Traffic management method and system for smart city

Abstract

The present invention discloses a traffic management method and system for a smart city, which relates to the field of smart traffic technology, including: generating multiple driving paths based on starting point information, dividing each driving path into multiple sub-paths with intersections as nodes, and obtaining the initial travel time of each sub-path, connecting each sub-path through nodes, and then setting the confluence coefficient, road condition coefficient and risk coefficient in turn for each sub-path, and adjusting the initial travel time of each sub-path; obtaining a panoramic image of each node by acquiring and processing, and obtaining the travel time of each node from different directions by using vehicle annotation and training; finally calculating the sum of the travel time of all sub-paths and the travel time of all nodes as the travel time of a single driving path, selecting the driving path corresponding to the shortest travel time as the optimal driving path of the vehicle, and greatly improving the accuracy of the optimal path planning of the vehicle.

Description

Technical Field

[0001] The present invention relates to the field of smart traffic technology, and in particular to a traffic management method and system for a smart city. Background Art

[0002] Smart city is a new form of city built with the goal of scientific development, more efficient management and better life, supported by information technology and communication technology, through transparent and sufficient information acquisition, extensive and secure information transmission and effective and scientific information processing, to improve the efficiency of urban operation, improve the level of public services and form a low-carbon urban ecosystem. As an important part of smart city, smart transportation integrates high-tech IT technologies such as the Internet of Things, cloud computing, big data, and mobile Internet, collects traffic information through high-tech, and provides traffic information services under real-time traffic data.

[0003] At present, in smart cities, when vehicles plan navigation routes, most of their data is based on the traffic volume and number of traffic lights on the navigation route, as well as the previous travel time of vehicles, to plan the optimal route for the vehicle. It does not consider the impact of multiple factors on vehicle traffic, such as road conditions, forks, confluences, road accident rates, and other factors such as the vehicle itself. As a result, the accuracy of optimal route planning for vehicles in existing smart cities needs to be improved. Summary of the invention

[0004] In order to solve the deficiencies of the prior art, the present invention provides a traffic management method for a smart city, the method comprising the following steps:

[0005] Acquire starting point information of the first vehicle, and generate multiple driving paths based on the starting point information;

[0006] Taking intersections as nodes, each driving path is divided into multiple sub-paths and the initial travel time of each sub-path is obtained;

[0007] Obtaining a confluence intersection of each sub-path in each driving path, processing each confluence intersection to obtain a first processing result, and setting a confluence coefficient for each sub-path based on the first processing result;

[0008] Acquire the road condition video of each sub-path in each driving path in real time and process it to obtain a second processing result, and set a road condition coefficient for each sub-path based on the second processing result;

[0009] Acquire vehicle traffic history data of each sub-path in each driving path, and set a risk coefficient for each sub-path based on the vehicle traffic history data;

[0010] Collect all images of each node in each driving path and process them to obtain a panoramic image of each node, and annotate the vehicles in the panoramic image, obtain multiple frames of panoramic images with vehicle annotation for training, and obtain the travel time of vehicles passing through each node from any entrance direction;

[0011] The first formula is used to calculate the travel time of each driving path to obtain the calculation result:

[0012]

[0013] is the travel time of each driving path, is the travel time of each sub-path in a single driving path, is the number of sub-paths in a single driving path, is the travel time of each node in a single driving path, is the number of nodes in a single driving path;

[0014] The second formula is used to calculate the travel time of each sub-path in a single driving path:

[0015]

[0016] is the confluence coefficient of each sub-path in a single driving path, is the risk factor of each sub-path in a single driving path, is the road condition coefficient of each sub-path in a single driving path, is the initial travel time of each sub-path in a single driving path;

[0017] Based on the calculation result, all driving paths are screened to obtain a first optimal driving path for the first vehicle.

[0018] The present invention is implemented through the following technical scheme: firstly, the starting point information of the first vehicle is obtained, and multiple driving paths are generated based on the starting point information. This can be achieved based on existing technologies (such as map software). However, the travel time of each of the multiple driving paths generated at this time is determined based on the path distance and the previous travel time of the vehicle, and other factors such as road conditions, forks, confluences and road accident rates are not considered. Therefore, this scheme first processes each driving path, takes the intersection as a node, divides the single driving path from the starting point to the end point into multiple sub-paths, obtains the initial travel time of each sub-path, and calculates the travel time of each sub-path. The reason for this division is that the travel time of vehicles is different at different intersections (such as three-way, four-way or five-way). If the traditional ratio between travel time and distance is used for calculation, the final node travel time is not accurate enough, which also leads to the travel time of a single driving path is not accurate enough. Therefore, this scheme divides a single driving path into multiple sub-paths, each of which is connected by nodes, and then calculates the travel time of multiple sub-paths and multiple nodes respectively. Finally, the sum of the travel time of all sub-paths and the travel time of all nodes is taken as the travel time of a single driving path.

