A monitoring-based vehicle-road collaboration method and device

By setting up cameras and processors on the road to identify vehicles, judging road congestion conditions and adjusting traffic light signals, the traffic congestion problem is solved, and efficient utilization of road resources and improvement of traffic conditions is achieved.

CN118470992BActive Publication Date: 2024-11-15JIANGSU DIANCHUANG INFORMATION TECH CO LTD
View PDF 6 Cites 0 Cited by

Patent Information

Application Number
CN202410918577.4
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-07-10
Publication Date
2024-11-15
Estimated Expiration
2044-07-10

AI Technical Summary

Technical Problem

The growth of urban population and vehicle numbers has led to traffic congestion, and it is difficult for existing technology to effectively control traffic flow, optimize the utilization of road resources, and improve traffic congestion.

Method used

By setting up cameras at road imports and exits, using processors and convolutional neural networks to identify vehicles, judge road congestion, and adjust traffic light signals according to traffic flow proportion, guide vehicles to evacuate roads, and transform congested roads into smooth passageways.

Benefits of technology

Effectively control traffic, reduce congestion, improve road traffic efficiency, and improve traffic conditions.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118470992B_ABST
    Figure CN118470992B_ABST
Patent Text Reader

Abstract

The present invention discloses a vehicle-road collaboration method and device based on monitoring, and relates to the technical field of vehicle-road collaboration. An import camera and an export camera are set on the road, and import vehicles on the road are acquired through the import camera, and export vehicles on the road are acquired through the export camera. Whether the road is a congested road is determined based on the import vehicles and export vehicles. If it is a congested road, the import vehicles and export vehicles on the congested road are acquired, and the mainstream vehicle source is acquired based on the import vehicles, and the diversion vehicle source is acquired based on the export vehicles. The mainstream road corresponding to the congested road is acquired based on the mainstream vehicle source, and the diversion road corresponding to the congested road is acquired based on the diversion vehicle source. The mainstream road and diversion road corresponding to the congested road are acquired, and the evacuation road is acquired based on the mainstream road and the diversion road, and the traffic light signal information of the congested road and the evacuation road is acquired, and the traffic light signal information is adjusted to transform the congested road into an unobstructed road.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of vehicle-road collaboration technology, and specifically to a vehicle-road collaboration method and device based on monitoring. Background Art

[0002] Vehicle-road collaboration adopts advanced wireless communications and new generation Internet technologies, implements all-round dynamic real-time information interaction between vehicles and roads, and collects and integrates dynamic traffic information in all time and space, to achieve active vehicle safety control and collaborative road management, fully implement effective collaboration between people, vehicles and roads, ensure traffic safety, improve traffic efficiency, and form a safe, efficient and environmentally friendly road traffic system.

[0003] Nowadays, with the continuous growth of urban population and the number of vehicles, traffic congestion has become a common problem in urban life. The dense traffic on the road has brought many inconveniences to people's travel; therefore, how to effectively control the direction of traffic through traffic lights, optimize the utilization of road resources, and improve congested traffic conditions is what we need to consider at present. Summary of the invention

[0004] In order to solve the above problems, the purpose of the present invention is to provide a monitoring-based vehicle-road collaboration method and device.

[0005] The purpose of the present invention can be achieved by the following technical solution: A monitoring-based vehicle-road collaboration method comprises the following steps:

[0006] Step S1: an import camera is set at the road import end, and an export camera is set at the road export end, and the import vehicles of the road are acquired through the import camera, and the export vehicles of the road are acquired through the export camera;

[0007] Step S2: judging whether there is congestion on the road according to the number of inbound vehicles and outbound vehicles; if there is congestion, marking the road as a congested road; otherwise, marking the road as a clear road;

[0008] Step S3: Obtain the inbound vehicles and outbound vehicles of the congested road, obtain the mainstream vehicle source according to the inbound vehicles, and obtain the diversion vehicle source according to the outbound vehicles; obtain the mainstream road corresponding to the congested road according to the mainstream vehicle source, and obtain the diversion road corresponding to the congested road according to the diversion vehicle source;

[0009] Step S4: Obtain the mainstream road and diversion road corresponding to the congested road, obtain the evacuation road according to the mainstream road and the diversion road, obtain the traffic light signal information of the congested road and the evacuation road, and adjust the traffic light signal information to transform the congested road into a smooth road.

