Path navigation method and system based on real-time traffic flow and related equipment

Abstract

The invention relates to the technical field of traffic flow detection, in particular to a path navigation method and system based on real-time traffic flow and related equipment.The method comprises the steps that lane division is carried out based on lane features, the lane category corresponding to each lane in a key road section is determined, preprocessing is carried out based on video monitoring images, and the video monitoring images are obtained; the traffic flow detection area is selected from the video monitoring image, and the traffic flow detection area is used for centralized processing and analysis of parts really related to the traffic flow, so that the accuracy of traffic flow detection is improved. And carrying out traffic flow detection based on the lane category corresponding to each lane in the target road section and the traffic flow detection area in the video monitoring image, determining the target traffic flow of the target road section, and selecting a corresponding traffic flow detection method for different lane categories when the traffic flow detection is carried out. The accuracy of the target traffic flow and the traffic flow detection efficiency are improved. And finally, integrating the target traffic flow of each target road section in the key road sections to obtain the total traffic flow of the key road sections.

Description

Technical Field

[0001] The present application relates to the field of route navigation technology, and in particular to a route navigation method, system and related equipment based on real-time vehicle flow. Background Art

[0002] The commonly used navigation method at present is to search for routes based on the shortest time or shortest route in a static road network to provide navigation routes. However, the navigation routes obtained based on the static road network are not updated in real time according to the actual road conditions. Sometimes, when driving along the navigation route, you may encounter emergencies such as road maintenance closures, traffic control or traffic accidents, which brings great inconvenience to travel. Considering the impact of real-time road conditions on road traffic, people later mainly learned about road traffic information of various sections by listening to the radio and paying attention to the news, but this method has poor real-time performance, small coverage and is not intuitive.

[0003] With the development of relevant technologies, through some current road traffic software, people can directly obtain the navigation path obtained by these software according to the current real-time traffic conditions of the path. This method can help people effectively avoid problem roads that are currently closed for maintenance or traffic control, or currently have pending traffic accidents; however, real-time traffic conditions can only represent whether the road can pass vehicles normally and the current congestion situation, and the road congestion situation will change in real time. Therefore, the method of selecting a path using traffic software still has the following situation: when driving along the navigation route to a certain road on the way, the road section that was not congested when previously planned is now lined with a long queue, which greatly prolongs people's travel time, reduces the accuracy of path navigation, and affects people's traffic experience. Summary of the invention

[0004] In order to improve the problems of low path navigation accuracy and poor user traffic experience when navigating according to current real-time traffic conditions when road congestion changes, the present application provides a path navigation method, system and related equipment based on real-time traffic flow.

[0005] In the first aspect, the present application provides a route navigation method based on real-time traffic flow, which adopts the following technical solution: A route navigation method based on real-time traffic flow, comprising: Acquire lane features of a target road section in the key road section, perform lane division based on the lane features, and determine a lane category corresponding to each lane in the key road section, wherein the lane categories include: a one-way straight lane and a multi-way lane, and the target road section is any branch road in the key road section; Acquire a video surveillance image of the target road section, perform preprocessing based on the video surveillance image, and select a vehicle flow detection area in the video surveillance image; Performing vehicle flow detection based on the lane category corresponding to each lane in the target road section and the vehicle flow detection area in the video surveillance image to determine the target vehicle flow of the target road section; The target traffic flow of each target section in the key section is integrated to obtain the total traffic flow of the key section.

[0006] By adopting the above technical solution, lane division is performed based on lane features to determine the lane category corresponding to each lane in the key section. At the same time, preprocessing is performed based on the video surveillance image, and the traffic flow detection area is framed in the video surveillance image. The framed traffic flow detection area is used to centrally process and analyze the part that is truly related to the traffic flow, which not only reduces false detection and missed detection, improves the accuracy of traffic flow detection, but also saves the computing resources and storage resources of electronic equipment. Then, based on the lane category corresponding to each lane in the target section and the traffic flow detection area in the video surveillance image, the traffic flow detection is performed to determine the target traffic flow of the target section. When performing traffic flow detection, the appropriate traffic flow detection method is selected for different lane categories to improve the accuracy of the target traffic flow and the efficiency of traffic flow detection. Finally, the target traffic flow of each target section in the key section is integrated to obtain the total traffic flow of the key section. In order to improve the accuracy of the traffic flow of the key section, traffic flow detection is performed for each target section in the key section to obtain the target traffic flow of the target section, and then the target traffic flow is summarized to obtain the total traffic flow of the key section.

[0007] In a preferred example, the present application may be further configured as follows: performing vehicle flow detection based on the lane category corresponding to each lane in the target road section and the vehicle flow detection area in the video surveillance image to determine the target vehicle flow of the target road section includes: Perform vehicle identification based on the vehicle flow detection area in the video surveillance image, and mark a vehicle identification frame corresponding to each vehicle; Based on the vehicle flow detection area in the video surveillance image, tracking each vehicle identification frame corresponding to the multi-directional lane to determine the first straight vehicle flow and turning vehicle flow in the multi-directional lane; Based on the vehicle flow detection area in the video surveillance image, tracking each vehicle identification frame corresponding to the one-way through lane to determine a second through-traffic flow in the one-way through lane; The through-traffic flow rate of the target road section is determined based on the sum of the first through-traffic flow rate and the second through-traffic flow rate, wherein the target traffic flow rate includes: the through-traffic flow rate and the turning traffic flow rate.

[0008] In a preferred example, the present application may be further configured as follows: based on the vehicle flow detection area in the video surveillance image, each vehicle identification frame corresponding to the multi-directional lane is tracked to determine the first straight vehicle flow and the turning vehicle flow in the multi-directional lane, including: Based on the vehicle flow detection area in the video surveillance image, tracking each of the vehicle identification frames corresponding to the multi-directional lanes, and determining a set of rotation direction angles corresponding to each of the vehicle identification frames; The first through-traffic flow and the turning vehicle flow in the multi-directional lane are determined based on the rotation direction angle set corresponding to each vehicle identification frame.