[0019] For multiple sub-paths, this scheme considers the influence of confluence intersections, road conditions and accident risk factors in turn, and sets the confluence coefficient, road condition coefficient and risk coefficient for each sub-path in turn.

[0020] For each node, this solution collects all images of each node and processes them to obtain a panoramic image of each node, then annotates the vehicles in the panoramic image, and then obtains multiple frames of panoramic images with vehicle annotation for training to obtain the travel time of the vehicle passing through the node from any entrance direction.

[0021] Then, the travel time of a single driving path from the starting point to the end point is calculated by the first formula, that is, the travel time of all sub-paths and the travel time of all nodes in the single driving path is calculated; the travel time of each sub-path is calculated by the second formula, that is, on the premise of obtaining the initial travel time of each sub-path, the initial travel time of each sub-path is adjusted by the confluence coefficient, road condition coefficient and risk coefficient.

[0022] Finally, the travel time of each driving path is calculated by the first formula, and the driving path corresponding to the shortest travel time is selected as the first optimal driving path.

[0023] As an optional technical solution, the method further includes:

[0024] Determine whether the road condition coefficient of each sub-path is less than a preset value. If so, record the corresponding sub-path in each driving path as a first sub-path and process it to obtain a second processing result. If not, do nothing.

[0025] Based on the second processing result, a pothole area of ​​each first sub-path in each driving path is obtained, and it is determined whether a vehicle accident has occurred in the pothole area of ​​each first sub-path. If so, all vehicle data corresponding to the vehicle accident occurring within the first preset time period is obtained and processed to obtain a third processing result. Based on the third processing result, the first vehicle data is screened and obtained as the first threshold value. If not, no action is taken.

[0026] Acquire the monitoring image of the first vehicle and identify it to obtain second vehicle data, and determine whether the second vehicle data is greater than a first threshold value. If so, do not take any action; if not, mark the corresponding first sub-path;

[0027] The total distance of a single driving path is obtained as S, and the distance and value of all marked first sub-paths in the single driving path are obtained as L.

[0028] As an optional technical solution, the method further includes:

[0029] Based on the calculation results, the travel time of all driving paths is obtained;

[0030] Calculate the L / S ratio in each driving path in turn and record it as the first ratio and put it into the preset set;

[0031] Acquire all first ratios corresponding to values ​​less than a set value in the preset set to obtain a first acquisition result;

[0032] Based on the first acquisition result, a driving path with the shortest travel time is screened out from all first ratios that are smaller than the set value as a second optimal driving path for the first vehicle;

[0033] Determine whether the second optimal driving path coincides with the first optimal driving path; if so, use the first optimal driving path as the optimal driving path of the first vehicle; if not, use the second optimal driving path as the optimal driving path of the first vehicle.

[0034] As an optional technical solution, the first vehicle data and the second vehicle data are both vehicle ground clearance data.

[0035] As an optional technical solution, based on the third processing result, screening and obtaining the first vehicle data as the first threshold includes:

[0036] Acquire the ground clearance data of all vehicles corresponding to the vehicle accidents occurring within the first preset time period to obtain a second acquisition result;

[0037] Based on the second acquisition result, a maximum value of the vehicle ground clearance data is screened and obtained as the first threshold.

[0038] As an optional technical solution, based on the first processing result, setting the confluence coefficient for each sub-path includes:

[0039] Acquire the vehicle inflow flow of each converging intersection in real time to obtain a third acquisition result;

[0040] Based on the third acquisition result, an initial confluence coefficient is set for each confluence intersection;

[0041] It is determined whether the number of confluence intersections in each sub-path is 1. If so, the initial confluence coefficient of the confluence intersection is used as the confluence coefficient of the corresponding sub-path. If not, the average of the initial confluence coefficients of multiple confluence intersections is used as the confluence coefficient of the corresponding sub-path.

