A real-time traffic assessment system based on image data analysis and processing

Abstract

The present invention discloses a real-time traffic assessment system based on image data analysis and processing, relates to the technical field of traffic assessment, and solves the technical problem in the prior art that shared roads cannot be monitored through image acquisition, which reduces the timeliness of road diversion and path division. Specifically, a network is built according to all the intricately connected roads inside and marked as a passable road network, single routes are divided according to various destinations outside the passable road network, and overlapping areas are marked as shared roads when each single route is in the passable road network at the same time, and at the same time, traffic flow pictures are collected on the shared roads during the continuous increase of traffic flow in the assessment area; shared road monitoring information is obtained by collecting traffic flow pictures, a traffic monitoring coefficient is calculated according to a formula, and whether shared road monitoring is abnormal is inferred according to threshold comparison.

Description

Technical Field

[0001] The present invention relates to the technical field of traffic assessment, and in particular to a real-time traffic assessment system based on image data analysis and processing. Background Art

[0002] Traffic assessment is the process of comprehensively analyzing and evaluating the traffic conditions of a specific area or project; the following are some important aspects of traffic assessment: determining traffic demand; optimizing the traffic system: evaluating the operating efficiency of existing traffic facilities, identifying bottlenecks and problems, and proposing improvement measures to improve the overall performance of the traffic system; ensuring traffic safety: analyzing traffic risks and safety hazards, proposing safety measures, and reducing the incidence of traffic accidents.

[0003] However, in the prior art, it is impossible to identify and monitor the traffic assessment road area through image acquisition and analysis during traffic assessment, so that real-time analysis of different types of communication entities cannot be performed; at the same time, it is impossible to track and analyze the motor vehicles among the passing entities after different types of passing entities in the assessment area have no traffic impact, and it is impossible to predict congestion risks in a timely manner; in addition, it is impossible to monitor shared roads through image acquisition, which reduces the timeliness of road diversion and path division, and cannot meet the needs of road traffic assessment.

[0004] In view of the above technical defects, a solution is now proposed. Summary of the invention

[0005] The purpose of the present invention is to solve the above-mentioned problems and to provide a real-time traffic assessment system based on image data analysis and processing.

[0006] The purpose of the present invention can be achieved through the following technical solutions:

[0007] A real-time traffic assessment system based on image data analysis and processing includes a real-time traffic assessment platform, and the real-time traffic assessment platform is connected to:

[0008] An image recognition monitoring unit is used to mark a road area where real-time traffic assessment is to be performed as an assessment area, collect road images in the assessment area and divide the collected images into a number of sub-frame images, and if there is no passing subject in the sub-frame image, the sub-frame image of the type with the highest definition is used as the background frame, and the sub-frame images with passing subjects are analyzed for the passing subjects, and the types are divided according to the passing subject analysis, and it is inferred whether there is a risk of passing in the assessment area according to the analysis of different types of passing subjects;

[0009] The monitoring, tracking and analysis unit is used to track and analyze the traffic entities in the assessment area, obtain tracking frequency data and tracking span data, and infer whether there is a congestion risk for the traffic entities in the assessment area based on data comparison;

[0010] The shared road monitoring unit is used to build a network based on all the intricately connected roads within the network and mark them as a passable road network, divide the single routes according to the destinations outside the passable road network, and mark the overlapping areas as shared roads when the single routes are in the passable road network at the same time. At the same time, when the traffic volume in the assessment area continues to increase, the shared roads are used to collect traffic pictures; the shared road monitoring information is obtained by collecting traffic pictures, the traffic monitoring coefficient is calculated according to the formula, and the shared road monitoring is inferred to be abnormal based on the threshold comparison.