[0009] In a preferred example, the present application may be further configured as follows: after performing vehicle flow detection based on the lane category corresponding to each lane in the target road section and the vehicle flow detection area in the video surveillance image, and determining the target vehicle flow of the target road section, the configuration further includes: Using a traffic flow prediction model to predict the traffic flow of the target road section, to obtain a predicted traffic flow corresponding to the target road section; Performing a traffic flow comparison analysis based on the predicted traffic flow corresponding to the target road section and the target traffic flow, and determining a comparison analysis result; When the comparison and analysis result is normal, the target traffic flow of the target road section is used as a training data set, and the traffic flow prediction model is retrained using the training data set to obtain an updated traffic flow prediction model.

[0010] In a preferred example, the present application may be further configured as follows: after performing a traffic flow comparison analysis based on the predicted traffic flow corresponding to the target road section and the target traffic flow, and determining the comparison analysis result, the following may be further configured: When the comparison and analysis result is abnormal, the vehicle flow of the target road section is re-detected using a positioning linkage method to obtain a precise target vehicle flow, wherein the positioning linkage method is to accurately detect the vehicle flow through the vehicle's positioning system.

[0011] In a preferred example, the present application may be further configured as follows: a method for collecting lane features of the target road section includes: When the lane feature collection cycle is reached, a road image of the target road section during a traffic trough period is collected, and feature extraction is performed based on the road image to obtain the current lane feature of the target road section; The stored lane features are obtained, and feature comparison is performed based on the current lane features and the stored lane features. When there are differences in features, the stored lane features are updated using the current lane features and recorded as the lane features of the target road section.

[0012] In a second aspect, the present application provides a vehicle flow detection device, which adopts the following technical solution: A vehicle flow detection device, comprising: A lane division module is used to obtain lane features of a target road section in a key road section, perform lane division based on the lane features, and determine a lane category corresponding to each lane in the key road section, wherein the lane categories include: a one-way straight lane and a multi-way lane, and the target road section is any branch road in the key road section; A region division module is used to obtain a video surveillance image of the target road section, perform preprocessing based on the video surveillance image, and select a vehicle flow detection area in the video surveillance image; A traffic flow detection module, configured to perform traffic flow detection based on the lane category corresponding to each lane in the target road section and the traffic flow detection area in the video surveillance image, and determine a target traffic flow of the target road section; The traffic aggregation module is used to integrate the target traffic volume of each target section in the key section to obtain the total traffic volume of the key section.

[0013] In a third aspect, the present application provides an electronic device, which adopts the following technical solution: An electronic device includes at least one processor; a memory; and at least one application, wherein the at least one application is stored in the memory and is configured to be executed by the at least one processor, and the at least one application is configured to: execute the above-mentioned path navigation method based on real-time traffic flow.

[0014] In a fourth aspect, the present application provides a computer-readable storage medium, which adopts the following technical solution: A computer-readable storage medium stores a computer program, which, when executed in a computer, causes the computer to execute the above-mentioned path navigation method based on real-time vehicle flow.

[0015] In summary, the present application includes at least one of the following beneficial technical effects: 1. Based on lane features, lane division is performed to determine the lane category corresponding to each lane in the key section. At the same time, preprocessing is performed based on the video surveillance image, and the traffic flow detection area is framed in the video surveillance image. The framed traffic flow detection area is used to centrally process and analyze the parts that are truly related to the traffic flow, which not only reduces false detection and missed detection, improves the accuracy of traffic flow detection, but also saves the computing resources and storage resources of electronic equipment. Then, based on the lane category corresponding to each lane in the target section and the traffic flow detection area in the video surveillance image, traffic flow detection is performed to determine the target traffic flow of the target section. When performing traffic flow detection, the appropriate traffic flow detection method is selected for different lane categories to improve the accuracy of the target traffic flow and the efficiency of traffic flow detection. Finally, the target traffic flow of each target section in the key section is integrated to obtain the total traffic flow of the key section. In order to improve the accuracy of the traffic flow of the key section, traffic flow detection is performed on each target section in the key section to obtain the target traffic flow of the target section. Then, the target traffic flow is summarized to obtain the total traffic flow of the key section; 2. Perform vehicle identification based on the traffic flow detection area in the video surveillance image and mark the vehicle identification frame corresponding to each vehicle. Then, based on the traffic flow detection area in the video surveillance image, track each vehicle identification frame corresponding to the multi-directional lane to determine the first straight traffic flow and turning traffic flow in the multi-directional lane. At the same time, based on the traffic flow detection area in the video surveillance image, track each vehicle identification frame corresponding to the one-way straight lane to determine the second straight traffic flow in the one-way straight lane. Finally, the straight traffic flow of the target section is determined by summing the first straight traffic flow and the second straight traffic flow. When performing traffic flow detection, select appropriate traffic flow detection methods for different lane categories to improve the accuracy of the target traffic flow and the efficiency of traffic flow detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 It is a structural schematic diagram of a vehicle flow detection device according to an embodiment of the present application.

[0017] Figure 2 It is a flowchart of a path navigation method based on real-time traffic flow according to another embodiment of the present application. DETAILED DESCRIPTION

[0018] In order to make the objectives, technical solutions and advantages of the present application clearer, the implementation methods of the present application will be further described in detail below with reference to the accompanying drawings.

[0019] The following is a further detailed description of an embodiment of a path navigation method, system and related equipment based on real-time vehicle flow in the present application in conjunction with the drawings of the specification.

[0020] An embodiment of the present application discloses a route navigation method based on real-time vehicle flow.