[0010] Furthermore, the process of acquiring the import vehicles on the road through the import camera and acquiring the export vehicles on the road through the export camera includes:

[0011] The inlet camera and the outlet camera are both provided with a processor, and the video footage collected by the inlet camera and the outlet camera is processed by the processor;

[0012] Collect vehicle images under different climate and lighting conditions, mark them as sample images, generate a sample image set, extract features from the sample image set through a convolutional neural network, obtain vehicle recognition features, and then build a vehicle recognition model based on the vehicle recognition features;

[0013] Set the processing cycle;

[0014] The vehicle driving video at the road entrance is obtained through the import camera. According to the processing cycle, the obtained vehicle driving video is regularly sent to the processor, and the processor performs the following operations on the vehicle driving video:

[0015] Step a1: cutting the vehicle driving video into frames to obtain a number of vehicle driving images, setting effective areas in the number of vehicle driving images, and generating a vehicle driving image set in chronological order;

[0016] Step a2: obtaining the first vehicle driving image in the first vehicle driving image set, identifying the valid area in the image according to the vehicle recognition model, marking the identified vehicles as valid driving vehicles, obtaining the number of valid driving vehicles, and marking the number as the vehicle initial value;

[0017] Step a3: obtain the first vehicle driving image in the vehicle driving image set, mark it as a sample image, identify the valid area in the sample image according to the vehicle recognition model, mark the identified vehicle as a valid driving vehicle, obtain the individual features of the valid driving vehicle through a convolutional neural network, the individual features are unique, mark the individual features as initial vehicle features, and generate an initial vehicle feature set; obtain the number of valid driving vehicles, and mark the number as the vehicle initial value; obtain the second vehicle driving image in the vehicle driving image set, mark it as a comparison image, identify the valid area in the comparison image according to the vehicle recognition model, obtain the valid driving vehicle in the comparison image, and then obtain the individual features of the valid driving vehicle through a convolutional neural network, mark them as the latest vehicle features, and generate the latest vehicle feature set; compare the latest vehicle feature set with the initial vehicle feature set, delete the latest vehicle features in the latest vehicle feature set that are consistent with the initial vehicle features in the initial vehicle feature set, obtain the number of the latest vehicle features remaining in the latest vehicle feature set, and mark the number as the vehicle newly added value;

[0018] Step a3: Repeat step a2 to obtain several vehicle added values ​​in turn, then the number of imported vehicles on the road = , where n is a positive integer greater than 1;

[0019] The vehicle driving video at the road exit is obtained through the exit camera, and the obtained vehicle driving video is sent to the processor regularly according to the processing cycle; the exit vehicle of the road is obtained by using the steps similar to those for obtaining the road import vehicle. , where n is a positive integer greater than 1.

[0020] Furthermore, the process of determining whether there is congestion on the road based on the import and export vehicles includes:

[0021] Set congestion period thresholds and congestion vehicle thresholds;

[0022] Obtain the import and export vehicles on the road in real time, and obtain the real-time capacity of the road based on the import and export vehicles on the road = import vehicles - export vehicles;

[0023] Obtain the duration of time when the real-time capacity of the road is greater than the congestion vehicle threshold, which is recorded as the congestion period;

[0024] If the congestion period is greater than the congestion period threshold, it is determined that the road is congested; otherwise, it is determined that the road is not congested.

[0025] Furthermore, the process of obtaining the import vehicles and export vehicles of the congested road, obtaining the mainstream vehicle source according to the import vehicles, and obtaining the diversion vehicle source according to the export vehicles includes:

[0026] Divide the import vehicles into a plurality of guided import vehicles according to the lanes, wherein the guided import vehicles include left-turn import vehicles, straight-ahead import vehicles and right-turn import vehicles; divide the export vehicles into a plurality of guided export vehicles according to the lanes, wherein the guided export vehicles include left-turn export vehicles, straight-ahead export vehicles and right-turn export vehicles;

[0027] Set the threshold for the proportion of congested traffic;

[0028] Get the ratio of left-turn import vehicles = ; Get the proportion of direct imported vehicles = ; Get the ratio of right-turn import vehicles = ;

[0029] If the proportion of guided import vehicles is greater than the congested traffic proportion threshold, the guided import vehicle is marked as a mainstream vehicle source, otherwise, the guided import vehicle is not marked as a mainstream vehicle source;

[0030] If none of the guided import vehicles are marked as mainstream vehicle sources, all guided import vehicles are marked as mainstream vehicle sources;

[0031] That is, the proportion of left-turn import vehicles, the proportion of straight import vehicles, and the proportion of right-turn import vehicles are all less than or equal to the congested traffic proportion threshold, so that left-turn import vehicles, straight import vehicles, and right-turn import vehicles are not marked as mainstream vehicle sources, then left-turn import vehicles, straight import vehicles, and right-turn import vehicles are all marked as mainstream vehicle sources;

[0032] Get the ratio of vehicles turning left at the exit = , the proportion of vehicles going straight out of the gate = and the proportion of vehicles turning right at the exit = ;

[0033] If the proportion of the exit-directed vehicle is greater than the congested traffic proportion threshold, the exit-directed vehicle is marked as a diversion vehicle source, otherwise, the exit-directed vehicle is not marked as a diversion vehicle source;

[0034] If none of the vehicles leading to the exit are marked as diversion vehicle sources, all vehicles leading to the exit are marked as diversion vehicle sources;

[0035] That is, the proportion of left-turn exit vehicles, the proportion of straight exit vehicles and the proportion of right-turn exit vehicles are all less than or equal to the congested traffic proportion threshold, so that the left-turn exit vehicles, the straight exit vehicles and the right-turn exit vehicles are not marked as diversion vehicle sources, then the left-turn exit vehicles, the straight exit vehicles and the right-turn exit vehicles are all marked as diversion vehicle sources.