[0011] As a preferred embodiment of the present invention, the process of classifying the types of traffic entities is as follows:

[0012] The subframe images with passing subjects are marked as passing analysis images, the moving trajectory of the passing subjects in the passing analysis images and the changed part of the moving trajectory at the adjacent positioning node positions in the corresponding moving trajectory are obtained, and trajectory analysis is performed:

[0013] If the horizontal angle floating span between the moving trajectory of the passing subject and the roadside trajectories on both sides of the road in the forward direction exceeds the horizontal angle floating span, it is inferred that the passing subject is a fixed-trajectory moving subject; if the horizontal angle floating span between the moving trajectory of the passing subject and the roadside trajectories on both sides of the road in the forward direction does not exceed the horizontal angle floating span, it is inferred that the passing subject is a variable-trajectory moving subject;

[0014] If the change part of the mobile trajectory at the adjacent positioning node positions in the corresponding mobile trajectory includes the real-time positioning node positions of the passing subject alternating back and forth, it is inferred that the passing subject is a multi-moving mobile subject; otherwise, the change part of the mobile trajectory at the adjacent positioning node positions in the corresponding mobile trajectory does not include the real-time positioning node positions of the passing subject alternating back and forth, then it is inferred that the passing subject is a less-moving mobile subject.

[0015] As a preferred embodiment of the present invention, trajectory overlapping data and trajectory handover data are collected, and the trajectory overlapping data and trajectory handover data are respectively the maximum overlapping area of ​​the area where the motor vehicle trajectory and the non-motor vehicle trajectory are located in the corresponding time period of the traffic analysis picture, and the duration of the intersection point between the motor vehicle trajectory and the non-motor vehicle trajectory in the corresponding time period of the traffic analysis picture.

[0016] As a preferred embodiment of the present invention, if the trajectory overlapping data exceeds the maximum value threshold of the overlapping area at the same moment, or the trajectory handover data exceeds the duration threshold of the handover point, an image recognition risk signal is generated; if the trajectory overlapping data does not exceed the maximum value threshold of the overlapping area at the same moment, and the trajectory handover data does not exceed the duration threshold of the handover point, an image recognition safety signal is generated.

[0017] As a preferred embodiment of the present invention, the tracking frequency data and the tracking span data are respectively the frequency ratio of the continuous overtaking frequency of the same type of passing subjects on the left side of the passing subject in the road area without speeding during the driving process of the passing subject in the evaluation area and the speed floating frequency of the overtaken passing subject, and the cumulative sum of the corresponding span values ​​of the increased span value of the continuous driving section of the passing subject in the evaluation area and the reduced span value of the overtaking passing subject's overtaking speed.

[0018] As a preferred embodiment of the present invention, if the tracking frequency data exceeds the frequency ratio threshold, or the tracking span data exceeds the span value cumulative sum threshold, a high risk signal for traffic congestion is generated; if the tracking frequency data does not exceed the frequency ratio threshold, and the tracking span data does not exceed the span value cumulative sum threshold, a low risk signal for traffic congestion is generated.

[0019] As a preferred embodiment of the present invention, the operation process of the shared road monitoring unit is as follows: the traffic entities traveling on the shared road are profiled and the traffic entities are set as single points when the profiles all belong to motor vehicles, and the traffic trajectory lines are constructed based on several single points in the collected traffic flow picture; and the forks in the shared roads are set as monitoring points.

[0020] As a preferred embodiment of the present invention, the shared road monitoring information includes the continuously increasing value of the number of single points in the lane in the opposite direction of the fork in the traffic trajectory line of any monitoring point in the shared road in the evaluation area that change their position in the direction of the fork, the continuous span of the vehicle average speed deviation value of the traffic trajectory lines at both ends of the current traffic trajectory line with the current fork direction as the midpoint, and the traffic flow value ratio corresponding to the traffic obstruction amount of the diversion obstruction line at the same time in the entire traffic trajectory line and the increased value of the number of passing entities in the current traffic trajectory line that drive out of the evaluation area at the corresponding time.

[0021] As a preferred embodiment of the present invention, if the traffic monitoring coefficient of the shared road corresponding to the traffic trajectory line in the evaluation area exceeds the traffic monitoring coefficient threshold, it is inferred that the shared road monitoring in the current evaluation area is abnormal; if the traffic monitoring coefficient of the shared road corresponding to the traffic trajectory line in the evaluation area does not exceed the traffic monitoring coefficient threshold, it is inferred that the shared road monitoring in the current evaluation area is normal.

