A full-time traffic flow state perception and evaluation system based on highway surveillance video

Abstract

The invention discloses a full-time traffic flow state perception and evaluation system for highway monitoring videos, which relates to the field of intelligent transportation technology and comprises: a road section division module, which divides the road sections based on monitoring positions and determines the vehicle dense areas in rainy and snowy weather; a slow-moving analysis module, which uses a windowed correlation coefficient to evaluate the stability of vehicle spacing and judge whether vehicle slow-moving occurs. If slow-moving occurs, a mutation determination module analyzes the correlation between road surface slipperiness and regional aggregation values ​​and locates aggregation mutation points; a behavior classification module, which tracks vehicle characteristics based on mutation points and divides driving behavior types; a perception evaluation module, which formulates a traffic warning response plan according to the diffusion trend of the slow-moving ripple effect caused by aggressive vehicles. The invention can locate congested areas, evaluate slow-moving stability, assist in traffic diversion, and improve the traffic management level and safety assurance capability of highways in rainy and snowy weather.

Description

Technical Field

[0001] The present invention relates to the field of intelligent transportation technology, and in particular to a full-time traffic flow state perception and evaluation system for highway monitoring videos. Background Art

[0002] In the field of intelligent transportation technology, with the continuous growth of traffic volume and the impact of climate change, especially under severe weather conditions such as rain and snow, traditional highway traffic monitoring and management technologies face many challenges.

[0003] In the existing technology, the highway traffic flow parameters are calculated based on the trajectory data obtained by vehicle detection and tracking, and the speed density index is used to comprehensively evaluate the highway traffic status. The slipperiness of the road surface and the vehicle gathering conditions are not correlated and analyzed, resulting in insufficient identification of high-risk sections and lack of targeted traffic induction strategies.

[0004] Moreover, the existing technology does not identify the characteristics of vehicles at the rear boundary of densely populated areas, does not quantify the spacing fluctuations and accident risks caused by different driving behaviors, and cannot provide a basis for risk classification on slippery roads.

[0005] To this end, the present invention provides a full-time traffic flow state perception and evaluation system for highway monitoring videos. Summary of the invention

[0006] The object of the present invention is to provide a full-time traffic flow state perception and evaluation system for highway monitoring videos to solve at least one of the above-mentioned prior art problems.

[0007] In a first aspect, the present invention provides a full-time traffic flow state perception and evaluation system for highway monitoring videos, comprising the following modules:

[0008] Road segmentation module: used to divide the highway into multiple sections, aggregate and evaluate the data of highway vehicles in rainy and snowy weather, and determine the vehicle-dense areas;

[0009] Slow-moving analysis module: used to evaluate the stability of the vehicle spacing in the densely populated area of ​​the road section, obtain the spacing stability value, and determine whether there is a slow-moving vehicle phenomenon in the road section;

[0010] Mutation determination module: If a vehicle slows down, it is used to conduct correlation analysis on the road section slipperiness and regional aggregation value to determine whether the road section slipperiness and regional aggregation value are positively correlated. If they are positively correlated, a standard aggregation sequence is constructed to determine the aggregation mutation point of the road section;

[0011] Behavior classification module: Based on the clustering mutation points, it is used to track the vehicle characteristics at the rear boundary of the vehicle-dense area, classify the vehicle's driving behavior, and identify aggressive vehicles;

[0012] Perception and evaluation module: used to perceive the slow-moving ripple effect of aggressive vehicles and evaluate the diffusion trend of the slow-moving ripples of aggressive vehicles.

[0013] In a second aspect, the present invention provides a method for perceiving and evaluating the full-time traffic flow state of a highway monitoring video, comprising the following steps:

[0014] Step 1: Divide the highway into multiple sections, aggregate and evaluate the data of highway vehicles in rainy and snowy weather, and determine the vehicle-dense areas;

[0015] Step 2: Evaluate the stability of the vehicle spacing in the densely populated area of ​​the road section to obtain a spacing stability value and determine whether there is a slow-moving vehicle phenomenon in the road section;

[0016] Step 3: If the vehicle slows down, a correlation analysis is performed on the road section slipperiness and the regional aggregation value to determine whether the road section slipperiness and the regional aggregation value are positively correlated. If they are positively correlated, a standard aggregation sequence is constructed to determine the aggregation mutation point of the road section.