[0036] Furthermore, the process of obtaining the mainstream road corresponding to the congested road according to the mainstream vehicle source and obtaining the diversion road corresponding to the congested road according to the diversion vehicle source includes:

[0037] Acquire aggregated roads and discrete roads of congested roads; the aggregated roads include left aggregated roads, straight aggregated roads and right aggregated roads; the discrete roads include left discrete roads, straight discrete roads and right discrete roads;

[0038] The left convergent road means that vehicles can enter the congested road by turning left on the left convergent road; the straight convergent road means that vehicles can enter the congested road by going straight on the straight convergent road; the right convergent road means that vehicles can enter the congested road by turning right on the right convergent road;

[0039] The left discrete road is a road that a vehicle can enter by turning left on a congested road; the straight discrete road is a road that a vehicle can enter by going straight on a congested road; the right discrete road is a road that a vehicle can enter by turning right on a congested road;

[0040] The guided import vehicles and the convergent roads have a one-to-one correspondence, that is, the left-turning import vehicles correspond to the left convergent road, the straight-going import vehicles correspond to the straight convergent road, and the right-turning import vehicles correspond to the right convergent road;

[0041] The guided exit vehicles have a one-to-one correspondence with the discrete roads, that is, the left-turn exit vehicles correspond to the left discrete road, the straight-going exit vehicles correspond to the straight discrete road, and the right-turn exit vehicles correspond to the right discrete road;

[0042] The aggregated roads corresponding to the mainstream vehicle sources are obtained as the mainstream roads of the congested road; and the discrete roads corresponding to the diversion vehicle sources are obtained as the diversion roads of the congested road.

[0043] Further, the process of obtaining the mainstream road and the diversion road corresponding to the congested road, obtaining the evacuation road according to the mainstream road and the diversion road, obtaining the traffic light signal information of the congested road and the evacuation road, and adjusting the traffic light signal information to transform the congested road into a smooth road includes:

[0044] Obtaining mainstream roads and diversion roads corresponding to congested roads; the evacuation roads include mainstream evacuation roads and diversion evacuation roads;

[0045] Step b1: obtaining inbound vehicles and outbound vehicles of a mainstream road, and determining whether the mainstream road is a congested road according to the inbound vehicles and outbound vehicles;

[0046] Step b2: If not, determine that the mainstream road is a mainstream evacuation road; if so, obtain the mainstream road corresponding to the mainstream road;

[0047] Step b3: determine whether the mainstream road corresponding to the mainstream road is a congested road. If not, determine both the mainstream road and the mainstream road corresponding to the mainstream road as mainstream evacuation roads. If yes, repeat steps b2 and b3 until a mainstream evacuation road is generated.

[0048] A similar process of obtaining the mainstream evacuation road based on the mainstream road is adopted to obtain the diversion evacuation road corresponding to the diversion road;

[0049] Set duration threshold A and duration threshold B, where duration threshold A>duration threshold B;

[0050] Obtaining traffic light signal information of congested roads and evacuated roads; the traffic light signal information includes red light duration and green light duration;

[0051] The mainstream evacuation roads are sorted according to the road relationship to generate a main sequence number; the diversion evacuation roads are sorted according to the road relationship to generate a sub-sequence number;

[0052] For example, if there is a congested road A and its corresponding mainstream evacuation road A, mainstream evacuation road B, diversion evacuation road A and diversion evacuation road B, and mainstream evacuation road B is the mainstream road of mainstream evacuation road A, and diversion evacuation road B is the diversion road of diversion evacuation road A, the mainstream evacuation roads are sorted according to the road relationship, then the main sequence number of mainstream evacuation road A is 2, and the main sequence number of mainstream evacuation road B is 1; the diversion evacuation roads are sorted according to the road relationship, then the sub-sequence number of diversion evacuation road A is 2, and the sub-sequence number of diversion evacuation road B is 1;

[0053] Change the red light duration of the congested road to duration threshold A, and change the green light duration to duration threshold B; obtain the main sequence number corresponding to the mainstream evacuation road, and change the red light duration of the mainstream evacuation road to , change the green light duration of the main evacuation road to ; Get the sub-sequence number corresponding to the diversion evacuation road, and change the red light duration of the diversion evacuation road to , change the green light duration of the diversion evacuation road to ;

[0054] Furthermore, a monitoring-based vehicle-road cooperative device includes:

[0055] The data acquisition module sets an import camera at the road import end and an export camera at the road export end, obtains the import vehicles of the road through the import camera, and obtains the export vehicles of the road through the export camera;

[0056] The road data analysis module determines whether the road is congested based on the number of inbound and outbound vehicles. If so, the road is marked as a congested road; otherwise, the road is marked as a clear road.