[0042] As an optional technical solution, based on the second processing result, setting the road condition coefficient for each sub-path includes:

[0043] Based on the traffic video of each sub-path, a multi-frame traffic image of each sub-path is obtained;

[0044] Crop and splice multiple frames of road condition images to obtain the road image of each sub-path;

[0045] Graying the road image, and binarizing it to extract objects in the road surface area of ​​each sub-path to obtain a corresponding binary image;

[0046] Calculating a second ratio of the area of ​​different objects in the binary image to the area of ​​the entire binary image;

[0047] Based on the second ratio, a road condition coefficient is set for each sub-path.

[0048] As an optional technical solution, setting a risk factor for each sub-path includes:

[0049] Based on the vehicle travel history data, the number of accidents occurring in each sub-path of each travel path within a second preset time period is obtained to obtain a fourth acquisition result;

[0050] Based on the fourth acquisition result, obtaining frequency data of accidents occurring in each sub-path in each driving path;

[0051] Based on the frequency data, a risk factor is set for each sub-path.

[0052] As an optional technical solution, all images of each node in each driving path are collected and processed including:

[0053] Obtain all images of each node and perform grayscale processing to obtain grayscale images;

[0054] All grayscale images are spliced ​​to obtain the node grayscale image;

[0055] Eliminating the overlapping areas in the node grayscale image;

[0056] Performing an inverse transformation on the eliminated node grayscale image to restore it into a color image;

[0057] The color image is used as a panoramic image of each node.

[0058] In order to solve the deficiencies of the prior art, the present invention also provides a traffic management system for a smart city, the system comprising:

[0059] A generating unit: used for acquiring the starting point information of the first vehicle, and generating a plurality of driving paths based on the starting point information;

[0060] Division unit: used to divide each driving path into multiple sub-paths with intersections as nodes and obtain the initial travel time of each sub-path;

[0061] A confluence coefficient unit: used for obtaining a confluence intersection of each sub-path in each driving path, and processing each confluence intersection to obtain a first processing result, and setting a confluence coefficient for each sub-path based on the first processing result;

[0062] A road condition coefficient unit: used for acquiring the road condition video of each sub-path in each driving path in real time and processing the video to obtain a second processing result, and setting a road condition coefficient for each sub-path based on the second processing result;

[0063] Risk coefficient unit: used to obtain the vehicle traffic history data of each sub-path in each driving path, and set the risk coefficient for each sub-path based on the vehicle traffic history data;

[0064] Node unit: collects all images of each node in each driving path and processes them to obtain a panoramic image of each node, and annotates the vehicles in the panoramic image, obtains multiple frames of panoramic images with vehicle annotation for training, and obtains the travel time of vehicles passing through each node from any entrance direction;

[0065] The first formula unit is used to calculate the travel time of each driving path using the first formula to obtain the calculation result:

[0066]

[0067] is the travel time of each driving path, is the travel time of each sub-path in a single driving path, is the number of sub-paths in a single driving path, is the travel time of each node in a single driving path, is the number of nodes in a single driving path;

[0068] The second formula unit is used to calculate the travel time of each sub-path in a single driving path using the second formula:

[0069]

[0070] is the confluence coefficient of each sub-path in a single driving path, is the risk factor of each sub-path in a single driving path, is the road condition coefficient of each sub-path in a single driving path, is the initial travel time of each sub-path in a single driving path;

[0071] Optimal driving path unit: used to screen all driving paths based on the calculation result to obtain a first optimal driving path for the first vehicle.

[0072] One or more technical solutions provided by the present invention have at least the following technical effects or advantages:

[0073] The present invention provides a traffic management method for a smart city. A plurality of driving paths are generated based on starting point information. Each driving path is divided into a plurality of sub-paths with intersections as nodes, and the initial travel time of each sub-path is obtained. Each sub-path is connected through nodes. Then, a confluence coefficient, a road condition coefficient and a risk coefficient are sequentially set for each sub-path, and the initial travel time of each sub-path is adjusted. A panoramic image of each node is obtained by acquiring and processing, and the travel time of each node from different directions is obtained by vehicle annotation and training. Finally, the sum of the travel time of all sub-paths and the travel time of all nodes is calculated as the travel time of a single driving path, and the driving path corresponding to the shortest travel time is selected as the optimal driving path of the vehicle, thereby greatly improving the accuracy of the optimal path planning of the vehicle.