[0017] Step 4: Based on the clustered mutation points, the vehicle features at the rear boundary of the vehicle-dense area are tracked to classify the vehicle's driving behavior and identify the aggressive vehicles;

[0018] Step 5: Sense the slow-moving ripple effect of aggressive vehicles and evaluate the diffusion trend of the slow-moving ripples of aggressive vehicles.

[0019] Beneficial effects of the present invention:

[0020] 1. Divide the highway into sections based on the monitoring position, obtain the vehicle spacing, calculate the regional aggregation value based on the average vehicle density and lane occupancy, determine the vehicle-dense area, provide key information for traffic management, and make positioning more timely, which improves the efficiency of congestion detection and enables traffic managers to grasp the congestion situation in a timely manner.

[0021] 2. Use the windowed correlation coefficient to process the vehicle spacing data, construct the vehicle spacing matrix, calculate the windowed correlation coefficient value to evaluate the stability of the vehicle spacing, and then obtain the spacing stability value. Compared with the traditional autocorrelation function, the windowed correlation coefficient is more suitable for non-stationary time series such as vehicle spacing. It can reveal the law of vehicle spacing changes and judge its trend. By analyzing the peak position of the function shape, the stability of traffic flow can be judged, so that the system can determine whether there is a slow-moving vehicle phenomenon on the road section, discover potential traffic congestion risks in advance, and provide a reliable basis for traffic control.

[0022] 3. When vehicle slowing occurs, the slipperiness of the road section and the regional aggregation value are correlated and analyzed. If there is a positive correlation, the aggregation mutation point is further determined, which helps to analyze the intrinsic relationship between slippery roads and vehicle aggregation, identify the aggregation mutation points of high-risk sections that are prone to accidents due to slippery roads and dense vehicles, track the characteristics of vehicles at the rear boundary of densely populated areas, and classify driving behaviors into aggressive, conservative, and conventional types. It can quantify the spacing fluctuations and accident risks caused by different driving behaviors, and provide a basis for risk classification on slippery roads.

[0023] 4. Calculate the stable value of the distance between adjacent sub-areas of aggressive vehicles to determine whether a slow-moving ripple effect occurs, and determine the diffusion trend of the ripple effect by calculating the first-order difference mean of the stable value of the distance between the next adjacent sub-areas. Based on different diffusion trends, different traffic warning response plans can be set, which is conducive to improving the traffic efficiency of highways. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings in the following description are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0025] Figure 1 It is a module diagram of a full-time traffic flow state perception and evaluation system of a highway monitoring video provided by the present invention;

[0026] Figure 2 is a vehicle slow-moving determination flow chart provided in the first embodiment of the present invention;

[0027] Figure 3 It is a flow chart of a method for perceiving and evaluating the full-time traffic flow state of a highway monitoring video provided by Embodiment 3 of the present invention;

[0028] Figure 4 It is a structural diagram of a computer device provided in Embodiment 4 of the present invention. DETAILED DESCRIPTION

[0029] 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 should fall within the scope of protection of the present invention.

[0030] Embodiment 1: Figure 1As shown, a full-time traffic flow state perception and evaluation system for highway monitoring video provided by an embodiment of the present invention specifically includes the following modules:

[0031] Road segmentation module: used to divide the highway into multiple sections, aggregate and evaluate the data of highway vehicles in rainy and snowy weather, and determine the vehicle-dense areas;

[0032] In some embodiments, the highway is divided into a plurality of sections based on the locations monitored on the highway;

[0033] The location of the highway monitoring section is used as the section monitoring point;

[0034] Obtain the monitoring screen of the highway in rainy and snowy weather to capture the vehicles at the monitoring points of the road section, and measure the distance between vehicles through the distance sensor;

[0035] If the distance between vehicles is within the preset distance range, merge the vehicles into the same area;