[0057] The road data processing module obtains the import and export vehicles of the congested road, obtains the mainstream vehicle source according to the import vehicles, and obtains the diversion vehicle source according to the export vehicles; obtains the mainstream road corresponding to the congested road according to the mainstream vehicle source, and obtains the diversion road corresponding to the congested road according to the diversion vehicle source; obtains the mainstream road and diversion road corresponding to the congested road, obtains the evacuation road according to the mainstream road and the diversion road, obtains the traffic light signal information of the congested road and the evacuation road, and adjusts the traffic light signal information to transform the congested road into an unobstructed road.

[0058] Compared with the prior art, the beneficial effects of the present invention are as follows: the present invention obtains the mainstream evacuation roads and diversion evacuation roads of congested roads, obtains the traffic light signal information of the congested roads and their corresponding mainstream evacuation roads and diversion evacuation roads, adjusts the traffic light signal information, indirectly controls the direction of vehicles, and transforms congested roads into unobstructed roads. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] Figure 1 It is a flow chart of the present invention. DETAILED DESCRIPTION

[0060] like Figure 1 As shown, the present invention provides a monitoring-based vehicle-road collaboration method, comprising the following steps:

[0061] Step S1: an import camera is set at the road import end, and an export camera is set at the road export end, and the import vehicles of the road are acquired through the import camera, and the export vehicles of the road are acquired through the export camera;

[0062] Step S2: judging whether there is congestion on the road according to the number of inbound vehicles and outbound vehicles; if there is congestion, marking the road as a congested road; otherwise, marking the road as a clear road;

[0063] Step S3: Obtain the inbound vehicles and outbound vehicles of the congested road, obtain the mainstream vehicle source according to the inbound vehicles, and obtain the diversion vehicle source according to the outbound vehicles; obtain the mainstream road corresponding to the congested road according to the mainstream vehicle source, and obtain the diversion road corresponding to the congested road according to the diversion vehicle source;

[0064] Step S4: Obtain the mainstream road and diversion road corresponding to the congested road, obtain the evacuation road according to the mainstream road and the diversion road, obtain the traffic light signal information of the congested road and the evacuation road, and adjust the traffic light signal information to transform the congested road into a smooth road.

[0065] It should be further explained that, in the specific implementation process, the process of obtaining the import vehicles on the road through the import camera and obtaining the export vehicles on the road through the export camera includes:

[0066] The inlet camera and the outlet camera are both provided with a processor, and the video footage collected by the inlet camera and the outlet camera is processed by the processor;

[0067] Collect vehicle images under different climate and lighting conditions, mark them as sample images, generate a sample image set, extract features from the sample image set through a convolutional neural network, obtain vehicle recognition features, and then build a vehicle recognition model based on the vehicle recognition features;

[0068] Set the processing cycle;

[0069] The vehicle driving video at the road entrance is obtained through the import camera. According to the processing cycle, the obtained vehicle driving video is regularly sent to the processor, and the processor performs the following operations on the vehicle driving video:

[0070] Step a1: cutting the vehicle driving video into frames to obtain a number of vehicle driving images, setting effective areas in the number of vehicle driving images, and generating a vehicle driving image set in chronological order;

[0071] Step a2: obtaining the first vehicle driving image in the first vehicle driving image set, identifying the valid area in the image according to the vehicle recognition model, marking the identified vehicles as valid driving vehicles, obtaining the number of valid driving vehicles, and marking the number as the vehicle initial value;

[0072] Step a3: obtain the first vehicle driving image in the vehicle driving image set, mark it as a sample image, identify the valid area in the sample image according to the vehicle recognition model, mark the identified vehicle as a valid driving vehicle, obtain the individual features of the valid driving vehicle through a convolutional neural network, the individual features are unique, mark the individual features as initial vehicle features, and generate an initial vehicle feature set; obtain the number of valid driving vehicles, and mark the number as the vehicle initial value; obtain the second vehicle driving image in the vehicle driving image set, mark it as a comparison image, identify the valid area in the comparison image according to the vehicle recognition model, obtain the valid driving vehicle in the comparison image, and then obtain the individual features of the valid driving vehicle through a convolutional neural network, mark them as the latest vehicle features, and generate the latest vehicle feature set; compare the latest vehicle feature set with the initial vehicle feature set, delete the latest vehicle features in the latest vehicle feature set that are consistent with the initial vehicle features in the initial vehicle feature set, obtain the number of the latest vehicle features remaining in the latest vehicle feature set, and mark the number as the vehicle newly added value;

[0073] Step a3: Repeat step a2 to obtain several vehicle added values ​​in turn, then the number of imported vehicles on the road = , where n is a positive integer greater than 1;

