Intelligent networked traffic flow stability analysis method

Abstract

The application relates to the technical field of traffic management, and provides a smart networked traffic flow stability analysis method, which comprises the following steps: collecting vehicle video data of vehicles driving in a traffic area by using a preset camera device, and calculating a speed variation value and a regional resistance coefficient of the traffic area; performing frame processing on the vehicle video data to obtain a target video frame, detecting a frame sequence corresponding to the target video frame, and calculating vehicle acceleration corresponding to the driving vehicles; extracting regional features corresponding to the target video frame, positioning vehicle positions of the driving vehicles in the traffic area, calculating a following vehicle braking distance corresponding to each vehicle in the driving vehicles, querying a vehicle reaction time lag corresponding to the driving vehicles, and calculating a preceding vehicle braking distance corresponding to each vehicle in the driving vehicles; and calculating a rear-end collision risk coefficient corresponding to each vehicle in the driving vehicles, and analyzing traffic flow stability corresponding to the traffic area. The application is characterized in that the accuracy of smart networked traffic flow stability analysis is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of traffic management, and particularly relates to a smart connected vehicle traffic flow stability analysis method. BACKGROUND

[0002] People, vehicles and roads are the three elements of road traffic, and are also the main reasons for frequent road safety problems. Therefore, in order to reduce the accident rate, it is necessary to analyze the traffic flow stability in the region in order to prevent accidents in advance.

[0003] The existing traffic flow stability analysis is a simulation method, which simulates the traffic flow behavior in the corresponding region by establishing a traffic simulation model, and predicts the traffic flow stability of the corresponding region according to the simulated traffic flow behavior. However, this method is an idealized situation and cannot completely reflect various factors and uncertainties in the real traffic system, resulting in a deviation between the predicted results and the actual situation, and thus reducing the accuracy of traffic flow stability analysis. SUMMARY

[0004] The present application provides a smart connected vehicle traffic flow stability analysis method, which mainly aims to improve the accuracy of smart connected vehicle traffic flow stability analysis.

[0005] To achieve the above purpose, the present application provides a smart connected vehicle traffic flow stability analysis method, which comprises:

[0006] Obtain the traffic area to be analyzed, use the preset camera device to collect the vehicle video data of the driving vehicle in the traffic area, use the preset intelligent network system to count the vehicle speed value and the number of vehicles corresponding to the driving vehicle, and combine the vehicle speed value and the number of vehicles to calculate the speed variation value and the regional resistance coefficient of the traffic area;

[0007] Frame processing is performed on the vehicle video data to obtain a target video frame, a frame sequence corresponding to the target video frame is detected, and the vehicle acceleration corresponding to the driving vehicle is calculated according to the frame sequence and the vehicle speed value;

[0008] Extract the region features corresponding to the target video frame, locate the vehicle position of the driving vehicle in the traffic area according to the region features, combine the vehicle acceleration, the vehicle speed value and the vehicle position to calculate the following brake travel of each vehicle in the driving vehicle, query the vehicle reaction time lag corresponding to the driving vehicle, and combine the vehicle reaction time lag, the vehicle speed value and the vehicle acceleration to calculate the front brake travel of each vehicle in the driving vehicle;

[0009] The rear-end risk coefficient corresponding to each vehicle in the traveling vehicle is calculated in combination with the rear-end brake stroke of the following vehicle and the front brake stroke of the leading vehicle, and the traffic flow stability corresponding to the traffic area is analyzed in combination with the rear-end risk coefficient, the speed variation value and the area drag coefficient.

[0010] Optionally, the speed variation value and the area drag coefficient of the traffic area are calculated in combination with the vehicle speed value and the number of vehicles, including:

[0011] The time sequence average speed corresponding to the time sequence of vehicle speed is calculated according to the vehicle speed value.

[0012] The speed variation value of the traffic area is calculated in combination with the time sequence average speed and the vehicle speed value.