[0036] For example, the monitoring screen of the road monitoring point captures three vehicles, the distance between vehicle A and the front vehicle B is 20 meters, the distance between vehicle B and the front vehicle C is 25 meters, and the distance between vehicle D and the rear vehicle E is 15 meters. The preset distance range is [10,30] meters, then vehicles A, B, C, D, and E are merged into the same area;

[0037] During the monitoring period, the average density of vehicles in the area and the average lane occupancy rate in the area are calculated;

[0038] The average density of vehicles and the mean of lane occupancy are de-dimensionalized, and the average density of vehicles and the mean of lane occupancy are summed to obtain the regional aggregation value;

[0039] Compare the regional aggregation value with a preset regional aggregation threshold, and if the regional aggregation value is higher than the preset regional aggregation threshold, mark the region as a vehicle-dense area;

[0040] It should be noted that the purpose of determining the vehicle-dense area is:

[0041] Function 1: Locate congested areas and identify densely populated sections of highways in rainy and snowy weather by quantitatively analyzing vehicle density and lane occupancy;

[0042] Function 2: Evaluate the stability of the vehicle support slow-moving, identify the densely populated area, and specifically analyze the stability of the internal vehicle spacing to determine whether the vehicle slow-moving phenomenon occurs;

[0043] Function 3: Providing data for slippery road risk assessment. When vehicle slowing occurs, the positioning of dense areas provides a spatial range for slippery road classification and visibility analysis.

[0044] The slow-moving analysis module is used to evaluate the stability of the vehicle spacing in the densely populated area of ​​the road section, obtain the spacing stability value, and determine whether the vehicle slow-moving phenomenon occurs in the road section;

[0045] Obtain the vehicle distance d between vehicle i and spatially adjacent vehicle j in the vehicle-dense area within multiple monitoring cycles i,j , construct the vehicle spacing matrix;

[0046] For example, if the monitoring point at the 6-kilometer section of the highway contains three vehicles connected end to end, in four consecutive monitoring cycles, each cycle is 10 seconds apart, the vehicle spacing matrix is:

[0047] = ;

[0048] where d 1,2 (1) represents the vehicle distance between vehicle 1 and vehicle 2 in the first monitoring cycle;

[0049] By formula: Get the windowed correlation coefficient between vehicle i and its spatially adjacent vehicle j in a densely populated area Value, where d i,j (t) represents the vehicle spacing of vehicle monitoring cycle t, t represents the number of monitoring cycles, T is the total number of monitoring cycles, represents the mean distance between vehicle i and its neighboring vehicle j, k is the time delay step between two adjacent monitoring cycles, and w is the width of the sliding window;

[0050] Preferably, the width of the sliding window is w=5;

[0051] It should be noted that the traditional autocorrelation function is suitable for stationary time series, while the change of vehicle spacing is non-stationary. The vehicle spacing is stabilized by sliding window, and the windowed correlation coefficient is used to quantify the dependency of the same signal at different time points, which reduces the impact of outliers on the global analysis.

[0052] The role of the windowed correlation coefficient in vehicle spacing analysis is:

[0053] Function 1: The windowed correlation coefficient reveals the vehicle spacing d between vehicle i and its spatially adjacent vehicle j in the vehicle-dense area in the vehicle spacing matrix i,j Regularity of changes:

[0054] Function 2: By calculating the correlation at different time steps (k), it is determined whether the change in vehicle spacing is periodic or trending. If the vehicle spacing changes little in adjacent periods, the autocorrelation value will be higher; conversely, if the change is drastic, the autocorrelation value will be lower.

[0055] Function 3: By analyzing the peak position of the windowed correlation coefficient, it is possible to determine whether the traffic flow is in a stable state, such as vehicles moving slowly;

[0056] Obtain the maximum value of the windowed correlation coefficient between all vehicles and spatially adjacent vehicles in the vehicle-dense area, and construct a spacing analysis sequence;

[0057] Calculate the mean of the spacing analysis series to obtain the spacing stability value;

[0058] like Figure 2 As shown, the spacing stability value is compared with the preset spacing stability threshold. If the spacing stability value is higher than or equal to the spacing stability threshold, it means that the vehicle spacing has changed little in multiple monitoring cycles, the system stability is strong, and it meets the characteristics of vehicle slowing down. Then, the vehicle slowing down phenomenon occurs on the current road section.

