory set acquisition module can be found in the description of S151 in the above embodiment.

[0115] In one embodiment, the wheel trajectory heatmap acquisition module includes an entrance wheel trajectory heatmap acquisition module, an exit wheel trajectory heatmap acquisition module, and a lane wheel trajectory heatmap acquisition module. The entrance wheel trajectory heatmap acquisition module is used to process each wheel trajectory in the entrance direction corresponding to the intersection using heatmap effects to obtain an entrance wheel trajectory heatmap. Alternatively, the exit wheel trajectory heatmap acquisition module is used to process each wheel trajectory in the exit direction corresponding to the intersection using heatmap effects to obtain an exit wheel trajectory heatmap. Alternatively, the lane wheel trajectory heatmap acquisition module is used to process each wheel trajectory in the lane wheel trajectory corresponding to the intersection using heatmap effects to obtain a lane wheel trajectory heatmap.

[0116] In this embodiment, the description of the inlet wheel trajectory heat map acquisition module can be found in the description of S161 in the above embodiment. The description of the outlet wheel trajectory heat map acquisition module can be found in the description of S162 in the above embodiment. The description of the lane wheel trajectory heat map acquisition module can be found in the description of S163 in the above embodiment.

[0117] In one embodiment, the analysis module includes a first analysis and judgment module, a second analysis and judgment module, and a third analysis and judgment module. The first analysis and judgment module is used to transfer the heat map of the approach wheel trajectory to map information for analysis based on the vehicle's latitude and longitude at each time, and to determine whether the guide lines for the approach direction of the intersection are drawn correctly. Alternatively, the second analysis and judgment module is used to transfer the heat map of the exit wheel trajectory to map information for analysis based on the vehicle's latitude and longitude at each time, and to determine whether the guide lines for the exit direction of the intersection are drawn correctly. Alternatively, the third analysis and judgment module is used to transfer the heat map of the lane wheel trajectory to map information for analysis based on the vehicle's latitude and longitude at each time, and to determine whether the guide lines for the lanes at the intersection are drawn correctly.

[0118] In this embodiment, the description of the first analysis and judgment module can be referred to the description of S171 in the above embodiment. The description of the second analysis and judgment module can be referred to the description of S172 in the above embodiment. The description of the third analysis and judgment module can be referred to the description of S173 in the above embodiment.

[0119] In one embodiment, this application provides a traffic flow analysis system for intersections based on radar-visual fusion. The system includes a data acquisition module, a relative spatiotemporal information acquisition module, an absolute spatiotemporal information acquisition module, a rut width acquisition module, a vehicle center point trajectory acquisition module, a second wheel trajectory acquisition module, a wheel trajectory heatmap acquisition module, and an analysis module. The data acquisition module acquires video image data and radar detection data. The relative spatiotemporal information acquisition module fuses the video image data and radar detection data to obtain the relative spatiotemporal information of each vehicle relative to the intersection.

[0120] The absolute spatiotemporal information acquisition module obtains map information of the intersection and, based on the map information and relative spatiotemporal information, acquires the absolute spatiotemporal information of all vehicles at the intersection. The rut width acquisition module obtains the actual rut width of each vehicle based on its model and width from the absolute spatiotemporal information. It then obtains the latitude and longitude of the actual rut width based on the vehicle's latitude and longitude from the absolute spatiotemporal information and projects this latitude and longitude onto the map information to obtain the simulated rut width for each vehicle. The vehicle center point trajectory acquisition module connects the vehicle's latitude and longitude from the absolute spatiotemporal information at each moment, using the vehicle as the center point, to obtain the trajectory of each vehicle's center point.