[0021] Reference Figure 1 , a path navigation method based on real-time traffic flow includes: The embodiment of the present application provides a route navigation method based on real-time traffic flow, which is executed by an electronic device, which can be a server or a terminal device, wherein the server can be an independent physical server, or a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides cloud computing services. The terminal device can be a smart phone, a tablet computer, a laptop computer, a desktop computer, etc., but is not limited to this. The terminal device and the server can be directly or indirectly connected via wired or wireless communication, which is not limited in the embodiment of the present application. Figure 1 As shown, the method includes step S101, step S102, step S103 and step S104, wherein: Step S101: Acquire lane features of a target section in a critical section, perform lane division based on the lane features, and determine the lane category corresponding to each lane in the critical section, wherein the lane categories include: one-way straight lanes and multi-way lanes, and the target section is any branch road in the critical section.

[0022] For the embodiment of the present application, the key road section is formed by the intersection of multiple branches, and each branch has vehicles entering and exiting. In order to improve the accuracy of the traffic flow of the key road section and reduce the calculation amount of the subsequent traffic flow detection by the electronic equipment, traffic flow detection is performed on each target section in the key road section to obtain the target traffic flow of the target section, and then the target traffic flow is summarized to obtain the total traffic flow of the key section.

[0023] Specifically, the lane features of the target road section in the key road section are obtained, and the lane features include but are not limited to: lane line position, lane width, lane marking, wherein the lane marking refers to the marking on the target road section, which is used to indicate the direction, function, restriction, etc. of the lane. Then, lane division is performed based on the lane marking in the lane features of the target road section, and the lane category corresponding to each lane in the key road section is determined. The lane categories include: one-way straight lane and multi-directional lane, wherein the one-way straight lane is a lane in which vehicles can only drive straight, and the multi-directional lane is a lane in which vehicles can not only drive straight but also turn. For vehicles on the one-way straight lane, the driving direction is determined; for vehicles on the multi-directional lane, the driving direction is uncertain, that is, it is uncertain whether it is straight or turning. Therefore, the two types of lanes are preferably divided, so that the appropriate traffic flow detection method is selected for different lane categories in the subsequent process, so as to ensure that the traffic flow on each type of lane is accurately and efficiently detected.

[0024] Step S102: Obtain a video surveillance image of the target road section, perform preprocessing based on the video surveillance image, and select a vehicle flow detection area in the video surveillance image.

[0025] For the embodiment of the present application, in order to improve the accuracy of vehicle flow detection, an image acquisition device is set for each target road section, and the image acquisition device is used to collect panoramic video images passing through the key road section on the target road section. The area covered by the panoramic video image collected by the image acquisition device is relatively large. If the vehicle flow detection is directly performed on the panoramic video image, it will not only consume high computing resources, but also introduce more irrelevant interference. Therefore, in the embodiment of the present application, pre-processing is performed based on the video surveillance image, and the vehicle flow detection area is framed in the video surveillance image, wherein the vehicle flow detection area is a part of the video surveillance image close to the key road section. For the size and position of the vehicle flow detection area, the user can set it according to his own needs, and the embodiment of the present application is no longer limited. The framed vehicle flow detection area is used to centrally process and analyze the parts that are truly related to the vehicle flow, and remove the areas that are irrelevant to the vehicle flow detection, such as the parts far away from the key road section. In this way, not only false detection and missed detection are reduced, the accuracy of vehicle flow detection is improved, but also the computing resources and storage resources of the electronic device are saved.

[0026] Step S103: performing vehicle flow detection based on the lane category corresponding to each lane in the target road section and the vehicle flow detection area in the video surveillance image to determine the target vehicle flow of the target road section; For the embodiments of the present application, for vehicles on a one-way straight lane, the driving direction is certain; for vehicles on a multi-directional lane, the driving direction is uncertain, that is, it is uncertain whether it is going straight or turning. Therefore, when performing vehicle flow detection, a suitable vehicle flow detection method is selected for different lane categories to improve the accuracy of the target vehicle flow and the efficiency of vehicle flow detection.

[0027] Specifically, vehicle identification is performed based on the traffic flow detection area in the video surveillance image, and the vehicle identification frame corresponding to each vehicle is marked. For multi-directional lanes, based on the traffic flow detection area in the video surveillance image, each vehicle identification frame corresponding to the multi-directional lane is tracked to determine the first straight traffic flow and the turning traffic flow in the multi-directional lane, wherein, when tracking the vehicles in the multi-directional lane, the vehicle trajectory analysis method is used to count the traffic flows in different directions respectively. For one-way straight lanes, based on the traffic flow detection area in the video surveillance image, each vehicle identification frame corresponding to the one-way straight lane is tracked to determine the second straight traffic flow in the one-way straight lane, wherein, when tracking the vehicles in the one-way straight lane, the second traffic flow is determined by counting the number of vehicles passing through per unit time. Finally, the first straight traffic flow and the second straight traffic flow are summed to determine the straight traffic flow of the target section, wherein the target traffic flow includes: straight traffic flow and turning traffic flow, and the straight traffic flow is the total traffic flow on all one-way straight lanes in the target section.

[0028] Step S104: The target traffic flow of each target section in the key section is integrated to obtain the total traffic flow of the key section.

[0029] For the embodiments of the present application, there are many ways to synthesize the target traffic flow of each target section, which are no longer limited in the embodiments of the present application, and users can set them by themselves. In one feasible method, the time period is divided based on the time period information of the target traffic flow, and the branch time period traffic flow arranged in time sequence is obtained. Then, the format of the branch time period traffic flow with the same time period information in each target section is unified to determine at least one traffic direction corresponding to the branch time period traffic flow and the traffic flow corresponding to the traffic direction. For any time period, the traffic direction corresponding to the branch time period traffic flow and the traffic flow corresponding to the traffic direction based on all target sections are summarized to obtain the total traffic flow of the key section, wherein the total traffic flow includes: the total traffic flow of the time period corresponding to each of the multiple time periods.