[0074] The vehicle driving video at the road exit is obtained through the exit camera, and the obtained vehicle driving video is sent to the processor regularly according to the processing cycle; the exit vehicle of the road is obtained by using the steps similar to those for obtaining the road import vehicle. , where n is a positive integer greater than 1;

[0075] It should be further explained that, in the specific implementation process, the process of judging whether there is congestion on the road based on the number of imported and exported vehicles includes:

[0076] Set congestion period thresholds and congestion vehicle thresholds;

[0077] Obtain the import and export vehicles on the road in real time, and obtain the real-time capacity of the road based on the import and export vehicles on the road = import vehicles - export vehicles;

[0078] Obtain the duration of time when the real-time capacity of the road is greater than the congestion vehicle threshold, which is recorded as the congestion period;

[0079] If the congestion period is greater than the congestion period threshold, it is determined that the road is congested; otherwise, it is determined that the road is not congested;

[0080] It should be further explained that, in the specific implementation process, the process of obtaining the import vehicles and export vehicles of the congested road, obtaining the mainstream vehicle source according to the import vehicles, and obtaining the diversion vehicle source according to the export vehicles includes:

[0081] Divide the import vehicles into a plurality of guided import vehicles according to the lanes, wherein the guided import vehicles include left-turn import vehicles, straight-ahead import vehicles and right-turn import vehicles; divide the export vehicles into a plurality of guided export vehicles according to the lanes, wherein the guided export vehicles include left-turn export vehicles, straight-ahead export vehicles and right-turn export vehicles;

[0082] Set the threshold for the proportion of congested traffic;

[0083] Get the ratio of left-turn import vehicles = ; Get the proportion of direct imported vehicles = ; Get the ratio of right-turn import vehicles = ;

[0084] If the proportion of guided import vehicles is greater than the congested traffic proportion threshold, the guided import vehicle is marked as a mainstream vehicle source, otherwise, the guided import vehicle is not marked as a mainstream vehicle source;

[0085] If none of the guided import vehicles are marked as mainstream vehicle sources, all guided import vehicles are marked as mainstream vehicle sources;

[0086] That is, the proportion of left-turn import vehicles, the proportion of straight import vehicles, and the proportion of right-turn import vehicles are all less than or equal to the congested traffic proportion threshold, so that left-turn import vehicles, straight import vehicles, and right-turn import vehicles are not marked as mainstream vehicle sources, then left-turn import vehicles, straight import vehicles, and right-turn import vehicles are all marked as mainstream vehicle sources;

[0087] Get the ratio of vehicles turning left at the exit = , the proportion of vehicles going straight out of the gate = and the proportion of vehicles turning right at the exit = ;

[0088] If the proportion of the exit-directed vehicle is greater than the congested traffic proportion threshold, the exit-directed vehicle is marked as a diversion vehicle source, otherwise, the exit-directed vehicle is not marked as a diversion vehicle source;

[0089] If none of the vehicles leading to the exit are marked as diversion vehicle sources, all vehicles leading to the exit are marked as diversion vehicle sources;

[0090] That is, the proportion of left-turn exit vehicles, the proportion of straight-through exit vehicles, and the proportion of right-turn exit vehicles are all less than or equal to the congested traffic proportion threshold, so that the left-turn exit vehicles, the straight-through exit vehicles, and the right-turn exit vehicles are not marked as diversion vehicle sources, then the left-turn exit vehicles, the straight-through exit vehicles, and the right-turn exit vehicles are all marked as diversion vehicle sources;

[0091] It should be further explained that, in a specific implementation process, the process of obtaining the mainstream road corresponding to the congested road according to the mainstream vehicle source and obtaining the diversion road corresponding to the congested road according to the diversion vehicle source includes:

[0092] Acquire aggregated roads and discrete roads of congested roads; the aggregated roads include left aggregated roads, straight aggregated roads and right aggregated roads; the discrete roads include left discrete roads, straight discrete roads and right discrete roads;

[0093] The left convergent road means that vehicles can enter the congested road by turning left on the left convergent road; the straight convergent road means that vehicles can enter the congested road by going straight on the straight convergent road; the right convergent road means that vehicles can enter the congested road by turning right on the right convergent road;

[0094] The left discrete road is a road that a vehicle can enter by turning left on a congested road; the straight discrete road is a road that a vehicle can enter by going straight on a congested road; the right discrete road is a road that a vehicle can enter by turning right on a congested road;

[0095] The guided import vehicles and the convergent roads have a one-to-one correspondence, that is, the left-turning import vehicles correspond to the left convergent road, the straight-going import vehicles correspond to the straight convergent road, and the right-turning import vehicles correspond to the right convergent road;

[0096] The guided exit vehicles have a one-to-one correspondence with the discrete roads, that is, the left-turn exit vehicles correspond to the left discrete road, the straight-going exit vehicles correspond to the straight discrete road, and the right-turn exit vehicles correspond to the right discrete road;

[0097] Obtain the aggregated roads corresponding to the mainstream vehicle sources as the mainstream roads of the congested road; obtain the discrete roads corresponding to the diversion vehicle sources as the diversion roads of the congested road;