[0121] The second wheel trajectory acquisition module uses each center point in the vehicle's center point trajectory as a vertex and the trajectories of adjacent center points as edges to construct angle bisectors with a length equal to the width of the ruts, obtaining the ruts of each vehicle at each time moment. It then connects the endpoints of the ruts of each vehicle at each time moment to form the wheel trajectory of each vehicle. The wheel trajectory heatmap acquisition module processes the wheel trajectories of each vehicle at the intersection into a heatmap, obtaining the wheel trajectory heatmap of the intersection. The analysis module uses the vehicle's latitude and longitude at each time moment to transfer the wheel trajectory heatmap of the intersection to map information for analysis, determining whether the guide lines of the intersection are drawn correctly.

[0122] In this embodiment, the description of the data acquisition module can refer to the description of S110 or S210 in the above embodiments. The description of the relative spatiotemporal information acquisition module can refer to the description of S120 or S220 in the above embodiments. The description of the absolute spatiotemporal information acquisition module can refer to the description of S130 or S230 in the above embodiments. The description of the rut width acquisition module can refer to the description of S140 or S240 in the above embodiments. The description of the vehicle center point trajectory acquisition module can refer to the description of S250 in the above embodiments. The description of the second wheel trajectory acquisition module can refer to the description of S260 in the above embodiments. The description of the wheel trajectory heatmap acquisition module can refer to the description of S270 or S160 in the above embodiments. The description of the analysis module can refer to the description of S280 or S170 in the above embodiments.

[0123] In the various embodiments described above, the specific order or hierarchy of steps in the disclosed process is an example of an exemplary method. Based on design preferences, it should be understood that the specific order or hierarchy of steps in the process may be rearranged without departing from the scope of this disclosure. The appended method claims provide elements of various steps in an exemplary order and are not intended to limit the scope to a specific order or hierarchy.

[0124] Those skilled in the art will also understand that the various illustrative logical blocks, modules, and steps listed in the embodiments of this application can be implemented by electronic hardware, computer software, or a combination of both. To clearly demonstrate the interchangeability of hardware and software, the functions of the various illustrative components, modules, and steps described above have been generally described. Whether such functionality is implemented through hardware or software depends on the specific application and the overall system design requirements. Those skilled in the art can implement the described functions using various methods for each specific application, but such implementation should not be construed as exceeding the scope of protection of the embodiments of this application.

[0125] The various illustrative logic blocks or modules described in the embodiments of this application can be implemented or operate the described functions using a general-purpose processor, digital signal processor, application-specific integrated circuit (ASIC), field-programmable gate array or other programmable logic device, discrete gate or transistor logic, discrete hardware components, or any combination thereof. The general-purpose processor can be a microprocessor; alternatively, it can be any conventional processor, controller, microcontroller, or state machine. The processor can also be implemented using a combination of computing devices, such as a digital signal processor and a microprocessor, multiple microprocessors, one or more microprocessors combined with a digital signal processor core, or any other similar configuration.

[0126] The steps of the methods or algorithms described in the embodiments of this application can be directly embedded in hardware, a software module executed by a processor, or a combination of both. The software module can be stored in RAM, flash memory, ROM, EPROM, EEPROM, registers, hard disk, removable disk, CD-ROM, or any other form of storage medium in the art. Exemplarily, the storage medium can be connected to the processor so that the processor can read information from and write information to the storage medium. Optionally, the storage medium can also be integrated into the processor. The processor and storage medium can be housed in an ASIC, which can be housed in a user terminal. Optionally, the processor and storage medium can also be housed in different components of the user terminal.