[0030] It can be seen that in the embodiment of the present application, lane division is performed based on lane features to determine the lane category corresponding to each lane in the key section. At the same time, preprocessing is performed based on the video surveillance image, and the traffic flow detection area is framed in the video surveillance image. The framed traffic flow detection area is used to centrally process and analyze the part that is truly related to the traffic flow, which not only reduces false detection and missed detection, improves the accuracy of traffic flow detection, but also saves the computing resources and storage resources of the electronic equipment. Then, based on the lane category corresponding to each lane in the target section and the traffic flow detection area in the video surveillance image, the traffic flow detection is performed to determine the target traffic flow of the target section. When performing traffic flow detection, the corresponding traffic flow detection method is selected for different lane categories to improve the accuracy of the target traffic flow and the efficiency of traffic flow detection. Finally, the target traffic flow of each target section in the key section is integrated to obtain the total traffic flow of the key section. In order to improve the accuracy of the traffic flow of the key section, the traffic flow detection is performed for each target section in the key section to obtain the target traffic flow of the target section, and then the target traffic flow is summarized to obtain the total traffic flow of the key section.

[0031] Further, in order to improve the accuracy of the target traffic flow and the efficiency of traffic flow detection, in the embodiment of the present application, traffic flow detection is performed based on the lane category corresponding to each lane in the target road section and the traffic flow detection area in the video surveillance image to determine the target traffic flow of the target road section, including: Steps S1031 to S1034, wherein: Step S1031: Vehicle identification is performed based on the vehicle flow detection area in the video surveillance image, and a vehicle identification frame corresponding to each vehicle is marked.

[0032] For the embodiment of the present application, computer vision technology is used for vehicle recognition, that is, a target detection algorithm is used to perform vehicle recognition on any frame image of the traffic flow detection area. The target detection algorithm includes but is not limited to: YOLO, SSD, Faster R-CNN, etc., to identify all vehicles in the traffic flow detection area, and mark each vehicle in the traffic flow detection area to obtain a vehicle identification box corresponding to each vehicle, wherein the vehicle identification box needs to select the entire vehicle and is used to mark the position and size of the vehicle.

[0033] Step S1032: Based on the vehicle flow detection area in the video surveillance image, the vehicle identification frame corresponding to the multi-directional lane is tracked to determine the first straight vehicle flow and the turning vehicle flow in the multi-directional lane.

[0034] For the embodiment of the present application, based on the traffic flow detection area in the video surveillance image, each vehicle identification frame corresponding to the multi-directional lane is tracked, that is, each vehicle identification in the multi-directional lane is tracked using a vehicle tracking algorithm, wherein the vehicle tracking algorithm predicts and updates the vehicle identification frame of each vehicle based on the position and appearance features of the vehicle between consecutive frames. There are many ways to determine the straight traffic flow and turning traffic flow in the multi-directional lane by vehicle tracking, and the embodiment of the present application is no longer limited. In one achievable method, based on the traffic flow detection area in the video surveillance image, each vehicle identification frame corresponding to the multi-directional lane is tracked, and the rotation direction angle set corresponding to each vehicle identification frame is determined, wherein the rotation direction angle set is composed of the rotation direction angles corresponding to the vehicle identification frame in different video frames, and the rotation direction angle is the angle formed between the vehicle's driving direction and the lane direction. Then, based on the rotation direction angle set corresponding to each vehicle identification frame, the first straight traffic flow and turning traffic flow in the multi-directional lane are determined. In another feasible manner, a target tracking algorithm, such as SORT, DeepSORT, etc., is used to track the vehicle identification frames in the multi-directional lanes to establish the vehicle's motion trajectory. Then, the vehicle's driving direction is analyzed based on the motion trajectory of each vehicle. Based on the analyzed vehicle driving direction, the first straight vehicle flow and turning vehicle flow in the multi-directional lanes are counted.

[0035] Step S1033: Based on the vehicle flow detection area in the video surveillance image, each vehicle identification frame corresponding to the one-way through lane is tracked to determine the second through-traffic flow in the one-way through lane.

[0036] For the embodiment of the present application, since the vehicles in the one-way straight lane can only go straight and cannot go in other directions, it is not necessary to analyze the driving direction of the vehicles in the one-way straight lane, which saves the computing resources of the electronic device. Therefore, based on the vehicle flow detection area in the video surveillance image, when tracking each vehicle identification frame corresponding to the one-way straight lane, only the target tracking algorithm is used to track each vehicle identification frame corresponding to the one-way straight lane. When it is detected that the vehicle identification frame of the vehicle disappears in the vehicle flow detection area, the vehicle flow counter is controlled to increase by one to obtain the second straight vehicle flow in the one-way straight lane.

[0037] Step S1034: Determine the through traffic flow of the target road section based on the sum of the first through traffic flow and the second through traffic flow, wherein the target traffic flow includes: through traffic flow and turning traffic flow.

[0038] For the embodiment of the present application, the straight traffic flow is the sum of the first straight traffic flow and the second straight traffic flow. At the same time, the straight traffic flow is the total traffic flow on all one-way straight lanes in the target section. The turning traffic flow is the non-straight traffic flow in the target section. The turning traffic flow includes but is not limited to: left-turning traffic flow, right-turning traffic flow, and U-turning traffic flow.

[0039] It can be seen that in the embodiment of the present application, vehicle identification is performed based on the traffic flow detection area in the video surveillance image, and the vehicle identification frame corresponding to each vehicle is marked. Then, based on the traffic flow detection area in the video surveillance image, each vehicle identification frame corresponding to the multi-directional lane is tracked to determine the first straight traffic flow and the turning traffic flow in the multi-directional lane. At the same time, based on the traffic flow detection area in the video surveillance image, each vehicle identification frame corresponding to the one-way straight lane is tracked to determine the second straight traffic flow in the one-way straight lane. Finally, the straight traffic flow of the target section is determined by summing the first straight traffic flow and the second straight traffic flow. When performing traffic flow detection, a suitable traffic flow detection method is selected for different lane categories to improve the accuracy of the target traffic flow and the efficiency of traffic flow detection.