[0059] If the spacing stability value is lower than the spacing stability threshold, it means that the vehicle spacing fluctuates greatly and has not yet formed a stable slow-moving phenomenon. In this case, there is no slow-moving phenomenon on the current road section, and it is necessary to continuously analyze the changes in the spacing stability value.

[0060] It should be noted that the spacing stability value is the average of the maximum values ​​of the windowed correlation coefficients between all vehicles and adjacent vehicles in the vehicle-dense area. The maximum value of the windowed correlation coefficient reflects the strongest correlation between vehicle spacing under a certain time delay, while the average value comprehensively summarizes the stability of the entire area.

[0061] The technical solution of this embodiment is: divide the highway into multiple sections, cluster and evaluate the data of highway vehicles in rainy and snowy weather, determine the vehicle-dense areas, evaluate the stability of the vehicle spacing in the vehicle-dense areas in the section, obtain the spacing stability value, and determine whether vehicles are slowing down in the section, which is conducive to discovering potential traffic congestion risks in advance and providing a reliable basis for traffic control.

[0062] Embodiment 2: Figure 1 As shown, a full-time traffic flow state perception and evaluation system for highway monitoring video also includes the following modules:

[0063] Mutation determination module: If a vehicle slows down, a correlation analysis is performed on the road section slipperiness and the regional aggregation value to determine whether the road section slipperiness and the regional aggregation value are positively correlated. If so, a standard aggregation sequence is constructed to determine the aggregation mutation point of the road section.

[0064] In some embodiments, the road surface slipperiness at each road section monitoring point is obtained by using a road surface slipperiness detector;

[0065] If vehicle slowing occurs, obtain the road surface slipperiness and regional aggregation values ​​at multiple road monitoring points;

[0066] Number the monitoring points on the road section according to their sequence;

[0067] Based on the road slipperiness and regional aggregation values ​​of multiple road monitoring points, the slipperiness change curve and the aggregation value change curve are drawn in a two-dimensional rectangular coordinate system, with the number of the road monitoring point as the X-axis and the dimensionless value corresponding to the road slipperiness and regional aggregation value as the Y-axis;

[0068] Construct a slipperiness sequence and an aggregation value sequence based on the slipperiness change curve and the aggregation value change curve;

[0069] Calculate the Pearson correlation coefficient of the slipperiness series and the aggregation value series to determine whether the road slipperiness and the regional aggregation value are positively correlated;

[0070] It should be noted that the role of judging whether the road slipperiness and regional aggregation value are positively correlated is:

[0071] Function 1: Analyze the reasons for wet and slippery roads and vehicle gathering. If there is a positive correlation, it means that the higher the wetness of the road, the denser the vehicles. This indicates that drivers tend to reduce speed and keep a smaller distance between vehicles under wet and slippery road conditions, thus forming a local traffic-intensive area.

[0072] Function 2: Improve accident prevention capabilities. The combined effect of slippery roads and densely packed vehicles will significantly increase the risk of rear-end collisions. Positive correlation analysis can identify high-risk sections and provide a basis for accident prevention.

[0073] Function 3: Support accurate traffic guidance. When the aggregation value of a slippery road section is detected to be abnormal in the navigation system, detour suggestions will be pushed first;

[0074] In some embodiments, the linear correlation between the slipperiness sequence and the clustering value sequence is determined. If the slipperiness sequence and the clustering value sequence are linearly correlated, the correlation coefficient between the road slipperiness and the regional clustering value is calculated using the Pearson correlation coefficient.