[0127] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of this application. It should be understood that the above description is only a specific embodiment of this application and is not intended to limit the scope of protection of this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. A method for analyzing traffic flow lines at intersections based on radar-visual fusion, characterized in that, include: Acquire video image data and radar detection data; The video image data and the radar detection data are fused to obtain the relative spatiotemporal information of each vehicle relative to the intersection. Obtain map information of the intersection, and obtain absolute spatiotemporal information of all vehicles at the intersection based on the map information and the relative spatiotemporal information; Based on the vehicle model and vehicle width in the absolute spatiotemporal information of each vehicle, the actual rut width of each vehicle is obtained. Based on the actual rut width and the vehicle latitude and longitude in the absolute spatiotemporal information, the latitude and longitude of the actual rut width are obtained. The latitude and longitude of the actual rut width are then projected onto the map information to obtain the simulated rut width of each vehicle. Using the vehicle's latitude and longitude at each time as the center point, extend a line segment with the simulated rut width in a direction perpendicular to the vehicle's direction angle to obtain the ruts of each vehicle at each time. Connect the endpoints of the ruts of each vehicle at each time to form the wheel track of each vehicle. or Connect the vehicle's latitude and longitude coordinates in the absolute spatiotemporal information of each vehicle at each moment as the center point to obtain the vehicle center point trajectory of each vehicle. Using each center point in the vehicle's center point trajectory as a vertex and the trajectories of adjacent center points as edges, construct angle bisectors with a length equal to the width of the ruts to the inner and outer angles to obtain the ruts of each vehicle at each time moment, and connect the endpoints of the ruts of each vehicle at each time moment to form the wheel trajectory of each vehicle. Heatmap processing is performed on the wheel trajectories of each vehicle at the intersection to obtain a heatmap of the wheel trajectories at the intersection. Based on the vehicle's latitude and longitude at each time, the heat map of the wheel trajectory at the intersection is transferred to the map information for analysis to determine whether the guide lines of the intersection are drawn correctly.

2. The intersection guide line analysis method based on radar-visual fusion according to claim 1, characterized in that, Using the vehicle's latitude and longitude in the absolute spatiotemporal information at each moment as the center point, a line segment extending to the width of the simulated ruts is drawn perpendicular to the vehicle's direction angle to obtain the ruts of each vehicle at each moment. After connecting the endpoints of the ruts of each vehicle at each moment to form the wheel track of each vehicle, before performing heatmap processing on the wheel tracks of each vehicle at the intersection to obtain the wheel track heatmap of the intersection, the method further includes: Based on the inbound / outbound directions and lane information in the map information, the wheel trajectories of each vehicle are classified to form an inbound direction wheel trajectory set, an outbound direction wheel trajectory set, and a lane wheel trajectory set.

3. The intersection guide line analysis method based on radar-visual fusion according to claim 2, characterized in that, The step of processing the wheel trajectories of each vehicle at the intersection to obtain a heatmap of the wheel trajectories at the intersection includes: The wheel tracks corresponding to the approach directions at the intersection are collected and processed into a heat map effect for each wheel track to obtain an approach wheel track heat map; or The exit wheel tracks corresponding to the intersections are collected and processed into a heatmap effect for each wheel track to obtain an exit wheel track heatmap; or Heatmap effects are applied to each wheel trajectory in the lane corresponding to the intersection to obtain a lane wheel trajectory heatmap.

4. The intersection guide line analysis method based on radar-visual fusion according to claim 3, characterized in that, The step of transferring the wheel trajectory heatmap of the intersection to the map information for analysis based on the vehicle's latitude and longitude at each time moment, and determining whether the guide lines of the intersection are drawn correctly, includes: Based on the latitude and longitude of each vehicle at each time moment, the heat map of the approaching wheel trajectory is transferred to the map information for analysis to determine whether the guide lines for the approaching direction of the intersection are drawn correctly; or Based on the latitude and longitude of each vehicle at each time moment, the heat map of the exit wheel trajectory is transferred to the map information for analysis to determine whether the guide lines for the exit direction of the intersection are drawn correctly; or Based on the vehicle's latitude and longitude at each time, the heat map of the lane wheel trajectory is transferred to the map information for analysis to determine whether the lane guide lines at the intersection are drawn correctly.