[0098] It should be further explained that, in the specific implementation process, the process of obtaining the mainstream road and the diversion road corresponding to the congested road, obtaining the evacuation road according to the mainstream road and the diversion road, obtaining the traffic light signal information of the congested road and the evacuation road, and adjusting the traffic light signal information to transform the congested road into a smooth road includes:

[0099] Obtaining mainstream roads and diversion roads corresponding to congested roads; the evacuation roads include mainstream evacuation roads and diversion evacuation roads;

[0100] Step b1: obtaining inbound vehicles and outbound vehicles of a mainstream road, and determining whether the mainstream road is a congested road according to the inbound vehicles and outbound vehicles;

[0101] Step b2: If not, determine that the mainstream road is a mainstream evacuation road; if so, obtain the mainstream road corresponding to the mainstream road;

[0102] Step b3: determine whether the mainstream road corresponding to the mainstream road is a congested road. If not, determine both the mainstream road and the mainstream road corresponding to the mainstream road as mainstream evacuation roads. If yes, repeat steps b2 and b3 until a mainstream evacuation road is generated.

[0103] A similar process of obtaining the mainstream evacuation road based on the mainstream road is adopted to obtain the diversion evacuation road corresponding to the diversion road;

[0104] Set duration threshold A and duration threshold B, where duration threshold A>duration threshold B;

[0105] Obtaining traffic light signal information of congested roads and evacuated roads; the traffic light signal information includes red light duration and green light duration;

[0106] The mainstream evacuation roads are sorted according to the road relationship to generate a main sequence number; the diversion evacuation roads are sorted according to the road relationship to generate a sub-sequence number;

[0107] For example, if there is a congested road A and its corresponding mainstream evacuation road A, mainstream evacuation road B, diversion evacuation road A and diversion evacuation road B, and mainstream evacuation road B is the mainstream road of mainstream evacuation road A, and diversion evacuation road B is the diversion road of diversion evacuation road A, the mainstream evacuation roads are sorted according to the road relationship, then the main sequence number of mainstream evacuation road A is 2, and the main sequence number of mainstream evacuation road B is 1; the diversion evacuation roads are sorted according to the road relationship, then the sub-sequence number of diversion evacuation road A is 2, and the sub-sequence number of diversion evacuation road B is 1;

[0108] Change the red light duration of the congested road to duration threshold A, and change the green light duration to duration threshold B; obtain the main sequence number corresponding to the mainstream evacuation road, and change the red light duration of the mainstream evacuation road to , change the green light duration of the main evacuation road to ; Get the sub-sequence number corresponding to the diversion evacuation road, and change the red light duration of the diversion evacuation road to , change the green light duration of the diversion evacuation road to ;

[0109] The present invention also provides a monitoring-based vehicle-road collaboration method device, the device comprising:

[0110] The data acquisition module sets an import camera at the road import end and an export camera at the road export end, obtains the import vehicles of the road through the import camera, and obtains the export vehicles of the road through the export camera;

[0111] The road data analysis module determines whether the road is congested based on the number of inbound and outbound vehicles. If so, the road is marked as a congested road; otherwise, the road is marked as a clear road.

[0112] A road data processing module obtains inbound and outbound vehicles of a congested road, obtains a mainstream vehicle source based on the inbound vehicles, and obtains a diversion vehicle source based on the outbound vehicles; obtains a mainstream road corresponding to the congested road based on the mainstream vehicle source, and obtains a diversion road corresponding to the congested road based on the diversion vehicle source; obtains a mainstream road and a diversion road corresponding to the congested road, obtains an evacuation road based on the mainstream road and the diversion road, obtains traffic light signal information of the congested road and the evacuation road, and adjusts the traffic light signal information to transform the congested road into an unobstructed road;

[0113] The above embodiments are only used to illustrate the technical method of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present invention.