[0075] If there is nonlinear correlation, the maximum information coefficient method is used to identify the correlation coefficient between the road slipperiness and the regional aggregation value;

[0076] If the correlation coefficient between the road surface slipperiness and the regional aggregation value is within the preset correlation range, the road surface slipperiness and the regional aggregation value are positively correlated, otherwise, the road surface slipperiness and the regional aggregation value are not considered to be positively correlated;

[0077] It should be noted that the preset correlation range is set by professionals in the field based on experience, and the maximum information coefficient method is obtained by calculating the mutual information value of the slipperiness sequence, the aggregation value sequence, and the grid dimension;

[0078] If the road surface slipperiness and the regional aggregation value show a positive correlation, the slipperiness change curve and the aggregation value change curve are divided into multiple curve segments based on the monitoring points of the road section;

[0079] Acquire synchronously rising curve segments in the slipperiness variation curve and the aggregation value variation curve as rising analysis segments;

[0080] Obtain the cluster value sequence corresponding to the rising analysis section, perform noise reduction processing using the sliding window method, and then perform Min-Max standardization processing to construct a standard cluster value sequence;

[0081] It should be noted that since the signature calculation method is sensitive to noise, a sliding window method is used for noise reduction;

[0082] By formula Get the signature difference Tb of the standard aggregate value sequence a,b , where a and b are the numbers of the monitoring points on the road section, n is the total number of monitoring points on the road section, and Z a represents the regional aggregation value of the monitoring point on the road section numbered a, Z b Indicates the regional aggregation value of the monitoring point on the road section numbered b;

[0083] where sign(Z a -Z b ) is a symbolic function;

[0084] By formula: Obtain the road monitoring point corresponding to the maximum value of the symbol difference as the aggregation mutation point;

[0085] It should be noted that the sign difference method detects trend reversal by counting the increasing ratio of aggregation values ​​between monitoring points and using the sign function to locate the aggregation mutation point;

[0086] The purpose of obtaining the aggregation mutation point is:

[0087] Function 1: Real-time congestion warning mechanism. The mutation point reflects the instantaneous jump-like increase in traffic density, which can be used as a precursor signal for congestion.

[0088] Function 2: Accident risk location: mutation points are often accompanied by sudden deceleration and lane change, which can assist in identifying accident-prone areas;

[0089] Function 3: Provide data support for further analysis of vehicle behavior;

[0090] Behavior classification module: Based on the clustering mutation points, it is used to track the vehicle characteristics at the rear boundary of the vehicle-dense area, classify the vehicle's driving behavior, and identify aggressive vehicles;

[0091] Based on the monitoring points of the road sections with clustered mutation points, the vehicles at the rear boundary of the densely populated area are identified;

[0092] Obtain the vehicle distance between the rear boundary vehicle and the spatially adjacent vehicles within multiple monitoring cycles, and calculate the windowed correlation coefficient value between the rear boundary vehicle and the spatially adjacent vehicles;

[0093] Count the number of times the windowed correlation coefficient crosses the zero point and calculate the zero-crossing frequency;

[0094] It should be noted that if the windowed correlation coefficient is positive, it means that the change trend of the vehicle spacing in adjacent cycles is consistent. If it is negative, the change trend of the vehicle spacing in adjacent cycles is opposite. If the windowed correlation coefficient crosses the zero point, it means that the change trend of the vehicle spacing is reversed.

[0095] The higher the zero-crossing frequency, the more violent the spacing fluctuation is and the more aggressive the driving behavior is. Aggressive driving will cause the change direction of the vehicle spacing in adjacent cycles to reverse frequently, such as frequent sudden acceleration and sudden braking. Conservative driving has a more stable change trend of vehicle spacing and fewer zero-crossing times, such as driving at a constant speed.

[0096] Based on the zero-crossing frequency, different frequency demarcation points are set to classify the driving behaviors of the rear boundary vehicles into aggressive, conservative, and conventional types.

[0097] It should be noted that the role of distinguishing driving behaviors is to: improve safety warning capabilities, quantify the spacing fluctuations and accident risks caused by them, and provide a basis for risk classification on slippery roads;

[0098] Perception and evaluation module: perceives the slow-moving ripple effect of aggressive vehicles and evaluates the diffusion trend of the slow-moving ripples of aggressive vehicles;

[0099] Divide the vehicle-dense area into multiple sub-areas and sort them in the order of advancement;

[0100] Calculate the spacing stability value of adjacent sub-areas of aggressive vehicles. If the spacing stability value in the adjacent sub-areas is lower than the spacing stability threshold, or the vehicle-dense area expands in the opposite direction of the forward order, it is considered that the aggressive vehicle produces a slow-moving ripple effect.