[0022] Compared with the prior art, the present invention has the following beneficial effects:

[0023] 1. In the present invention, the road area for traffic assessment is identified and monitored through image acquisition and analysis, and the real-time traffic process of the passing subjects in the current area is inferred. According to the real-time monitoring of the passing subjects in the area, it is inferred whether the current road traffic environment meets the demand. It can also infer the real-time development trend of the traffic conditions in the current road area according to the real-time traffic process, thereby improving the efficiency of real-time traffic assessment;

[0024] Track and analyze the traffic entities in the assessment area. After different types of traffic entities have no traffic impact in the assessment area, track and analyze the motor vehicles among the traffic entities. Based on the real-time motor vehicle traffic analysis, it is inferred whether there are congestion risks for the traffic entities in the current assessment area, thereby improving the traffic monitoring efficiency of the traffic entities; motor vehicle traffic entities are taken as the object.

[0025] 2. In the present invention, shared road images are collected in the evaluation area and the shared roads are monitored based on the collected images, wherein the shared roads are roads connecting multiple destinations in the evaluation area, such as viaducts and other types of roads in the prior art; through this type of road, it is inferred whether there is a road traffic impact when the current traffic entities in different destination directions are driving, so that the original traffic supply performance of the shared road is reduced, and it is impossible to divert and divide the paths in time, resulting in a congestion trend in real time traffic;

[0026] Dynamic exchange analysis is performed on the road network. By analyzing the traffic conditions of traffic entities inside and outside the road network, it is inferred whether the dynamic exchange analysis of the current road network is qualified, so as to facilitate timely traffic control based on the dynamic exchange analysis and improve the accuracy of traffic condition assessment. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] In order to facilitate understanding by those skilled in the art, the present invention is further described below with reference to the accompanying drawings.

[0028] Figure 1 This is a principle block diagram of Embodiment 1 of the present invention;

[0029] Figure 2 This is a principle block diagram of the second embodiment of the present invention. DETAILED DESCRIPTION

[0030] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0031] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present invention. The appearance of the phrase in various places in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0032] Embodiment 1

[0033] This embodiment uses image data collection and analysis to conduct real-time assessment of road traffic to ensure the accuracy and timeliness of road traffic control. Figure 1 As shown, a real-time traffic assessment system based on image data analysis and processing includes a real-time traffic assessment platform, wherein the real-time traffic assessment platform is communicatively connected to an image recognition monitoring unit, a monitoring tracking analysis unit, and a shared road monitoring unit;

[0034] The real-time traffic assessment platform generates an image recognition monitoring signal and sends it to the image recognition monitoring unit. After receiving the image recognition monitoring signal, the image recognition monitoring unit identifies and monitors the traffic assessment road area through image acquisition and analysis, and infers the real-time traffic process of the passing subjects in the current area, where the passing subjects include people, motor vehicles, non-motor vehicles, etc.; based on the real-time monitoring of the passing subject process in the area, it can infer whether the current road traffic environment meets the demand, and it can also infer the real-time development trend of the traffic conditions in the current road area based on the real-time traffic process, thereby improving the efficiency of real-time traffic assessment;

[0035] The road area where the real-time traffic assessment is to be carried out is marked as the assessment area, and road images are collected in the assessment area and the collected images are divided into a number of sub-frame images. If there is no passing subject in the sub-frame image, the sub-frame image with the highest definition of this type is used as the background frame, and the sub-frame image with passing subjects is analyzed for the passing subjects;

[0036] The subframe images with passing subjects are marked as passing analysis images, the moving trajectory of the passing subjects in the passing analysis images and the changed part of the moving trajectory at the adjacent positioning node positions in the corresponding moving trajectory are obtained, and trajectory analysis is performed:

[0037] If the horizontal angle floating span between the moving trajectory of the passing subject and the trajectory of the roadside on both sides of the road's forward direction exceeds the horizontal angle floating span, it is inferred that the passing subject is a fixed-trajectory moving subject; that is, if it is a motor vehicle, the passing trajectory fits the road trajectory and cannot change at a large angle; if the horizontal angle floating span between the moving trajectory of the passing subject and the trajectory of the roadside on both sides of the road's forward direction does not exceed the horizontal angle floating span, it is inferred that the passing subject is a variable-trajectory moving subject; that is, it is a non-motor vehicle; the horizontal angle is expressed as the horizontal angle between the two sides of the passing subject in the moving trajectory and the corresponding roadside on both sides of the road after infinite extension;