Claims

1. A monitoring-based vehicle-road collaboration method. It is characterized in that The following steps are involved: Step S1: an import camera is set at the road import end, and an export camera is set at the road export end, and the import vehicles of the road are acquired through the import camera, and the export vehicles of the road are acquired through the export camera; Step S2: judging whether there is congestion on the road according to the number of inbound vehicles and outbound vehicles; if there is congestion, marking the road as a congested road; otherwise, marking the road as a clear road; Step S3: Obtain the inbound vehicles and outbound vehicles of the congested road, obtain the mainstream vehicle source according to the inbound vehicles, and obtain the diversion vehicle source according to the outbound vehicles; obtain the mainstream road corresponding to the congested road according to the mainstream vehicle source, and obtain the diversion road corresponding to the congested road according to the diversion vehicle source; Step S4: Obtaining the mainstream road and diversion road corresponding to the congested road, obtaining the evacuation road according to the mainstream road and the diversion road, obtaining the traffic light signal information of the congested road and the evacuation road, and adjusting the traffic light signal information to transform the congested road into a smooth road; The process of obtaining an evacuation road according to the mainstream road and the diversion road, obtaining traffic light signal information of the congested road and the evacuation road, and adjusting the traffic light signal information includes: Acquire evacuation roads according to mainstream roads and diversion roads, wherein the evacuation roads include mainstream evacuation roads and diversion evacuation roads; acquire traffic light signal information of congested roads and evacuation roads, wherein the traffic light signal information includes red light duration and green light duration; set duration threshold A and duration threshold B, change the red light duration of the congested road to duration threshold A, and change the green light duration to duration threshold B; acquire the main sequence number of the mainstream evacuation road, and acquire the sub-sequence number of the diversion evacuation road; generate a new red light duration according to the main sequence number and duration threshold A, and Generate a new green light duration according to the main sequence number and duration threshold B, and use the new red light duration to cover the original red light duration of the mainstream evacuation road, and use the new green light duration to cover the original green light duration of the mainstream evacuation road; generate a new red light duration according to the sub-sequence number and duration threshold B, and generate a new green light duration according to the sub-sequence number and duration threshold A, and use the new red light duration to cover the original red light duration of the diversion evacuation road, and use the new green light duration to cover the original green light duration of the diversion evacuation road, and adjust the traffic light signal information to transform the congested road into a smooth road; The process of obtaining the import vehicles on the road through the import camera and the export vehicles on the road through the export camera includes: The inlet camera and the outlet camera are both provided with a processor; Collect vehicle images under different climate and lighting conditions, mark them as sample images, generate a sample image set, obtain vehicle recognition features based on the sample image set, and then build a vehicle recognition model based on the vehicle recognition features; Set the processing cycle; The vehicle driving video at the road entrance is obtained through the import camera. According to the processing cycle, the obtained vehicle driving video is sent to the processor regularly to perform the following operations: Step a1: cutting the vehicle driving video into frames to obtain a number of vehicle driving images, setting effective areas in the number of vehicle driving images, and generating a vehicle driving image set in chronological order; Step a2: obtaining the first vehicle driving image in the first vehicle driving image set, identifying the valid area in the image according to the vehicle recognition model, obtaining valid driving vehicles, obtaining the number of valid driving vehicles, and marking the number as the vehicle initial value; Step a3: obtain the first vehicle driving image in the vehicle driving image set, mark it as a sample image, obtain valid driving vehicles in the sample image according to the vehicle recognition model, obtain individual features of the valid driving vehicles, mark them as initial vehicle features, generate an initial vehicle feature set, and obtain the number of valid driving vehicles, and mark the number as the vehicle initial value; obtain the second vehicle driving image in the vehicle driving image set, mark it as a comparison image, obtain valid driving vehicles in the comparison image according to the vehicle recognition model, then obtain individual features of the valid driving vehicles, mark them as latest vehicle features, and generate an latest vehicle feature set; compare the latest vehicle feature set with the initial vehicle feature set, delete the latest vehicle features in the latest vehicle feature set that are consistent with the initial vehicle features in the initial vehicle feature set, obtain the number of the latest vehicle features remaining in the latest vehicle feature set, and mark the number as the vehicle newly added value; Step a3: Repeat step a2 to obtain several vehicle added values ​​in turn, then the number of imported vehicles on the road = , where n is a positive integer greater than 1; The vehicle driving video at the road exit is obtained through the exit camera, and the obtained vehicle driving video is sent to the processor regularly according to the processing cycle; the exit vehicles of the road are obtained by adopting the steps similar to those for obtaining the vehicles entering the road; The process of determining whether a road is congested based on inbound and outbound vehicles includes: Set congestion period thresholds and congestion vehicle thresholds; Obtain the import and export vehicles on the road in real time, and obtain the real-time capacity of the road based on the import and export vehicles on the road = import vehicles - export vehicles; Obtain the duration of time when the real-time capacity of the road is greater than the congestion vehicle threshold, which is recorded as the congestion period; If the congestion period is greater than the congestion period threshold, it is determined that the road is congested; otherwise, it is determined that the road is not congested; The process of obtaining the import vehicles and export vehicles of the congested road, obtaining the mainstream vehicle source according to the import vehicles, and obtaining the diversion vehicle source according to the export vehicles includes: Dividing the imported vehicles into a plurality of guided imported vehicles according to the lanes, wherein the guided imported vehicles include left-turning imported vehicles, straight-going imported vehicles and right-turning imported vehicles; Dividing the exit vehicles into a plurality of guided exit vehicles according to the lanes, wherein the guided exit vehicles include left-turn exit vehicles, straight-ahead exit vehicles and right-turn exit vehicles; Set the threshold for the proportion of congested traffic; Obtain the ratio of left-turning imported vehicles, the ratio of straight-going imported vehicles, and the ratio of right-turning imported vehicles; If the proportion of guided import vehicles is greater than the congested traffic proportion threshold, the guided import vehicle is marked as a mainstream vehicle source, otherwise, the guided import vehicle is not marked as a mainstream vehicle source; If none of the guided import vehicles are marked as mainstream vehicle sources, all guided import vehicles are marked as mainstream vehicle sources; Obtain the ratio of vehicles turning left, vehicles going straight and vehicles turning right; If the proportion of the exit-directed vehicle is greater than the congested traffic proportion threshold, the exit-directed vehicle is marked as a diversion vehicle source, otherwise, the exit-directed vehicle is not marked as a diversion vehicle source; If none of the vehicles leading to the exit are marked as diversion vehicle sources, all vehicles leading to the exit are marked as diversion vehicle sources; The process of obtaining the mainstream road corresponding to the congested road according to the mainstream vehicle source and obtaining the diversion road corresponding to the congested road according to the diversion vehicle source includes: Get the aggregated roads and discrete roads of congested roads; The converged roads include a left converged road, a straight converged road and a right converged road; The discrete roads include a left discrete road, a straight discrete road and a right discrete road; Left-turning import vehicles correspond to the left convergent road, straight-going import vehicles correspond to the straight convergent road, and right-turning import vehicles correspond to the right convergent road; The left-turn exit vehicles correspond to the left discrete road, the straight-going exit vehicles correspond to the straight discrete road, and the right-turn exit vehicles correspond to the right discrete road; Obtain the aggregated roads corresponding to the mainstream vehicle sources as the mainstream roads of the congested road; obtain the discrete roads corresponding to the diversion vehicle sources as the diversion roads of the congested road; The process of obtaining the mainstream road and the diversion road corresponding to the congested road, obtaining the evacuation road according to the mainstream road and the diversion road, obtaining the traffic light signal information of the congested road and the evacuation road, and adjusting the traffic light signal information includes: Obtaining mainstream roads and diversion roads corresponding to congested roads; the evacuation roads include mainstream evacuation roads and diversion evacuation roads; Step b1: obtaining inbound vehicles and outbound vehicles of a mainstream road, and determining whether the mainstream road is a congested road according to the inbound vehicles and outbound vehicles; Step b2: If not, determine that the mainstream road is a mainstream evacuation road; if so, obtain the mainstream road corresponding to the mainstream road; Step b3: determine whether the mainstream road corresponding to the mainstream road is a congested road. If not, determine both the mainstream road and the mainstream road corresponding to the mainstream road as mainstream evacuation roads. If yes, repeat steps b2 and b3 until a mainstream evacuation road is generated. Obtain the diversion evacuation road corresponding to the diversion road by adopting the similar steps of obtaining the mainstream evacuation road according to the mainstream road; Set duration threshold A and duration threshold B, where duration threshold A>duration threshold B; Obtaining traffic light signal information of congested roads and evacuated roads; the traffic light signal information includes red light duration and green light duration; The mainstream evacuation roads are sorted according to the road relationship to generate a main sequence number; the diversion evacuation roads are sorted according to the road relationship to generate a sub-sequence number; Change the red light duration of the congested road to duration threshold A, and change the green light duration to duration threshold B; obtain the main sequence number corresponding to the mainstream evacuation road, and change the red light duration of the mainstream evacuation road to , change the green light duration of the main evacuation road to ; Get the sub-sequence number corresponding to the diversion evacuation road, and change the red light duration of the diversion evacuation road to , change the green light duration of the diversion evacuation road to .