[0040] Furthermore, in order to improve the accuracy of the traffic flow in the multi-directional lanes, in the embodiment of the present application, based on the traffic flow detection area in the video surveillance image, each vehicle identification frame corresponding to the multi-directional lane is tracked to determine the first straight traffic flow and the turning traffic flow in the multi-directional lanes, including: Based on the vehicle flow detection area in the video surveillance image, each vehicle identification frame corresponding to the multi-directional lane is tracked to determine the rotation direction angle set corresponding to each vehicle identification frame; Based on the rotation direction angle set corresponding to each vehicle identification frame, the first straight vehicle flow and the turning vehicle flow in the multi-directional lane are determined.

[0041] For the embodiment of the present application, since the video surveillance image is composed of continuous multi-frame images, the traffic flow detection area is also composed of continuous multi-frame images. Therefore, based on the continuous multi-frame images of the detection area, the target vehicle identification frame corresponding to the multi-directional lane is tracked to obtain the rotation direction angle of the target vehicle identification frame in each frame image, and the rotation direction angles in each frame image arranged in time sequence together constitute the rotation direction angle set corresponding to the target vehicle identification frame, wherein the target vehicle identification frame is any vehicle identification frame. There are many ways to determine the rotation direction angle of the target vehicle identification frame. In one achievable method, the rotation direction angle is calculated according to the position change of the target vehicle identification frame between continuous frame images, that is, the motion trajectory of the target vehicle identification frame is fitted using the least squares method, and then the rotation direction angle is calculated according to the change of the trajectory. Of course, the rotation direction angle can also be estimated more accurately based on the method of the three-dimensional rigid body motion model. In another achievable method, for any frame image, the angle between the lane line and the target vehicle identification frame is calculated using computer vision technology, and the angle between the two is recorded as the rotation direction angle corresponding to the frame image.

[0042] Due to the specificity of the target section in the key section, the actual road conditions of the target section determine the steering angle of the turning vehicle. For example, the turning radius and layout of the road will determine the angle change of the vehicle during the turning process, and the width of the lane will affect the direction angle. Therefore, the electronic device will pre-store the steering direction angle range corresponding to the target section, and the size of the steering direction angle range is obtained by technicians in this field based on a large number of experimental operations. Therefore, based on the steering direction angle range and the rotation direction angle set corresponding to each vehicle identification frame, the first straight traffic flow and the turning traffic flow in the multi-directional lane are determined. Specifically, when the number of the rotation direction angle set that falls in the steering direction angle range is greater than the preset number, the vehicle identification frame is determined to be turning; otherwise, the vehicle identification frame is determined to be straight, wherein the size of the preset number is determined by technicians in this field based on a large number of experimental operations. The first straight traffic flow and the turning traffic flow in the multi-directional lane are obtained by combining the driving direction of each vehicle identification frame.

[0043] It can be seen that in the embodiment of the present application, based on the traffic flow detection area in the video surveillance image, each vehicle identification frame corresponding to the multi-directional lane is tracked, and the rotation direction angle set corresponding to each vehicle identification frame is determined. Then, based on the rotation direction angle set corresponding to each vehicle identification frame, the first straight traffic flow and turning traffic flow in the multi-directional lane are determined. When determining the first straight traffic flow and turning traffic flow in the multi-directional lane by the rotation direction angle of the vehicle, the specificity of the target section in the key section is comprehensively considered, thereby improving the accuracy of the traffic flow in the multi-directional lane.

[0044] Furthermore, in order to enable the updated vehicle flow prediction model to better respond to changes in traffic conditions and adjust the prediction in a timely manner, in an embodiment of the present application, vehicle flow detection is performed based on the lane category corresponding to each lane in the target road section and the vehicle flow detection area in the video surveillance image, and after determining the target vehicle flow of the target road section, the following is also included: The traffic flow prediction model is used to predict the traffic flow of the target road section to obtain the predicted traffic flow corresponding to the target road section; Perform traffic flow comparison analysis based on the predicted traffic flow corresponding to the target road section and the target traffic flow, and determine the comparison analysis result; When the comparison and analysis result is normal, the target traffic flow of the target road section is used as a training data set, and the traffic flow prediction model is retrained using the training data set to obtain an updated traffic flow prediction model.

[0045] For the embodiments of the present application, due to various special situations that often occur during road driving, errors in traffic flow detection due to special reasons such as weather, etc., all of which may cause the traffic flow statistics results of key sections to be inconsistent with the actual situation. Therefore, after determining the target traffic flow, a traffic flow comparison analysis is performed based on the target traffic flow and the predicted traffic flow. When there is a large difference, the target traffic flow corresponding to the target section is adjusted so that the target traffic flow can more accurately represent the traffic conditions of the target section.

[0046] The traffic flow prediction model is obtained by training the convolutional neural network based on the historical traffic flow training samples of the target section. Therefore, when the relevant information of the target section is input into the traffic flow prediction model, the traffic flow prediction model automatically predicts the traffic flow and obtains the predicted traffic flow corresponding to the target section. That is, when the predicted traffic flow is obtained by the traffic flow prediction model, the unique characteristics and influencing factors of the target section are taken into account, so that the predicted traffic flow is more accurate and can better reflect the actual situation of the target section. Then, the traffic flow comparison analysis is performed based on the predicted traffic flow corresponding to the target section and the target traffic flow. When the difference between the predicted traffic flow and the target traffic flow is greater than the traffic flow threshold, the comparison analysis result is determined to be abnormal; otherwise, the comparison analysis result is determined to be normal, wherein the traffic flow threshold is a reasonable traffic flow difference value that can exist between the predicted situation and the actual situation. When the comparison analysis result is normal, it indicates that the traffic flow detection determines that the target traffic flow is consistent with the usual situation of the target section, then the target traffic flow of the branch road is used as a training data set, and the traffic flow prediction model is retrained using the training data set to obtain an updated traffic flow prediction model. Since traffic conditions are changing dynamically, incorporating the latest target traffic flow into the training data set can capture the latest changes in traffic conditions, so that the updated traffic flow prediction model can better respond to changes in traffic conditions and adjust the prediction in a timely manner.