[0013] The area drag coefficient of the traffic area is calculated in combination with the time sequence of vehicle speed, the vehicle speed value and the number of vehicles.

[0014] Optionally, the area drag coefficient of the traffic area is calculated in combination with the time sequence of vehicle speed, the vehicle speed value and the number of vehicles, including:

[0015] The time sequence length corresponding to the time sequence of vehicle speed is calculated.

[0016] The drag vehicle speed value in the vehicle speed value is identified according to a preset drag speed threshold.

[0017] The number of drag vehicles corresponding to the drag vehicle speed value is determined.

[0018] The area drag coefficient of the traffic area is calculated in combination with the number of drag vehicles, the time sequence length and the number of vehicles.

[0019] Optionally, the vehicle video data is frame-processed to obtain a target video frame, including:

[0020] The vehicle video data is frame-processed by using a preset frame processing tool to obtain an initial video frame.

[0021] The initial video frame is denoised to obtain a denoised video frame.

[0022] The denoised video frame is deblurred to obtain a clear video frame.

[0023] The clear video frame is color-enhanced to obtain a target video frame.

[0024] Optionally, the vehicle acceleration corresponding to the traveling vehicle is calculated according to the frame sequence and the vehicle speed value, including:

[0025] Map the vehicle speed value with the frame sequence to obtain a frame vehicle speed value;

[0026] According to the frame sequence, the frame vehicle speed value corresponding to the vehicle speed interval is calculated;

[0027] In combination with the frame vehicle speed value, the frame sequence corresponding to the vehicle speed amplitude is calculated;

[0028] In combination with the vehicle speed amplitude and the vehicle speed interval, the vehicle acceleration corresponding to the driving vehicle is calculated.

[0029] Optionally, the extraction of the region feature corresponding to the target video frame comprises:

[0030] Vehicle identification is performed on the target video frame to obtain an identified vehicle, and a vehicle surrounding area corresponding to the identified vehicle is determined;

[0031] According to the vehicle surrounding area and the identified vehicle, image segmentation processing is performed on the target video frame to obtain a segmented region image;

[0032] The region color feature corresponding to the segmented region image is extracted, and the segmented region image is subjected to gray processing to obtain a gray region image;

[0033] The region texture feature corresponding to the gray region image is extracted;

[0034] In combination with the region color feature and the region texture feature, the region feature corresponding to the target video frame is determined.

[0035] Optionally, the combination of the region color feature and the region texture feature to determine the region feature corresponding to the target video frame comprises:

[0036] The color feature value corresponding to the region color feature is calculated, and according to the color feature value, the color dispersion of each feature in the region color feature is calculated;

[0037] According to the color dispersion, the region color feature is subjected to filtering processing to obtain a first region feature;

[0038] The region texture feature is subjected to quantization processing to obtain a texture feature value, and the probability density corresponding to the texture feature value is calculated;

[0039] According to the probability density, the texture entropy corresponding to the region texture feature is calculated;

[0040] According to the texture entropy, the region texture feature is subjected to screening processing to obtain a second region feature;

[0041] The first regional feature and the second regional feature are fused to obtain a regional feature corresponding to the target video frame.

[0042] Optionally, the vehicle acceleration, the vehicle speed value and the vehicle position are combined to calculate a front vehicle brake stroke corresponding to each vehicle in the traveling vehicles.

[0043] The vehicle body length parameter corresponding to the traveling vehicles is queried.

[0044] The headway between the traveling vehicles is determined according to the vehicle position.

[0045] The headway, the vehicle body length parameter, the vehicle acceleration and the vehicle speed value are combined to calculate the front vehicle brake stroke number corresponding to each vehicle in the traveling vehicles.

[0046] Optionally, the vehicle reaction time lag, the vehicle speed value and the vehicle acceleration are combined to calculate a following vehicle brake stroke corresponding to each vehicle in the traveling vehicles.