[0101] If a slow-moving ripple effect occurs, the first-order difference mean of the spacing stability value in different monitoring periods is calculated for the next adjacent sub-area in the order of the overall forward direction;

[0102] If the first-order difference mean is negative, the slow-moving ripple effect of the sub-region follows the overall forward direction of the vehicle-dense area, that is, the slow-moving ripple effect of the aggressive vehicles shows a tendency to spread forward;

[0103] It should be noted that if the first-order difference mean is negative, the stability value of the spacing between adjacent monitoring cycles shows a downward trend, reflecting that the fluctuation amplitude of the vehicle spacing is increasing and the stability in the region is decreasing;

[0104] If the densely populated area of ​​vehicles expands in the opposite direction of the forward sequence, the slow-moving ripple effect of aggressive vehicles will tend to spread backward;

[0105] For example, if the slow-moving ripple effect of aggressive vehicles tends to spread backward, the nearest highway entrance ramp can be closed, or safe driving tips for rainy and snowy weather can be generated; if it spreads forward, traffic can be diverted;

[0106] Based on the diffusion trend of the slow-moving ripple effect, different traffic warning response plans are set up;

[0107] It should be noted that the role of analyzing the diffusion trend of the slow-moving ripple effect is:

[0108] Function 1: Predict the direction of congestion propagation. If the slow-moving ripple effect spreads in the forward direction, it means that the slow-moving congestion is gradually expanding forward, otherwise it is expanding backward;

[0109] Function 2: Optimize emergency response strategies;

[0110] If the slow-moving ripples of aggressive vehicles show a tendency to spread forward, the regional aggregation value of the vehicle-dense area is obtained;

[0111] The technical solution of this embodiment is: if vehicle slowing down occurs, a correlation analysis is performed on the road section slipperiness and the regional aggregation value to determine whether the road section slipperiness and the regional aggregation value are positively correlated; if they are positively correlated, a standard aggregation sequence is constructed to determine the aggregation mutation point of the road section; based on the aggregation mutation point, the vehicle characteristics at the rear boundary of the vehicle-dense area are tracked, the vehicle driving behavior is classified, the aggressive vehicles are determined, the slowing down ripple effect of the aggressive vehicles is perceived, and the diffusion trend of the slowing down ripples of the aggressive vehicles is evaluated; based on different diffusion trends, the system can set different traffic warning response plans to improve the traffic efficiency of the highway.

[0112] Embodiment 3: Figure 3 As shown, a method for perceiving and evaluating the full-time traffic flow state of a highway monitoring video is used to implement a system for perceiving and evaluating the full-time traffic flow state of a highway monitoring video, comprising the following steps:

[0113] Step 1: Divide the highway into multiple sections, aggregate and evaluate the data of highway vehicles in rainy and snowy weather, and determine the vehicle-dense areas;

[0114] Step 2: Evaluate the stability of the vehicle spacing in the densely populated area of ​​the road section to obtain a spacing stability value and determine whether there is a slow-moving vehicle phenomenon in the road section;

[0115] Step 3: If the vehicle slows down, a correlation analysis is performed on the road section slipperiness and the regional aggregation value to determine whether the road section slipperiness and the regional aggregation value are positively correlated. If they are positively correlated, a standard aggregation sequence is constructed to determine the aggregation mutation point of the road section.

[0116] Step 4: Based on the clustered mutation points, the vehicle features at the rear boundary of the vehicle-dense area are tracked to classify the vehicle's driving behavior and identify the aggressive vehicles;

[0117] Step 5: Sense the slow-moving ripple effect of aggressive vehicles and evaluate the diffusion trend of the slow-moving ripples of aggressive vehicles.

[0118] Example 4: Reference Figure 4 The embodiment of the present invention further provides a computer device 3, comprising: a memory 302 and a processor 301 and a computer program 303 stored in the memory 302. When the computer program 303 is executed on the processor 301, a full-time traffic flow state perception and evaluation method of a highway monitoring video as described in any one of the above methods is implemented.