[0047] It can be seen that in the embodiments of the present application, due to various special circumstances such as the frequent presence of various emergencies during road driving and errors in traffic flow detection due to special reasons such as weather, the traffic flow statistics of key sections will be inconsistent with the actual situation. Therefore, the traffic flow prediction model is used to predict the traffic flow of the target section to obtain the predicted traffic flow corresponding to the target section. At the same time, a traffic flow comparison analysis is performed based on the predicted traffic flow corresponding to the target section and the target traffic flow to determine the comparison analysis result. When the comparison analysis result is normal, the target traffic flow of the target section is used as a training data set, and the traffic flow prediction model is retrained using the training data set to obtain an updated traffic flow prediction model. Including the latest target traffic flow in the training data set can capture the latest changes in traffic conditions, so that the updated traffic flow prediction model can better respond to changes in traffic conditions and adjust the prediction in a timely manner.

[0048] Furthermore, in order to improve the accuracy and reliability of the traffic flow data, in the embodiment of the present application, a traffic flow comparison analysis is performed based on the predicted traffic flow corresponding to the target road section and the target traffic flow. After the comparison analysis result is determined, the following is further included: When the comparison and analysis result is abnormal, the positioning linkage method is used to re-detect the traffic flow of the target section to obtain the precise target traffic flow. The positioning linkage method is to accurately detect the traffic flow through the vehicle's positioning system.

[0049] For the embodiments of the present application, when the comparison and analysis result is abnormal, it indicates that the target traffic flow determined by the traffic flow detection is very different from the usual situation of the target road section. There are many reasons for this situation, such as data collection errors, special events affecting the normal driving of vehicles, severe weather conditions, changes in road infrastructure, etc. In order to improve the accuracy and reliability of traffic flow detection, when the target traffic flow determined by the traffic flow detection is very different from the usual situation of the target road section, the positioning linkage method is used as a supplementary means of traffic flow detection, and the precise target traffic flow corresponding to the target road section is obtained, which greatly improves the accuracy and reliability of traffic flow data.

[0050] Specifically, when the comparison and analysis result is abnormal, the positioning data of vehicles traveling on the target road section is collected through the interface with the vehicle positioning system (such as GPS), wherein the positioning data includes but is not limited to: vehicle position, vehicle speed, driving direction and vehicle ID. Then, the collected positioning data is preprocessed to obtain preprocessed positioning data, wherein the preprocessing includes but is not limited to: data cleaning, data filtering, data conversion and data analysis. Furthermore, based on the preprocessed positioning data, vehicle tracking is performed to determine the driving trajectory of each vehicle, and traffic statistics are performed based on the driving trajectories corresponding to all vehicles in the target road section to obtain accurate target traffic flow, wherein the accurate target traffic flow includes: directional traffic flow corresponding to each driving direction in the target road section.

[0051] It can be seen that in the embodiment of the present application, in order to improve the accuracy and reliability of vehicle flow detection, when the target vehicle flow determined by the vehicle flow detection is very different from the normal situation of the target road section, the positioning linkage method is used as a supplementary means of vehicle flow detection, and the precise target vehicle flow corresponding to the target road section is obtained, which greatly improves the accuracy and reliability of the vehicle flow data.

[0052] Furthermore, in order to improve the accuracy and automation of lane feature extraction and provide reliable data support for subsequent traffic flow analysis and application, in an embodiment of the present application, a method for collecting lane features of a target road section includes: When the lane feature collection cycle is reached, the road image of the target road section during the traffic trough period is collected, and feature extraction is performed based on the road image to obtain the current lane feature of the target road section; Obtain the stored lane features, perform feature comparison based on the current lane features and the stored lane features, and when there are differences in the features, use the current lane features to update the stored lane features and record them as the lane features of the target section.

[0053] For the embodiments of the present application, with the continuous construction of roads, the lane distribution of the target section in the key section may change. In order to ensure that the lane characteristics can accurately represent the current situation of the target section, the lane characteristics will be re-collected after each lane characteristic collection cycle to ensure the accuracy and timeliness of the lane characteristics of the target section.

[0054] Specifically, when the lane feature collection cycle is reached, the road image of the target road section during the traffic trough period is collected. There are many ways to collect road images, such as using a camera installed at the target road section to shoot, and using a drone to collect road images. At the same time, the selection of the traffic trough period can ensure that there are fewer vehicles on the road, reduce interference factors, and facilitate accurate extraction of lane features. Then, feature extraction is performed based on the road image to obtain the current lane feature of the target road section, that is, feature extraction is performed using computer vision and image processing technology to extract lane line information, lane width, lane markings, etc., and the lane line information, lane width, and lane markings of the target road section are recorded as current lane features. Then, the stored lane features pre-stored in the electronic device are obtained. The stored lane features are the lane features of the target road section previously saved, and feature comparison is performed based on the current lane features and the stored lane features to determine whether there are differences between the two. When there are differences in the features, the current lane features are used to update the stored lane features and are recorded as the lane features of the target road section for use in subsequent traffic flow detection. The feature extraction method based on image processing and computer vision technology can improve the accuracy and automation of lane feature extraction, and provide reliable data support for subsequent traffic flow analysis and applications.