[0038] If the change part of the moving trajectory at the adjacent positioning node position in the corresponding moving trajectory includes the real-time positioning node position of the passing subject alternating back and forth, it is inferred that the passing subject is a multi-moving moving subject, that is, a pedestrian, and the corresponding moving trajectory can move forward and backward at any time; on the contrary, if the change part of the moving trajectory at the adjacent positioning node position in the corresponding moving trajectory does not include the real-time positioning node position of the passing subject alternating back and forth, it is inferred that the passing subject is a less-moving moving subject, that is, a non-motor vehicle, and the horizontal angle of the corresponding moving trajectory can change with the rotation of the vehicle head, but it still needs to move according to the direction of the road passage route;

[0039] The maximum value of the overlapping area of ​​the area where the motor vehicle trajectory and the non-motor vehicle trajectory are located in the corresponding time period of the traffic analysis picture is obtained, and the duration of the intersection point of the motor vehicle trajectory and the non-motor vehicle trajectory in the corresponding time period of the traffic analysis picture is obtained. The maximum value of the overlapping area of ​​the area where the motor vehicle trajectory and the non-motor vehicle trajectory are located in the corresponding time period of the traffic analysis picture and the duration of the intersection point of the motor vehicle trajectory and the non-motor vehicle trajectory in the corresponding time period of the traffic analysis picture are marked as trajectory overlapping data and trajectory intersection data, respectively, and compared with the maximum value threshold of the overlapping area at the same time and the duration threshold of the intersection point respectively:

[0040] If the maximum value of the simultaneous overlapping area of ​​the area where the motor vehicle passage trajectory and the non-motor vehicle passage trajectory are located in the corresponding time period of the traffic analysis image exceeds the maximum value threshold of the simultaneous overlapping area, or the duration of the intersection point between the motor vehicle passage trajectory and the non-motor vehicle passage trajectory in the corresponding time period of the traffic analysis image exceeds the intersection duration threshold, it is inferred that the corresponding overlap of different types of traffic entities in the current assessment area has a large impact, which is likely to increase the risk of road congestion in the area, and an image recognition risk signal is generated and sent to the real-time traffic assessment platform. After receiving the image recognition risk signal, the real-time traffic assessment platform performs traffic control on the assessment area corresponding to the image recognition risk signal;

[0041] If the maximum value of the simultaneous overlapping area of ​​the area where the motor vehicle trajectory and the non-motor vehicle trajectory are located in the corresponding time period of the traffic analysis image does not exceed the maximum value threshold of the simultaneous overlapping area, and the duration of the intersection point between the motor vehicle trajectory and the non-motor vehicle trajectory in the corresponding time period of the traffic analysis image does not exceed the intersection point duration threshold, it is inferred that the corresponding overlapping impact of different types of traffic entities in the current assessment area is small, and an image recognition safety signal is generated and sent to the real-time traffic assessment platform;

[0042] At the same time, a monitoring tracking and analysis signal is generated and sent to the monitoring tracking and analysis unit. After receiving the monitoring tracking and analysis signal, the monitoring tracking and analysis unit tracks and analyzes the passing subjects in the assessment area. After different types of passing subjects in the assessment area have no traffic impact, the motor vehicles in the passing subjects are tracked and analyzed. According to the real-time motor vehicle traffic analysis, it is inferred whether there is a congestion risk in the passing of the passing subjects in the current assessment area, thereby improving the traffic monitoring efficiency of the passing subjects; the motor vehicle passing subjects are taken as the object;