[0119] The computer device 3 may be a computing device such as a desktop computer, a notebook, a PDA, a cloud server, etc. The computer device 3 may include, but is not limited to, a processor 301 and a memory 302. Those skilled in the art may understand that;

[0120] Figure 4 It is only an example of computer device 3 and does not constitute a limitation on computer device 3. It may include more or fewer components than shown in the figure, or a combination of certain components, or different components, for example, it may also include input and output devices, network access devices, etc.

[0121] The processor 301 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc.

[0122] In some embodiments, the memory 302 may be an internal storage unit of the computer device 3, such as a hard disk or memory of the computer device 3. In other embodiments, the memory 302 may also be an external storage device of the computer device 3, such as a plug-in hard disk, a smart memory card (SmartMediaCard, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), etc. equipped on the computer device 3. Further, the memory 302 may also include both an internal storage unit of the computer device 3 and an external storage device. The memory 302 is used to store an operating system, an application program, a boot loader (BootLoader), data, and other programs, such as the program code of the computer program. The memory 302 may also be used to temporarily store data that has been output or is to be output.

[0123] Embodiment 5: The embodiment of the present invention further provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, it implements a method for perceiving and evaluating the full-time traffic flow state of a highway monitoring video as described in any one of the above methods.

[0124] In this embodiment, if the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present application implements all or part of the processes in the above-mentioned embodiment method, which can be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned various method embodiments when executed by the processor. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium can at least include: any entity or device that can carry the computer program code to the camera / terminal device, recording medium, computer memory, read-only memory (ROM, Read-Only Memory, ROM), random access memory (RAM, Random Access Memory), electric carrier signal, telecommunication signal and software distribution medium. For example, USB flash drive, mobile hard disk, disk or optical disk. In some jurisdictions, according to legislation and patent practice, computer-readable media cannot be electric carrier signals and telecommunication signals.

[0125] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described or recorded in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0126] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0127] In the embodiments disclosed in the present application, it should be understood that the disclosed devices / terminal equipment and methods can be implemented in other ways. For example, the device / terminal equipment embodiments described above are only schematic. For example, the division of the modules or units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0128] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0129] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters in the formula are set by technicians in this field according to actual conditions.

[0130] The above is a detailed description of an embodiment of the present invention, but the content is only a preferred embodiment of the present invention and cannot be considered to limit the scope of implementation of the present invention. All equivalent changes and improvements made within the scope of the present invention should still fall within the scope of the patent coverage of the present invention.

Claims

1. A full-time traffic flow state perception and evaluation system for highway monitoring videos, characterized in that: Includes the following modules: Slow-moving analysis module: used to evaluate the stability of the vehicle spacing in the densely populated area of ​​the road section, obtain the spacing stability value, and determine whether there is a slow-moving vehicle phenomenon in the road section; Mutation determination module: If a vehicle slows down, it is used to conduct correlation analysis on the road section slipperiness and regional aggregation value to determine whether the road section slipperiness and regional aggregation value are positively correlated. If they are positively correlated, a standard aggregation sequence is constructed to determine the aggregation mutation point of the road section; Behavior classification module: Based on the clustering mutation points, it is used to track the vehicle characteristics at the rear boundary of the vehicle-dense area, classify the vehicle's driving behavior, and identify aggressive vehicles; Perception and evaluation module: used to perceive the slow-moving ripple effect of aggressive vehicles and evaluate the diffusion trend of the slow-moving ripples of aggressive vehicles.

2. The full-time traffic flow state perception and evaluation system of highway monitoring video according to claim 1 is characterized in that: The method for determining the vehicle dense area is as follows: Divide the expressway into multiple road section monitoring points and obtain the vehicle spacing between the road section monitoring points; The vehicle spacing is merged, the regional aggregation value after the merger is calculated, and the vehicle-dense area is determined.