[0074] The present invention further processes the road conditions of the driving path. In the past, when the optimal vehicle path was planned, the connection between the vehicle and the road was not considered. In the actually planned road, potholes are inevitable. Different vehicles also have different situations when passing through potholes of different depths. Therefore, the present invention uses the vehicle ground clearance as a basis to obtain the pothole area in each first sub-path, and obtain the highest vehicle ground clearance corresponding to the vehicle accident in each pothole area, and then obtain the vehicle ground clearance of the first vehicle through the Internet of Things and cloud computing and other technologies. If the vehicle ground clearance of the first vehicle is less than the highest vehicle ground clearance, then the first vehicle passes through the pothole. There is a risk of scratching the first vehicle when passing through the pothole area, but if only the pothole area in a single section of the first sub-path will cause the risk of scratching the vehicle, taking measures such as reducing the road condition coefficient or detouring will also affect the accuracy of the optimal driving path planning. Therefore, the present invention obtains the sum of all first sub-paths in a single driving path that may cause scratches to the first vehicle, calculates the ratio of the sum to the total distance of the single driving path, and uses the ratio to judge the risk of scratching the first vehicle when passing the corresponding driving path, and adjusts the optimal driving path based on the judgment result, thereby improving the safety of the vehicle passing the driving path while ensuring the accuracy of the optimal driving path planning of the vehicle. BRIEF DESCRIPTION OF THE DRAWINGS

[0075] The drawings described herein are used to provide a further understanding of the embodiments of the present invention, constitute a part of the present invention, and do not constitute a limitation on the embodiments of the present invention;

[0076] Figure 1 It is a flow chart of a traffic management method for a smart city in the present invention;

[0077] Figure 2 It is a schematic diagram of the composition of a traffic management system for a smart city in the present invention. DETAILED DESCRIPTION

[0078] In order to more clearly understand the above-mentioned purpose, features and advantages of the present invention, the present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be noted that the embodiments of the present invention and the features in the embodiments can be combined with each other without conflict.

[0079] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those within the scope of this description. Therefore, the protection scope of the present invention is not limited to the specific embodiments disclosed below.

[0080] Embodiment 1

[0081] Please refer to Figure 1 , Figure 1 The present invention is a flow chart of a traffic management method for a smart city, wherein the method comprises the following steps:

[0082] Acquire starting point information of the first vehicle, and generate multiple driving paths based on the starting point information;

[0083] Taking intersections as nodes, each driving path is divided into multiple sub-paths and the initial travel time of each sub-path is obtained;

[0084] Obtaining a confluence intersection of each sub-path in each driving path, processing each confluence intersection to obtain a first processing result, and setting a confluence coefficient for each sub-path based on the first processing result;

[0085] Acquire the road condition video of each sub-path in each driving path in real time and process it to obtain a second processing result, and set a road condition coefficient for each sub-path based on the second processing result;

[0086] Acquire vehicle traffic history data of each sub-path in each driving path, and set a risk coefficient for each sub-path based on the vehicle traffic history data;

[0087] Collect all images of each node in each driving path and process them to obtain a panoramic image of each node, and annotate the vehicles in the panoramic image, obtain multiple frames of panoramic images with vehicle annotation for training, and obtain the travel time of vehicles passing through each node from any entrance direction;

[0088] The first formula is used to calculate the travel time of each driving path to obtain the calculation result:

[0089]

[0090] is the travel time of each driving path, is the travel time of each sub-path in a single driving path, is the number of sub-paths in a single driving path, is the travel time of each node in a single driving path, is the number of nodes in a single driving path;

[0091] The second formula is used to calculate the travel time of each sub-path in a single driving path:

[0092]

[0093] is the confluence coefficient of each sub-path in a single driving path, is the risk factor of each sub-path in a single driving path, is the road condition coefficient of each sub-path in a single driving path, is the initial travel time of each sub-path in a single driving path;

[0094] Based on the calculation result, all driving paths are screened to obtain a first optimal driving path for the first vehicle.

[0095] The specific embodiments of the present invention are as follows:

[0096] The starting point information of the first vehicle is obtained, and multiple driving paths are generated based on the starting point information, such as generating five driving paths A, B, C, D and E, and the five driving paths are divided based on intersections to obtain multiple sub-paths and obtain the initial travel time of each sub-path.

[0097] Process each sub-path: obtain the confluence intersection of each sub-path in each driving path, and process each confluence intersection to obtain a first processing result, and set a confluence coefficient for each sub-path based on the first processing result; obtain the road condition video of each sub-path in each driving path in real time and process it to obtain a second processing result, and set a road condition coefficient for each sub-path based on the second processing result; obtain the vehicle traffic history data of each sub-path in each driving path, and set a risk coefficient for each sub-path based on the vehicle traffic history data.