[0043] The frequency ratio of the frequency of continuous overtaking on the left side of the passing subject without speeding and the speed floating frequency of the passing subject being overtaken is obtained during the driving process of the passing subject in the assessment area. The continuous overtaking of the passing subject on the premise of not exceeding the speed indicates that there are low-speed vehicles on the road, so there is a risk of traffic jam. At the same time, the cumulative sum of the corresponding span values ​​of the increased span value of the continuous driving section of the passing subject being overtaken in the assessment area and the reduced span value of the overtaking speed of the overtaking passing subject is obtained. It can be understood that the increase in the number of vehicles on the road where the low-speed passing subject is located indicates the risk of congestion. If the risk increases and the speed of the overtaking subject decreases, it means that the road is congested and the overtaking speed continues to decrease; during the driving process of the subjects in the assessment area, the subjects of the same type are in the road area, and the frequency ratio of the continuous overtaking frequency of the subjects on the left side without speeding to the speed fluctuation frequency of the overtaken subjects, and the cumulative sum of the span values ​​of the increase in traffic volume on the road section where the overtaken subjects continue to drive in the assessment area and the reduced span value of the overtaking speed of the overtaking subjects are marked as tracking frequency data and tracking span data, and compared with the frequency ratio threshold and the span value cumulative sum threshold respectively:

[0044] If the same type of passing subjects are in the road area during the driving process of the passing subject in the assessment area, and the frequency ratio of the continuous overtaking frequency of the passing subject on the left side and the speed fluctuation frequency of the passing subject being overtaken without speeding exceeds the frequency ratio threshold, or the cumulative sum of the span values ​​of the traffic flow increase span value of the continuous driving section of the passing subject being overtaken in the assessment area and the reduced span value of the overtaking passing subject's overtaking speed exceeds the cumulative sum of span values ​​threshold, it is inferred that the passing subject in the assessment area has a congestion risk, and a high-risk driving congestion signal is generated and sent to the real-time traffic assessment platform. After receiving the high-risk driving congestion signal, the real-time traffic assessment platform uses traffic control to drive the vehicle at the specified speed of the lane, and broadcasts the traffic flow fluctuation report of the intersection to the road passing subject when the traffic flow increases;

[0045] If the same type of passing subjects are in the road area during the driving process of the passing subject in the assessment area, and the frequency ratio of the continuous overtaking frequency of the passing subject on the left side and the speed fluctuation frequency of the passing subject being overtaken does not exceed the frequency ratio threshold value without speeding, and the cumulative sum of the span values ​​corresponding to the span value of the traffic volume increase of the continuous driving section of the passing subject being overtaken in the assessment area and the span value of the reduced speed of the overtaking passing subject does not exceed the cumulative sum threshold value of the span value, it is inferred that there is no congestion risk for the passing subject in the assessment area, and a low risk signal for driving congestion is generated and sent to the real-time traffic assessment platform;

[0046] At the same time, a shared road monitoring signal is generated and sent to a shared road monitoring unit. After receiving the shared road monitoring signal, the shared road monitoring unit collects shared road images of the evaluation area and monitors the shared road based on the collected images, wherein the shared road is represented by a road connecting multiple destinations in the evaluation area, such as viaducts and other types of roads in the prior art; through this type of road, it is inferred whether there is a road traffic impact when the current traffic entities in different destination directions are driving, thereby reducing the original traffic supply performance of the shared road, and cannot be diverted and routed in time, resulting in real-time traffic congestion;

[0047] A network is built based on all the intricately connected roads and marked as a passable road network. Single routes are divided according to each destination outside the passable road network, and the overlapping areas of each single route in the passable road network are marked as shared roads. At the same time, when the traffic volume in the assessment area continues to increase, traffic flow images of shared roads are collected;

[0048] Perform profile analysis on the traffic entities on the shared road and set the traffic entities as single points when the profiles all belong to motor vehicles. Construct traffic trajectory lines based on several single points in the collected traffic flow images. Set the forks in the shared road as monitoring points.

[0049] Obtain the continuously increasing value of the number of single points in the lane in the opposite direction of the fork of the traffic trajectory line of any monitoring point in the shared road of the assessment area that change their positions in the direction of the fork, and the continuous span of the deviation value of the average speed of vehicles corresponding to the traffic trajectory lines at both ends of the current traffic trajectory line with the current fork direction as the midpoint, and mark the continuously increasing value of the number of single points in the lane in the opposite direction of the fork of the traffic trajectory line of any monitoring point in the shared road of the assessment area that change their positions in the direction of the fork, and the continuous span of the deviation value of the average speed of vehicles corresponding to the traffic trajectory lines at both ends of the current traffic trajectory line with the current fork direction as the midpoint as SLZ and CXZ respectively;