[0055] It can be seen that in the embodiment of the present application, when the lane feature collection cycle is reached, the road image of the target section during the traffic trough period is collected, and feature extraction is performed based on the road image to obtain the current lane feature of the target section. The lane feature will be recollected after each lane feature collection cycle to ensure the accuracy and timeliness of the lane feature of the target section. Then, feature comparison is performed based on the current lane feature and the stored lane feature. When there is a difference in the feature, the current lane feature is used to update the stored lane feature and recorded as the lane feature of the target section. The feature extraction method based on image processing and computer vision technology can improve the accuracy and automation of lane feature extraction and provide reliable data support for subsequent traffic flow analysis and application.

[0056] Reference Figure 2 The present application also discloses a vehicle flow detection device, which specifically includes: A lane division module is used to obtain lane features of a target road section in a key road section, divide lanes based on the lane features, and determine the lane category corresponding to each lane in the key road section, wherein the lane categories include: one-way straight lanes and multi-way lanes, and the target road section is any branch road in the key road section; The area division module is used to obtain the video surveillance image of the target road section, perform preprocessing based on the video surveillance image, and select the vehicle flow detection area in the video surveillance image; A traffic flow detection module, used to detect traffic flow based on the lane category corresponding to each lane in the target road section and the traffic flow detection area in the video surveillance image, and determine the target traffic flow of the target road section; The traffic aggregation module is used to integrate the target traffic flow of each target section in the key section to obtain the total traffic flow of the key section.

[0057] For the embodiment of the present application, lane division is performed based on lane features to determine the lane category corresponding to each lane in the key section. At the same time, preprocessing is performed based on the video surveillance image, and the traffic flow detection area is framed in the video surveillance image. The framed traffic flow detection area is used to centrally process and analyze the part that is truly related to the traffic flow, which not only reduces false detection and missed detection, improves the accuracy of traffic flow detection, but also saves the computing resources and storage resources of the electronic equipment. Then, based on the lane category corresponding to each lane in the target section and the traffic flow detection area in the video surveillance image, the traffic flow detection is performed to determine the target traffic flow of the target section. When performing the traffic flow detection, the corresponding traffic flow detection method is selected for different lane categories to improve the accuracy of the target traffic flow and the efficiency of traffic flow detection. Finally, the target traffic flow of each target section in the key section is integrated to obtain the total traffic flow of the key section. In order to improve the accuracy of the traffic flow of the key section, the traffic flow detection is performed for each target section in the key section to obtain the target traffic flow of the target section, and then the target traffic flow is summarized to obtain the total traffic flow of the key section.

[0058] In a possible implementation of the embodiment of the present application, when the traffic flow detection module performs traffic flow detection based on the lane category corresponding to each lane in the target road section and the traffic flow detection area in the video surveillance image to determine the target traffic flow of the target road section, it is used to: Vehicle identification is performed based on the vehicle flow detection area in the video surveillance image, and a vehicle identification frame corresponding to each vehicle is marked; Based on the vehicle flow detection area in the video surveillance image, each vehicle identification frame corresponding to the multi-directional lane is tracked to determine the first straight vehicle flow and turning vehicle flow in the multi-directional lane; Based on the vehicle flow detection area in the video surveillance image, each vehicle identification frame corresponding to the one-way through lane is tracked to determine the second through-traffic flow in the one-way through lane; The through traffic flow of the target road section is determined based on the sum of the first through traffic flow and the second through traffic flow, wherein the target traffic flow includes: through traffic flow and turning traffic flow.

[0059] In a possible implementation of the embodiment of the present application, when the traffic detection module performs vehicle tracking for each vehicle identification frame corresponding to the multi-directional lane based on the vehicle flow detection area in the video surveillance image, and determines the first straight vehicle flow and turning vehicle flow in the multi-directional lane, it is used to: Based on the vehicle flow detection area in the video surveillance image, each vehicle identification frame corresponding to the multi-directional lane is tracked to determine the rotation direction angle set corresponding to each vehicle identification frame; Based on the rotation direction angle set corresponding to each vehicle identification frame, the first straight vehicle flow and the turning vehicle flow in the multi-directional lane are determined.

[0060] In a possible implementation of the embodiment of the present application, the vehicle flow detection device for a key road section further includes: A comparison and analysis module is used to predict the traffic flow of a target road section using a traffic flow prediction model to obtain the predicted traffic flow corresponding to the target road section; Perform traffic flow comparison analysis based on the predicted traffic flow corresponding to the target road section and the target traffic flow, and determine the comparison analysis result; When the comparison and analysis result is normal, the target traffic flow of the target road section is used as a training data set, and the traffic flow prediction model is retrained using the training data set to obtain an updated traffic flow prediction model.

[0061] In a possible implementation of the embodiment of the present application, the vehicle flow detection device for a key road section further includes: The precise detection module is used to re-detect the traffic flow of the target road section by using the positioning linkage method when the comparison and analysis result is abnormal, so as to obtain the precise target traffic flow. The positioning linkage method is to accurately detect the traffic flow through the vehicle's positioning system.

[0062] In a possible implementation of the embodiment of the present application, the vehicle flow detection device for a key road section further includes: A feature acquisition module is used to acquire a road image of a target road section during a traffic trough period when a lane feature acquisition cycle is reached, and to extract features based on the road image to obtain current lane features of the target road section; Obtain the stored lane features, perform feature comparison based on the current lane features and the stored lane features, and when there are differences in the features, use the current lane features to update the stored lane features and record them as the lane features of the target section.

[0063] Based on the same inventive concept mentioned above, an embodiment of the present application also discloses an intelligent terminal, which includes a memory and a processor, wherein the memory stores at least one instruction, at least one program, code set or instruction set, and the at least one instruction, at least one program, code set or instruction set is loaded and executed by the processor to implement a path navigation method based on real-time traffic flow as provided in the above method embodiment.

[0064] Based on the same inventive concept mentioned above, an embodiment of the present application also discloses a computer-readable storage medium, which stores at least one instruction, at least one program, code set or instruction set. The at least one instruction, at least one program, code set or instruction set can be loaded and executed by a processor to implement a path navigation method based on real-time traffic flow provided in the above method embodiment.