[0047] An intelligent networked traffic flow stability analysis system, characterized in that the system comprises:

[0048] A regional lag coefficient calculation module is configured to acquire a traffic region to be analyzed, collect vehicle video data of traveling vehicles in the traffic region by using a preset camera device, and calculate a speed variation value and a regional lag coefficient of the traffic region by combining a vehicle speed value and a vehicle number of the traveling vehicles, which are counted by a preset intelligent network system.

[0049] A vehicle acceleration calculation module is configured to perform frame processing on the vehicle video data to obtain a target video frame, detect a frame sequence corresponding to the target video frame, and calculate a vehicle acceleration of the traveling vehicles according to the frame sequence and the vehicle speed value.

[0050] A brake stroke calculation module is configured to extract a regional feature corresponding to the target video frame, locate a vehicle position of the traveling vehicles in the traffic region according to the regional feature, calculate a following vehicle brake stroke corresponding to each vehicle in the traveling vehicles by combining the vehicle acceleration, the vehicle speed value and the vehicle position, query a vehicle reaction time lag of the traveling vehicles, and calculate a front vehicle brake stroke corresponding to each vehicle in the traveling vehicles by combining the vehicle reaction time lag, the vehicle speed value and the vehicle acceleration.

[0051] The stability analysis module is used for combining the following vehicle brake stroke and the front vehicle brake stroke to calculate a rear-end risk coefficient corresponding to each vehicle in the running vehicle, and combining the rear-end risk coefficient, the speed variation value and the regional resistance coefficient to analyze traffic flow stability corresponding to the traffic region.

[0052] The speed variation value and the regional resistance coefficient of the traffic region can be calculated by combining the vehicle speed value and the vehicle quantity, so that the speed variation degree of the traffic region and the congestion degree of the region can be known, and the traffic flow stability of the traffic region can be analyzed subsequently, the vehicle video data can be converted into corresponding individual frames by frame processing, so that each frame can be processed individually, and the calculation complexity is reduced, the region representation in the target video frame can be obtained by extracting the region features corresponding to the target video frame, and the vehicle position of the running vehicle in the traffic region is provided for subsequent positioning, the rear-end risk coefficient corresponding to each vehicle in the running vehicle can be calculated by combining the following vehicle brake stroke and the front vehicle brake stroke, so that the rear-end risk of each vehicle in the running vehicle can be known, and the analysis accuracy of the traffic flow stability is improved. Therefore, the intelligent networked traffic flow stability analysis method provided by the embodiment of the present application can improve the accuracy of intelligent networked traffic flow stability analysis. BRIEF DESCRIPTION OF DRAWINGS

[0053] Figure 1 A flowchart of an intelligent networked traffic flow stability analysis method provided by an embodiment of the present application is shown.

[0054] Figure 2 A functional module diagram of an intelligent networked traffic flow stability analysis system provided by an embodiment of the present application is shown.

[0055] The implementation, functional features and advantages of the present application will be further described with reference to the embodiments and the accompanying drawings. DETAILED DESCRIPTION

[0056] It should be understood that the specific embodiments described herein are merely intended to explain the present application, and are not intended to limit the present application.

[0057] The embodiment of the present application provides a kind of intelligent network connection traffic flow stability analysis method.The execution subject of the kind of intelligent network connection traffic flow stability analysis method in the embodiment of the present application includes but is not limited to at least one of the electronic equipment that can be configured to execute the method provided in the embodiment of the present application, such as server, terminal etc.It is said in other words, the kind of intelligent network connection traffic flow stability analysis method can be executed by software or hardware installed in terminal equipment or server equipment, and the software can be blockchain platform.The server includes but is not limited to: single server, server cluster, cloud server or cloud server cluster etc.The server can be independent server, can also be cloud server that provides cloud service, cloud database, cloud computing, cloud function, cloud storage, network service, cloud communication, middleware service, domain name service, security service, content distribution network (Content Delivery Network, CDN), and basic cloud computing services such as big data and artificial intelligence platform.