[0098] Based on the first processing result, setting the confluence coefficient for each sub-path includes:

[0099] Acquire the vehicle inflow flow of each converging intersection in real time to obtain a third acquisition result;

[0100] Based on the third acquisition result, an initial confluence coefficient is set for each confluence intersection;

[0101] It is determined whether the number of confluence intersections in each sub-path is 1. If so, the initial confluence coefficient of the confluence intersection is used as the confluence coefficient of the corresponding sub-path. If not, the average of the initial confluence coefficients of multiple confluence intersections is used as the confluence coefficient of the corresponding sub-path.

[0102] The initial confluence coefficient is set for each confluence intersection by acquiring the vehicle confluence flow at each confluence intersection in real time. The initial confluence coefficient is proportional to the vehicle confluence flow. The confluence coefficient of each sub-path is calculated by calculating the average of all initial confluence coefficients on the sub-path. It should be noted that if there is no confluence intersection in a single sub-path, the confluence coefficient of the corresponding sub-path is recorded as 1.

[0103] Based on the second processing result, setting a road condition coefficient for each sub-path includes:

[0104] Based on the traffic video of each sub-path, a multi-frame traffic image of each sub-path is obtained;

[0105] Crop and splice multiple frames of road condition images to obtain the road image of each sub-path;

[0106] Graying the road image, and binarizing it to extract objects in the road surface area of ​​each sub-path to obtain a corresponding binary image;

[0107] Calculating a second ratio of the area of ​​different objects in the binary image to the area of ​​the entire binary image;

[0108] Based on the second ratio, a road condition coefficient is set for each sub-path.

[0109] Among them, by acquiring the road condition video of each sub-path, multiple frames of road condition images of each sub-path are obtained, all road condition images of a single sub-path are cropped and spliced ​​to obtain the road image of each sub-path, and then the single road image is grayed and binarized, and then the second ratio of the area of ​​different objects (i.e. the area of ​​the pothole area) to the area of ​​the entire binary image (i.e. the entire road section) is obtained. The ratio is used to judge the road condition of the corresponding sub-path and set the road condition coefficient.

[0110] Setting the risk factor for each sub-path includes:

[0111] Based on the vehicle travel history data, the number of accidents occurring in each sub-path of each travel path within a second preset time period is obtained to obtain a fourth acquisition result;

[0112] Based on the fourth acquisition result, obtaining frequency data of accidents occurring in each sub-path in each driving path;

[0113] Based on the frequency data, a risk factor is set for each sub-path.

[0114] Among them, based on the vehicle traffic history data, the number of accidents in each sub-path within the second preset time period (such as one year) is obtained, so as to obtain the frequency data of accidents in each sub-path. Based on the frequency data of the accident, a risk coefficient is set for each sub-path. However, it should be noted that the setting of the risk coefficient only prompts the risk of an accident when passing through the path, which is ultimately reflected in the impact on the travel time. The higher the risk coefficient, the more it is necessary to reduce the speed appropriately, that is, the corresponding travel time becomes longer. Similarly, if the risk coefficient is low, the speed can be appropriately increased, and the corresponding travel time becomes shorter.

[0115] Processing each node: Collect all images of each node on each driving path and process them to obtain a panoramic image of each node, and annotate the vehicles in the panoramic image, obtain multiple frames of panoramic images after vehicle annotation for training, and obtain the travel time of passing through each node from any entrance direction. Training multiple frames of panoramic images after vehicle annotation includes obtaining the travel time of multiple annotated vehicles entering other roads from any direction of the node panoramic image to obtain an acquisition result, and based on the acquisition result, obtain the travel time of the vehicle passing through each node.

[0116] Collecting and processing all images of each node in each driving path includes:

[0117] Obtain all images of each node and perform grayscale processing to obtain grayscale images;

[0118] All grayscale images are spliced ​​to obtain the node grayscale image;

[0119] Eliminating the overlapping areas in the node grayscale image;

[0120] Performing an inverse transformation on the eliminated node grayscale image to restore it into a color image;

[0121] The color image is used as a panoramic image of each node.

[0122] The first formula is used to calculate the travel time of each driving path to obtain the calculation result:

[0123]

[0124] is the travel time of each driving path, is the travel time of each sub-path in a single driving path, is the number of sub-paths in a single driving path, is the travel time of each node in a single driving path, is the number of nodes in a single driving path;

[0125] The second formula is used to calculate the travel time of each sub-path in a single driving path:

[0126]

[0127] is the confluence coefficient of each sub-path in a single driving path, is the risk factor of each sub-path in a single driving path, is the road condition coefficient of each sub-path in a single driving path, is the initial travel time of each sub-path in a single driving path;

[0128] Based on the calculation result, all driving paths are screened to obtain a first optimal driving path for the first vehicle.