[0065] It should be understood that the "plurality" mentioned in this article refers to two or more. "And / or" describes the association relationship of the associated objects, indicating that there can be three relationships. For example, A and / or B can mean: A exists alone, A and B exist at the same time, and B exists alone. The character " / " generally indicates that the associated objects are in an "or" relationship.

[0066] Those skilled in the art will appreciate that all or part of the steps to implement the above embodiments may be accomplished by hardware or by instructing related hardware through a program, and the program may be stored in a computer-readable storage medium. The storage medium mentioned above may include, for example, various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0067] The above description is only an optional embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application shall be included in the protection scope of the present application.

Claims

1. A route navigation method based on real-time traffic flow, characterized in that: include: Acquire lane features of a target road section in the key road section, perform lane division based on the lane features, and determine a lane category corresponding to each lane in the key road section, wherein the lane categories include: a one-way straight lane and a multi-way lane, and the target road section is any branch road in the key road section; Acquire a video surveillance image of the target road section, perform preprocessing based on the video surveillance image, and select a vehicle flow detection area in the video surveillance image; Performing vehicle flow detection based on the lane category corresponding to each lane in the target road section and the vehicle flow detection area in the video surveillance image to determine the target vehicle flow of the target road section; The target traffic flow of each target section in the key section is integrated to obtain the total traffic flow of the key section.

2. The route navigation method based on real-time traffic flow according to claim 1 is characterized in that: The performing vehicle flow detection based on the lane category corresponding to each lane in the target road section and the vehicle flow detection area in the video surveillance image to determine the target vehicle flow of the target road section includes: Perform vehicle identification based on the vehicle flow detection area in the video surveillance image, and mark a vehicle identification frame corresponding to each vehicle; Based on the vehicle flow detection area in the video surveillance image, tracking each vehicle identification frame corresponding to the multi-directional lane to determine the first straight vehicle flow and turning vehicle flow in the multi-directional lane; Based on the vehicle flow detection area in the video surveillance image, tracking each vehicle identification frame corresponding to the one-way through lane to determine a second through-traffic flow in the one-way through lane; The through-traffic flow rate of the target road section is determined based on the sum of the first through-traffic flow rate and the second through-traffic flow rate, wherein the target traffic flow rate includes: the through-traffic flow rate and the turning traffic flow rate.

3. The route navigation method based on real-time traffic flow according to claim 2 is characterized in that: The method of tracking each vehicle identification frame corresponding to the multi-directional lane based on the vehicle flow detection area in the video surveillance image to determine the first straight vehicle flow and turning vehicle flow in the multi-directional lane includes: Based on the vehicle flow detection area in the video surveillance image, tracking each of the vehicle identification frames corresponding to the multi-directional lanes, and determining a set of rotation direction angles corresponding to each of the vehicle identification frames; The first through-traffic flow and the turning vehicle flow in the multi-directional lane are determined based on the rotation direction angle set corresponding to each vehicle identification frame.

4. The route navigation method based on real-time traffic flow according to claim 1, characterized in that: After performing vehicle flow detection based on the lane category corresponding to each lane in the target road section and the vehicle flow detection area in the video surveillance image to determine the target vehicle flow of the target road section, the method further includes: Using a traffic flow prediction model to predict the traffic flow of the target road section, to obtain a predicted traffic flow corresponding to the target road section; Performing a traffic flow comparison analysis based on the predicted traffic flow corresponding to the target road section and the target traffic flow, and determining a comparison analysis result; When the comparison and analysis result is normal, the target traffic flow of the target road section is used as a training data set, and the traffic flow prediction model is retrained using the training data set to obtain an updated traffic flow prediction model.

5. The route navigation method based on real-time traffic flow according to claim 4 is characterized in that: After comparing and analyzing the predicted traffic volume corresponding to the target road section with the target traffic volume and determining the comparison and analysis result, the method further includes: When the comparison and analysis result is abnormal, the vehicle flow of the target road section is re-detected using a positioning linkage method to obtain a precise target vehicle flow, wherein the positioning linkage method is to accurately detect the vehicle flow through the vehicle's positioning system.

6. The route navigation method based on real-time traffic flow according to claim 1, characterized in that: The method for collecting the lane features of the target road section includes: When the lane feature collection cycle is reached, a road image of the target road section during a traffic trough period is collected, and feature extraction is performed based on the road image to obtain the current lane feature of the target road section; The stored lane features are obtained, and feature comparison is performed based on the current lane features and the stored lane features. When there are differences in features, the stored lane features are updated using the current lane features and recorded as the lane features of the target road section.

7. A vehicle flow detection device, characterized in that: include: A lane division module is used to obtain lane features of a target road section in a key road section, perform lane division based on the lane features, and determine a lane category corresponding to each lane in the key road section, wherein the lane categories include: a one-way straight lane and a multi-way lane, and the target road section is any branch road in the key road section; A region division module is used to obtain a video surveillance image of the target road section, perform preprocessing based on the video surveillance image, and select a vehicle flow detection area in the video surveillance image; A traffic flow detection module, configured to perform traffic flow detection based on the lane category corresponding to each lane in the target road section and the traffic flow detection area in the video surveillance image, and determine a target traffic flow of the target road section; The traffic aggregation module is used to integrate the target traffic volume of each target section in the key section to obtain the total traffic volume of the key section.

8. An electronic device, characterized in that: It includes at least one processor; a memory; and at least one application, wherein the at least one application is stored in the memory and is configured to be executed by the at least one processor, and the at least one application is configured to execute the path navigation method based on real-time vehicle flow as described in any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that: A computer program is stored thereon, and when the computer program is executed in a computer, the computer is caused to execute the path navigation method based on real-time vehicle flow as described in any one of claims 1 to 6.

Images