3. The full-time traffic flow state perception and evaluation system of highway monitoring video according to claim 1 is characterized in that: The identification method of the vehicle slowing down phenomenon is as follows: Obtain the vehicle spacing in the vehicle-dense area during multiple monitoring cycles and construct a vehicle spacing matrix; The vehicle spacing matrix is ​​numerically analyzed to obtain the spacing stability value, and the spacing stability value is compared and analyzed to determine the vehicle slow-moving phenomenon within the road section.

4. The full-time traffic flow state perception and evaluation system of highway monitoring video according to claim 3 is characterized in that: The method for obtaining the spacing stability value is as follows: Based on the vehicle spacing matrix, the maximum value of the windowed correlation coefficients between all vehicles and spatially adjacent vehicles is obtained through the windowed autocorrelation function to construct a spacing analysis sequence; Calculate the mean of the spacing analysis series to obtain the spacing stability value.

5. The full-time traffic flow state perception and evaluation system of highway monitoring video according to claim 1 is characterized in that: The aggregation mutation point is determined in the following manner: If vehicle slowing occurs, curve analysis is performed on the road surface slipperiness and regional aggregation values ​​of multiple road monitoring points to construct slipperiness sequence and aggregation value sequence; If the road surface slipperiness and the regional aggregation value show a positive correlation, the slipperiness change curve and the aggregation value change curve are divided into multiple curve segments based on the monitoring points of the road section; Screen and analyze the curve segments to determine the rising analysis segment; Obtain the aggregation value sequence corresponding to the rising analysis section, perform noise reduction processing using the sliding window method, and then perform Min-Max standardization processing to construct a standard aggregation value sequence; The standard aggregation value sequence is processed by the signature calculation formula to obtain the road section monitoring point corresponding to the maximum signature value as the aggregation mutation point.

6. A full-time traffic flow state perception and evaluation system for highway monitoring video according to claim 5, characterized in that: The method for judging whether the road surface slipperiness and the regional aggregation value are positively correlated is as follows: A correlation analysis is performed on the slipperiness sequence and the clustering value sequence. If they are within the preset correlation range, the road slipperiness and the regional clustering value are positively correlated.

7. The full-time traffic flow state perception and evaluation system of highway monitoring video according to claim 5 is characterized in that: The rising analysis section is determined as follows: The curve segments that rise synchronously in the slipperiness change curve and the aggregation value change curve are taken as the rising analysis segments.

8. The full-time traffic flow state perception and evaluation system of highway monitoring video according to claim 1 is characterized in that: The aggressive vehicle is determined as follows: Based on the monitoring points of the road sections with clustered mutation points, the vehicles at the rear boundary of the densely populated area are identified; Obtain the vehicle distance between the rear boundary vehicle and the spatially adjacent vehicles within multiple monitoring cycles, and calculate the windowed correlation coefficient value between the rear boundary vehicle and the spatially adjacent vehicles; Count the number of times the windowed correlation coefficient crosses the zero point and calculate the zero-crossing frequency; Based on the zero-crossing frequency, different frequency demarcation points are set to determine the aggressive vehicles among the rear boundary vehicles.

9. The full-time traffic flow state perception and evaluation system of highway monitoring video according to claim 1 is characterized in that: The diffusion trend of the slow-moving ripples is obtained as follows: Divide the densely populated area into multiple sub-areas, compare and analyze the stability values ​​of the distances between adjacent sub-areas of aggressive vehicles, and determine the slow-moving ripple effect caused by aggressive vehicles; If a ripple effect occurs, the first-order difference mean of the stable value of the spacing between the next adjacent sub-regions is calculated according to the order of the overall forward direction; If the first-order difference mean is negative, the ripple effect of aggressive vehicles tends to spread forward; If the vehicle-dense area expands in the opposite direction of the forward order, the slow-moving ripple effect of aggressive vehicles will tend to spread backward.

10. A full-time traffic flow state perception and evaluation system for highway monitoring video according to claim 9, characterized in that: The method for comparing and analyzing the spacing stability value is as follows: If the spacing stability value in the adjacent sub-area is lower than the spacing stability threshold, or the vehicle-dense area expands in the opposite direction of the forward order, it is considered that the aggressive vehicle produces a slow-moving ripple effect.

Images