[0129] If the travel time of the five driving paths A, B, C, D and E is calculated by the first and second formulas, they are T A 、T B 、T C 、T D and T E , and T B <T C <T A <T E <T D , then based on the calculation results, the first optimal driving path of the first vehicle is path B.

[0130] In the past, when the optimal vehicle path planning was performed, the connection between the vehicle and the road was not considered. In the actual planned road, potholes are inevitable. Different vehicles also have different situations when passing through potholes of different depths. Therefore, this embodiment also provides an adjustment method for the optimal driving path planning, including:

[0131] Determine whether the road condition coefficient of each sub-path is less than a preset value. If so, record the corresponding sub-path in each driving path as a first sub-path and process it to obtain a second processing result. If not, do nothing.

[0132] Based on the second processing result, a pothole area of ​​each first sub-path in each driving path is obtained, and it is determined whether a vehicle accident has occurred in the pothole area of ​​each first sub-path. If so, all vehicle data corresponding to the vehicle accident occurring within the first preset time period is obtained and processed to obtain a third processing result. Based on the third processing result, the first vehicle data is screened and obtained as the first threshold value. If not, no action is taken.

[0133] Acquire the monitoring image of the first vehicle and identify it to obtain second vehicle data, and determine whether the second vehicle data is greater than a first threshold value. If so, do not take any action; if not, mark the corresponding first sub-path;

[0134] The total distance of a single driving path is obtained as S, and the distance and value of all marked first sub-paths in the single driving path are obtained as L.

[0135] This embodiment further includes: obtaining the travel time of all driving paths based on the calculation result;

[0136] Calculate the L / S ratio in each driving path in turn and record it as the first ratio and put it into the preset set;

[0137] Acquire all first ratios corresponding to values ​​less than a set value in the preset set to obtain a first acquisition result;

[0138] Based on the first acquisition result, a driving path with the shortest travel time is screened out from all first ratios that are smaller than the set value as a second optimal driving path for the first vehicle;

[0139] Determine whether the second optimal driving path coincides with the first optimal driving path; if so, use the first optimal driving path as the optimal driving path of the first vehicle; if not, use the second optimal driving path as the optimal driving path of the first vehicle.

[0140] This embodiment first makes a judgment based on the road condition coefficient, obtains the sub-path corresponding to the road condition coefficient lower than the preset value as the first sub-path, and then obtains the pothole area in each section of the first sub-path. The specific acquisition method has been explained in the method of setting the road condition coefficient, so it will not be expanded. It is judged whether a vehicle accident has occurred in each pothole area. If so, the highest ground clearance data of the vehicle corresponding to the accident is obtained as the first threshold value. If not, no action is taken; then the image of the first vehicle is obtained and identified to obtain the vehicle ground clearance data of the first vehicle, and it is judged whether the vehicle ground clearance data is greater than the first threshold value. If If yes, no action is taken. If no, the corresponding first sub-path is marked. The ratio between the distance sum value L of all marked first sub-paths in each driving path and the distance S of a single driving path is calculated to determine whether the ratio is less than the set value. If so, it means that the risk of scratching the potholes in the driving path is relatively small when the vehicle passes through. Therefore, the L / S ratio of the five driving paths needs to be calculated. If the L / S ratio of the B driving path is calculated to exceed the set value, and the L / S ratio of the C driving path is less than the set value, the C driving path is used as the new optimal driving path. Under the premise of ensuring the optimal driving path planning of the vehicle, the safety of the vehicle passing through the optimal driving path is further improved.

[0141] Embodiment 2

[0142] Please refer to Figure 2 , Figure 2 This is a schematic diagram of a traffic management system for a smart city in the present invention, the system comprising:

[0143] A generating unit: used for acquiring the starting point information of the first vehicle, and generating a plurality of driving paths based on the starting point information;

[0144] Division unit: used to divide each driving path into multiple sub-paths with intersections as nodes and obtain the initial travel time of each sub-path;

[0145] A confluence coefficient unit: used for obtaining a confluence intersection of each sub-path in each driving path, and processing each confluence intersection to obtain a first processing result, and setting a confluence coefficient for each sub-path based on the first processing result;