5. A traffic flow analysis system for intersections based on radar-visual fusion, characterized in that, include: The data acquisition module is used to acquire video image data and radar detection data; The relative spatiotemporal information acquisition module is used to fuse the video image data with the radar detection data to obtain the relative spatiotemporal information of each vehicle relative to the intersection. An absolute spatiotemporal information acquisition module is used to acquire map information of the intersection and obtain absolute spatiotemporal information of all vehicles at the intersection based on the map information and the relative spatiotemporal information. The rut width acquisition module is used to obtain the actual rut width of each vehicle based on the vehicle model and vehicle width in the absolute spatiotemporal information of each vehicle, obtain the latitude and longitude of the actual rut width based on the actual rut width and the vehicle latitude and longitude in the absolute spatiotemporal information, and project the latitude and longitude of the actual rut width onto the map information to obtain the simulated rut width of each vehicle. The first wheel trajectory acquisition module is used to extend the simulated rut width of a line segment in a direction perpendicular to the vehicle's direction angle, with the vehicle's latitude and longitude as the center point, to obtain the ruts of each vehicle at each time, and connect the endpoints of the ruts of each vehicle at each time to form the wheel trajectory of each vehicle. or The vehicle center point trajectory acquisition module is used to connect the vehicle latitude and longitude of each vehicle in the absolute spatiotemporal information at each time moment with the vehicle latitude and longitude as the center point to obtain the vehicle center point trajectory of each vehicle. The second wheel trajectory acquisition module is used to construct angle bisectors with a length equal to the width of the ruts to the inner and outer angles, with each center point in the vehicle center point trajectory as a vertex and the trajectories of adjacent center points as edges, to obtain the ruts of each vehicle at each time moment, and connect the ruts of each vehicle at each time moment to form the wheel trajectory of each vehicle. The wheel trajectory heatmap acquisition module is used to process the wheel trajectory of each vehicle at the intersection to obtain a wheel trajectory heatmap of the intersection. The analysis module is used to transfer the wheel trajectory heatmap of the intersection to the map information for analysis based on the vehicle latitude and longitude of each vehicle at each time, and to determine whether the guide lines of the intersection are drawn correctly.

6. The intersection traffic flow analysis system based on radar-visual fusion according to claim 5, characterized in that, The intersection traffic flow analysis system based on radar-visual fusion also includes: The wheel trajectory set acquisition module is used to classify the wheel trajectory of each vehicle according to the entrance / exit direction and lane information in the map information, forming an entrance direction wheel trajectory set, an exit direction wheel trajectory set, and a lane wheel trajectory set.

7. The intersection traffic flow analysis system based on radar-visual fusion according to claim 6, characterized in that, The wheel trajectory heatmap acquisition module includes: An import wheel trajectory heatmap acquisition module is used to collect the import wheel trajectories corresponding to the intersection in the import direction, process each wheel trajectory to obtain an import wheel trajectory heatmap; or The exit wheel trajectory heatmap acquisition module is used to collect the wheel trajectories of the exit direction corresponding to the intersection, process each wheel trajectory to obtain an exit wheel trajectory heatmap; or The lane wheel trajectory heatmap acquisition module is used to process each wheel trajectory in the lane wheel trajectory set corresponding to the intersection with a heatmap effect to obtain a lane wheel trajectory heatmap.

8. The intersection traffic flow analysis system based on radar-visual fusion according to claim 7, characterized in that, The analysis module includes: The first analysis and judgment module is used to transfer the heat map of the approaching wheel trajectory to the map information for analysis based on the vehicle's latitude and longitude at each time, and to determine whether the guide line drawn in the approach direction of the intersection is correct; or The second analysis and judgment module is used to transfer the heat map of the exit wheel trajectory to the map information for analysis based on the vehicle latitude and longitude of each vehicle at each time, and to determine whether the guide line of the exit direction of the intersection is drawn correctly; or The third analysis and judgment module is used to transfer the heat map of the lane wheel trajectory to the map information for analysis based on the vehicle latitude and longitude of each vehicle at each time, and to determine whether the lane guide lines of the intersection are drawn correctly.

Images