[0050] When there is a lane change at a fork in the shared road, the lane change route of the lane-changing subject at the fork is marked as a diversion barrier line, and the traffic flow barrier amount of the diversion barrier line at the same time in the entire traffic flow trajectory line and the traffic flow value ratio corresponding to the increase in the number of traffic subjects that have left the evaluation area in the current traffic flow trajectory line at the corresponding time are obtained, and the traffic flow barrier amount of the diversion barrier line at the same time in the entire traffic flow trajectory line and the traffic flow value ratio corresponding to the increase in the number of traffic subjects that have left the evaluation area in the current traffic flow trajectory line at the corresponding time are marked as CLB;

[0051] The above collected data are uniformly marked as shared road monitoring information, and substituted into the formula to obtain the traffic monitoring coefficient of the corresponding vehicle trajectory line of the shared road in the evaluation area, where the formula is:

[0052] , G is the traffic monitoring coefficient, fvb1, fvb2, fvb3 are the preset proportional coefficients of the continuous increase value of the number of single points, the continuous span of the average speed deviation value of vehicles, and the traffic flow value ratio, e is a natural constant, and α is an error correction factor with a value of 0.87;

[0053] Compare the traffic monitoring coefficient of the corresponding traffic trajectory line of the shared road in the assessment area with the traffic monitoring coefficient threshold:

[0054] If the traffic monitoring coefficient of the corresponding vehicle flow trajectory line of the shared road in the assessment area exceeds the traffic monitoring coefficient threshold, it is inferred that the shared road monitoring in the current assessment area is abnormal, and a shared road monitoring abnormality signal is generated and sent to the real-time traffic assessment platform. After receiving the shared road monitoring abnormality signal, the real-time traffic assessment platform analyzes the shared roads in the assessment area. If the vehicle flow trajectory line in the current shared road corresponds to multiple destinations and the number exceeds the set threshold, targeted diversion is performed according to the geographical location of the corresponding destination and the traffic route except the current road; at the same time, the shortest distance between adjacent diversion obstacle lines in the shared road is controlled, and the distance is extended to increase the traffic buffer time;

[0055] If the traffic monitoring coefficient of the corresponding vehicle flow trajectory line of the shared road in the assessment area does not exceed the traffic monitoring coefficient threshold, it is inferred that the shared road monitoring in the current assessment area is normal, and a shared road monitoring normal signal is generated and sent to the real-time traffic assessment platform;

[0056] Embodiment 2

[0057] The previous embodiment collects and analyzes images within the assessment area to infer whether the real-time road traffic conditions in the assessment area are qualified. Based on the previous embodiment, this embodiment adds a dynamic network analysis outside the entire assessment area. Please refer to Figure 2As shown, a network dynamic exchange analysis unit is connected between the shared road monitoring unit and the real-time traffic assessment platform;

[0058] After the shared road monitoring unit generates a shared road monitoring normal signal, it generates a network dynamic exchange analysis signal and sends the network dynamic exchange analysis signal to the network dynamic exchange analysis unit. After receiving the network dynamic exchange analysis signal, the network dynamic exchange analysis unit performs a dynamic exchange analysis on the traffic road network, and analyzes the traffic conditions of traffic entities inside and outside the traffic road network to infer whether the dynamic exchange analysis of the current traffic road network is qualified, so as to facilitate timely traffic control according to the dynamic exchange analysis and improve the accuracy of traffic condition assessment;

[0059] The numerical ratio of the continuous reduction span of the input vehicle speed in the ramp driving section of the access road network to the continuous increase span of the output vehicle output number in the off-ramp driving section is obtained, and the traffic flow increase speed of the non-input vehicles that use the access road network as the nearest route is obtained. The numerical ratio of the continuous reduction span of the input vehicle speed in the ramp driving section of the access road network to the continuous increase span of the output vehicle output number in the off-ramp driving section and the traffic flow increase speed of the non-input vehicles that use the access road network as the nearest route are marked as network exchange information and network change information, respectively, and compared with the span value ratio threshold and the traffic flow increase speed threshold, respectively:

[0060] If the ratio of the span of continuous reduction in the speed of input vehicles in the ramp section of the access road network to the span of continuous increase in the number of output vehicles in the ramp section exceeds the span ratio threshold, or the traffic flow increase rate of the non-input vehicles taking the detour as the nearest route of the access road network exceeds the traffic flow increase rate threshold, it is inferred that the dynamic exchange analysis of the access road network is abnormal, and a dynamic warning signal is generated and sent to the real-time traffic assessment platform. After receiving the dynamic warning signal, the real-time traffic assessment platform controls the traffic subject of the access road network;

[0061] If the corresponding numerical ratio of the span of continuous reduction of the speed of input vehicles in the ramp-entering driving section of the access road network to the span of continuous increase of the number of output vehicles in the ramp-out driving section does not exceed the span numerical ratio threshold, and the traffic flow increase speed of the non-input vehicles taking the detour of the access road network as the nearest route does not exceed the traffic flow increase speed threshold, it is inferred that the dynamic exchange analysis of the access road network is normal, and a dynamic safety signal is generated and sent to the real-time traffic assessment platform;

[0062] The above formulas are obtained by collecting a large amount of data and performing software simulation, and a formula close to the actual value is selected. The coefficients in the formula are set by technicians in this field according to actual conditions;

[0063] When the present invention is in use, the image recognition monitoring unit is used to classify the types according to the analysis of the passing subjects, and to infer whether there is a risk of passing in the assessment area based on the analysis of different types of passing subjects; the monitoring tracking and analysis unit is used to track and analyze the passing subjects in the assessment area, obtain tracking frequency data and tracking span data, and infer whether there is a congestion risk for the passing subjects in the assessment area based on data comparison; the shared road monitoring unit is used to collect traffic pictures on the shared roads during the process of continuous increase in traffic volume in the assessment area; shared road monitoring information is obtained by collecting traffic pictures, the traffic monitoring coefficient is calculated according to the formula, and it is inferred whether the shared road monitoring is abnormal based on the threshold comparison.

[0064] The preferred embodiments of the present invention disclosed above are only used to help explain the present invention. The preferred embodiments do not describe all the details in detail, nor do they limit the invention to only specific implementation methods. Obviously, many modifications and changes can be made according to the content of this specification. This specification selects and specifically describes these embodiments in order to better explain the principles and practical applications of the present invention, so that those skilled in the art can understand and use the present invention well. The present invention is limited only by the claims and their full scope and equivalents.

Claims

1. A real-time traffic assessment system based on image data analysis and processing, characterized in that: Including the real-time traffic assessment platform, the communication connections of the real-time traffic assessment platform are: An image recognition monitoring unit is used to mark a road area where real-time traffic assessment is to be performed as an assessment area, collect road images in the assessment area and divide the collected images into a number of sub-frame images, and if there is no passing subject in the sub-frame image, the sub-frame image of the type with the highest definition is used as the background frame, and the sub-frame images with passing subjects are analyzed for the passing subjects, and the types are divided according to the passing subject analysis, and it is inferred whether there is a risk of passing in the assessment area according to the analysis of different types of passing subjects; The monitoring tracking and analysis unit is used to track and analyze the passing subjects in the assessment area, obtain tracking frequency data and tracking span data, and infer whether the passing subjects in the assessment area are at risk of congestion based on data comparison; the tracking frequency data and tracking span data are respectively the frequency ratio of the frequency of continuous overtaking on the left side of the passing subject in the assessment area when the same type of passing subjects are in the road area without speeding and the speed fluctuation frequency of the passing subject being overtaken, and the cumulative sum of the span values ​​corresponding to the span value of the traffic flow increase of the continuous driving section of the passing subject being overtaken in the assessment area and the span value of the reduced speed of the overtaking passing subject; if the tracking frequency data exceeds the frequency ratio threshold, or the tracking span data exceeds the cumulative sum threshold of the span value, a high risk signal of driving congestion is generated; If the tracking frequency data does not exceed the frequency ratio threshold, and the tracking span data does not exceed the span value cumulative sum threshold, a low risk signal for driving congestion is generated; The shared road monitoring unit is used to build a network based on all the intricately connected roads within the network and mark them as a passable road network, divide the single routes according to the destinations outside the passable road network, and mark the overlapping areas as shared roads when the single routes are in the passable road network at the same time. At the same time, when the traffic volume in the assessment area continues to increase, the shared roads are used to collect traffic pictures; the shared road monitoring information is obtained by collecting traffic pictures, the traffic monitoring coefficient is calculated according to the formula, and the shared road monitoring is inferred to be abnormal based on the threshold comparison.