[0146] A road condition coefficient unit: used for acquiring the road condition video of each sub-path in each driving path in real time and processing the video to obtain a second processing result, and setting a road condition coefficient for each sub-path based on the second processing result;

[0147] Risk coefficient unit: used to obtain the vehicle traffic history data of each sub-path in each driving path, and set the risk coefficient for each sub-path based on the vehicle traffic history data;

[0148] Node unit: collects all images of each node in each driving path and processes them to obtain a panoramic image of each node, and annotates the vehicles in the panoramic image, obtains multiple frames of panoramic images with vehicle annotation for training, and obtains the travel time of vehicles passing through each node from any entrance direction;

[0149] The first formula unit is used to calculate the travel time of each driving path using the first formula to obtain the calculation result:

[0150]

[0151] is the travel time of each driving path, is the travel time of each sub-path in a single driving path, is the number of sub-paths in a single driving path, is the travel time of each node in a single driving path, is the number of nodes in a single driving path;

[0152] The second formula unit is used to calculate the travel time of each sub-path in a single driving path using the second formula:

[0153]

[0154] is the confluence coefficient of each sub-path in a single driving path, is the risk factor of each sub-path in a single driving path, is the road condition coefficient of each sub-path in a single driving path, is the initial travel time of each sub-path in a single driving path;

[0155] Optimal driving path unit: used to screen all driving paths based on the calculation result to obtain a first optimal driving path for the first vehicle.

[0156] Although the preferred embodiments of the present invention have been described, those skilled in the art may make other changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention.

[0157] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A traffic management method for a smart city, characterized in that: The method comprises the following steps: Acquire starting point information of the first vehicle, and generate multiple driving paths based on the starting point information; Taking intersections as nodes, each driving path is divided into multiple sub-paths and the initial travel time of each sub-path is obtained; Obtaining a confluence intersection of each sub-path in each driving path, processing each confluence intersection to obtain a first processing result, and setting a confluence coefficient for each sub-path based on the first processing result; Acquire the road condition video of each sub-path in each driving path in real time and process it to obtain a second processing result, and set a road condition coefficient for each sub-path based on the second processing result; Acquire vehicle traffic history data of each sub-path in each driving path, and set a risk coefficient for each sub-path based on the vehicle traffic history data; Collect all images of each node in each driving path and process them to obtain a panoramic image of each node, and annotate the vehicles in the panoramic image, obtain multiple frames of panoramic images with vehicle annotation for training, and obtain the travel time of vehicles passing through each node from any entrance direction; The first formula is used to calculate the travel time of each driving path to obtain the calculation result: is the travel time of each driving path, is the travel time of each sub-path in a single driving path, is the number of sub-paths in a single driving path, is the travel time of each node in a single driving path, is the number of nodes in a single driving path; The second formula is used to calculate the travel time of each sub-path in a single driving path: is the confluence coefficient of each sub-path in a single driving path, is the risk factor of each sub-path in a single driving path, is the road condition coefficient of each sub-path in a single driving path, is the initial travel time of each sub-path in a single driving path; Based on the calculation result, all driving paths are screened to obtain a first optimal driving path for the first vehicle; The method further comprises: Determine whether the road condition coefficient of each sub-path is less than a preset value. If so, record the corresponding sub-path in each driving path as a first sub-path and process it to obtain a fourth processing result. If not, do nothing. Based on the fourth processing result, the pothole area of ​​each first sub-path in each driving path is obtained, and it is determined whether a vehicle accident has occurred in the pothole area of ​​each first sub-path. If so, all vehicle data corresponding to the vehicle accident occurring within the first preset time period is obtained and processed to obtain a third processing result. Based on the third processing result, the first vehicle data is screened and obtained as the first threshold value. If not, no action is taken. Acquire the monitoring image of the first vehicle and identify it to obtain second vehicle data, and determine whether the second vehicle data is greater than a first threshold value. If so, do not take any action; if not, mark the corresponding first sub-path; The total distance of a single driving path is obtained as S, and the distance and value of all marked first sub-paths in the single driving path are obtained as L; The method further comprises: Based on the calculation results, the travel time of all driving paths is obtained; Calculate the L / S ratio in each driving path in turn and record it as the first ratio and put it into the preset set; Acquire all first ratios corresponding to values ​​less than a set value in the preset set to obtain a first acquisition result; Based on the first acquisition result, a driving path with the shortest travel time is screened out from all first ratios that are smaller than the set value as a second optimal driving path for the first vehicle; Determine whether the second optimal driving path coincides with the first optimal driving path, if so, use the first optimal driving path as the optimal driving path of the first vehicle, if not, use the second optimal driving path as the optimal driving path of the first vehicle; The first vehicle data and the second vehicle data are both vehicle ground clearance data.