2. A monitoring-based vehicle-road cooperative device, used to implement the vehicle-road cooperative method according to claim 1, It is characterized in that The device includes: The data acquisition module sets an import camera at the road import end and an export camera at the road export end, obtains the import vehicles of the road through the import camera, and obtains the export vehicles of the road through the export camera; The road data analysis module determines whether the road is congested based on the number of inbound and outbound vehicles. If so, the road is marked as a congested road; otherwise, the road is marked as a clear road. The road data processing module obtains the import and export vehicles of the congested road, obtains the mainstream vehicle source according to the import vehicles, and obtains the diversion vehicle source according to the export vehicles; obtains the mainstream road corresponding to the congested road according to the mainstream vehicle source, and obtains the diversion road corresponding to the congested road according to the diversion vehicle source; obtains the mainstream road and diversion road corresponding to the congested road, obtains the evacuation road according to the mainstream road and the diversion road, obtains the traffic light signal information of the congested road and the evacuation road, and adjusts the traffic light signal information to transform the congested road into an unobstructed road.

Citation Information

Patent Citations

  • System for intelligently evacuating traffic congestion

    CN110390824A

  • Traffic jam processing method, device, equipment and medium

    CN114023084A

  • Internet-of-things traffic monitoring system in intelligent traffic system and use method of Internet-of-things traffic monitoring system

    CN114170795A

  • Urban traffic jam optimization method and system

    CN116959275A

  • Road real-time monitoring method, system and equipment based on unmanned aerial vehicle and medium

    CN118262508A