2. A real-time traffic assessment system based on image data analysis and processing according to claim 1, characterized in that: The classification process of traffic subject types is as follows: The subframe images with passing subjects are marked as passing analysis images, the moving trajectory of the passing subjects in the passing analysis images and the changed part of the moving trajectory at the adjacent positioning node positions in the corresponding moving trajectory are obtained, and trajectory analysis is performed: If the horizontal angle floating span between the moving trajectory of the passing subject and the roadside trajectories on both sides of the road in the forward direction exceeds the horizontal angle floating span, it is inferred that the passing subject is a fixed-trajectory moving subject; if the horizontal angle floating span between the moving trajectory of the passing subject and the roadside trajectories on both sides of the road in the forward direction does not exceed the horizontal angle floating span, it is inferred that the passing subject is a variable-trajectory moving subject; If the change part of the mobile trajectory at the adjacent positioning node positions in the corresponding mobile trajectory includes the real-time positioning node positions of the passing subject alternating back and forth, it is inferred that the passing subject is a multi-moving mobile subject; otherwise, the change part of the mobile trajectory at the adjacent positioning node positions in the corresponding mobile trajectory does not include the real-time positioning node positions of the passing subject alternating back and forth, then it is inferred that the passing subject is a less-moving mobile subject.

3. A real-time traffic assessment system based on image data analysis and processing according to claim 2, characterized in that: The trajectory overlapping data and trajectory handover data are collected, and the trajectory overlapping data and trajectory handover data are respectively the maximum overlapping area of ​​the areas where the motor vehicle trajectory and the non-motor vehicle trajectory are located in the corresponding time period of the traffic analysis picture, and the duration of the intersection point between the motor vehicle trajectory and the non-motor vehicle trajectory in the corresponding time period of the traffic analysis picture.

4. A real-time traffic assessment system based on image data analysis and processing according to claim 3, characterized in that: If the trajectory overlap data exceeds the maximum overlap area threshold at the same time, or the trajectory intersection data exceeds the intersection duration threshold, an image recognition risk signal is generated; If the trajectory overlap data does not exceed the maximum overlap area threshold at the same time, and the trajectory intersection data does not exceed the intersection point duration threshold, an image recognition safety signal is generated.

5. A real-time traffic assessment system based on image data analysis and processing according to claim 1, characterized in that: The operation process of the shared road monitoring unit is as follows: the outline of the traffic entities traveling on the shared road is analyzed and the traffic entities are set as single points when the outlines all belong to motor vehicles, and the traffic trajectory lines are constructed based on several single points in the collected traffic flow pictures; and monitoring points are set at the forks where the shared roads are connected.

6. A real-time traffic assessment system based on image data analysis and processing according to claim 5, characterized in that: The shared road monitoring information includes the continuously increasing value of the number of single points in the lane in the opposite direction of the fork in the traffic trajectory line of any monitoring point in the shared road in the assessment area that change their position in the direction of the fork, the continuous span of the deviation value of the average speed of vehicles on the traffic trajectory lines at both ends of the current traffic trajectory line with the current fork direction as the midpoint, and the traffic flow value ratio corresponding to the increase in the number of traffic entities that leave the assessment area in the current traffic trajectory line at the same time.

7. A real-time traffic assessment system based on image data analysis and processing according to claim 6, characterized in that: If the traffic monitoring coefficient of the corresponding traffic trajectory line of the shared road in the assessment area exceeds the traffic monitoring coefficient threshold, it is inferred that the shared road monitoring in the current assessment area is abnormal; if the traffic monitoring coefficient of the corresponding traffic trajectory line of the shared road in the assessment area does not exceed the traffic monitoring coefficient threshold, it is inferred that the shared road monitoring in the current assessment area is normal.

Images