2. A traffic management method for a smart city according to claim 1, characterized in that: Based on the third processing result, filtering and obtaining the first vehicle data as the first threshold includes: Acquire the ground clearance data of all vehicles corresponding to the vehicle accidents occurring within the first preset time period to obtain a second acquisition result; Based on the second acquisition result, a maximum value of the vehicle ground clearance data is screened and obtained as the first threshold.

3. A traffic management method for a smart city according to claim 1, characterized in that: Based on the first processing result, setting the confluence coefficient for each sub-path includes: Acquire the vehicle inflow flow of each converging intersection in real time to obtain a third acquisition result; Based on the third acquisition result, an initial confluence coefficient is set for each confluence intersection; It is determined whether the number of confluence intersections in each sub-path is 1. If so, the initial confluence coefficient of the confluence intersection is used as the confluence coefficient of the corresponding sub-path. If not, the average of the initial confluence coefficients of multiple confluence intersections is used as the confluence coefficient of the corresponding sub-path.

4. A traffic management method for a smart city according to claim 1, characterized in that: Based on the second processing result, setting a road condition coefficient for each sub-path includes: Based on the traffic video of each sub-path, a multi-frame traffic image of each sub-path is obtained; Crop and splice multiple frames of road condition images to obtain the road image of each sub-path; Graying the road image, and binarizing it to extract objects in the road surface area of ​​each sub-path to obtain a corresponding binary image; Calculating a second ratio of the area of ​​different objects in the binary image to the area of ​​the entire binary image; Based on the second ratio, a road condition coefficient is set for each sub-path.

5. The traffic management method for a smart city according to claim 1, characterized in that: Setting the risk factor for each sub-path includes: Based on the vehicle travel history data, the number of accidents occurring in each sub-path of each travel path within a second preset time period is obtained to obtain a fourth acquisition result; Based on the fourth acquisition result, obtaining frequency data of accidents occurring in each sub-path in each driving path; Based on the frequency data, a risk factor is set for each sub-path.

6. A traffic management method for a smart city according to claim 1, characterized in that: Collecting and processing all images of each node in each driving path includes: Obtain all images of each node and perform grayscale processing to obtain grayscale images; All grayscale images are spliced ​​to obtain the node grayscale image; Eliminating the overlapping areas in the node grayscale image; Performing an inverse transformation on the eliminated node grayscale image to restore it into a color image; The color image is used as a panoramic image of each node.

7. A traffic management system for a smart city, implemented by a traffic management method for a smart city as claimed in claim 1, characterized in that: The system comprises: A generating unit: used for acquiring the starting point information of the first vehicle, and generating a plurality of driving paths based on the starting point information; Division unit: used to divide each driving path into multiple sub-paths with the intersection as a node and obtain the initial travel time of each sub-path; A confluence coefficient unit: used for obtaining a confluence intersection of each sub-path in each driving path, and processing each confluence intersection to obtain a first processing result, and setting a confluence coefficient for each sub-path based on the first processing result; A road condition coefficient unit: used for acquiring the road condition video of each sub-path in each driving path in real time and processing the video to obtain a second processing result, and setting a road condition coefficient for each sub-path based on the second processing result; Risk coefficient unit: used to obtain the vehicle traffic history data of each sub-path in each driving path, and set the risk coefficient for each sub-path based on the vehicle traffic history data; Node unit: collects all images of each node in each driving path and processes them to obtain a panoramic image of each node, and annotates the vehicles in the panoramic image, obtains multiple frames of panoramic images with vehicle annotation for training, and obtains the travel time of vehicles passing through each node from any entrance direction; The first formula unit is used to calculate the travel time of each driving path using the first formula to obtain the calculation result: is the travel time of each driving path, is the travel time of each sub-path in a single driving path, is the number of sub-paths in a single driving path, is the travel time of each node in a single driving path, is the number of nodes in a single driving path; The second formula unit is used to calculate the travel time of each sub-path in a single driving path using the second formula: is the confluence coefficient of each sub-path in a single driving path, is the risk factor of each sub-path in a single driving path, is the road condition coefficient of each sub-path in a single driving path, is the initial travel time of each sub-path in a single driving path; Optimal driving path unit: used to screen all driving paths based on the calculation result to obtain a first optimal driving path for the first vehicle.

Images