[0058] Referring to Figure 1 As shown in the flowchart of the intelligent network connection traffic flow stability analysis method provided by an embodiment of the present application.In this embodiment, the kind of intelligent network connection traffic flow stability analysis method includes steps S1-S4.

[0059] S1, obtain the traffic area to be analyzed, collect the vehicle video data of the vehicle driving in the traffic area by using the preset camera equipment, and calculate the speed variation value and the regional resistance coefficient of the traffic area by using the preset intelligent network connection system to count the vehicle speed value and the vehicle quantity corresponding to the driving vehicle in combination with the vehicle speed value and the vehicle quantity.

[0060] The present application can understand the speed variation degree of the traffic area and the congestion degree of the region by calculating the speed variation value and the regional resistance coefficient of the traffic area in combination with the vehicle speed value and the vehicle quantity, so as to facilitate subsequent analysis of the traffic flow stability of the traffic area, wherein the camera equipment is a monitoring device in the traffic area, such as camera, the vehicle video data is the video data of the driving vehicle in the traffic area shot by the camera equipment, and the intelligent network connection system is a vehicle-mounted device system integrating modern communication and network technology, with functions of vehicle state monitoring, environment perception, path planning and automatic driving, etc.The speed variation value represents the speed variation degree of the traffic area, and the regional resistance coefficient represents the congestion degree of the vehicle driving in the traffic area.

[0061] As one embodiment of the present application, the speed variation value and the area resistance coefficient of the traffic area are calculated by combining the vehicle speed value and the vehicle quantity, comprising: identifying the vehicle speed time sequence corresponding to the vehicle speed value, calculating the time sequence average speed corresponding to the vehicle speed time sequence according to the vehicle speed value, combining the time sequence average speed and the vehicle speed value to calculate the speed variation value of the traffic area, and combining the vehicle speed time sequence, the vehicle speed value and the vehicle quantity to calculate the area resistance coefficient of the traffic area.

[0062] Wherein, the vehicle speed time sequence is the measurement time corresponding to the vehicle speed value, the time sequence average speed is the average value of all speeds in each time sequence in the vehicle speed time sequence, further, the vehicle speed time sequence corresponding to the vehicle speed value can be realized by a time sequence identification tool, and the time sequence identification tool is compiled by a script language, such as JS language; the time sequence average speed corresponding to the vehicle speed time sequence can be realized by an average function, the sum of all vehicle speed values in each time sequence in the vehicle speed time sequence is calculated, and finally the sum is averaged to obtain the time sequence average speed corresponding to each time sequence.

[0063] Further, as one optional embodiment of the present application, the area resistance coefficient of the traffic area is calculated by combining the time sequence average speed and the vehicle speed value, comprising: counting the time sequence vehicle quantity corresponding to each time sequence in the vehicle speed time sequence, and combining the time sequence vehicle quantity, the time sequence average speed and the vehicle speed value to calculate the speed variation value of the traffic area by the following formula:

[0064]

[0065] Wherein, A represents the speed variation value of the traffic area, a represents the time sequence vehicle serial number corresponding to the Bth vehicle speed time sequence, B, B+1 and B+e respectively represent the serial number of the vehicle speed time sequence, a, b and d respectively represent the vehicle serial number in the vehicle speed time sequence corresponding to B, B+1 and B+e, r, y and x respectively represent the vehicle quantity in the vehicle speed time sequence corresponding to B, B+1 and B+e, respectively represent the time sequence average speed of the vehicle speed time sequence corresponding to B, B+1 and B+e, B a , (B+1) b , (B+e) b respectively represent the vehicle speed value corresponding to the a th vehicle, the b th vehicle and the d th vehicle in the vehicle speed time sequence corresponding to B, B+1 and B+e.

[0066] Further, as an optional embodiment of the present application, the combination of the vehicle speed time sequence, the vehicle speed value and the vehicle quantity is used to calculate the regional resistance coefficient of the traffic area, which comprises: calculating the time sequence length corresponding to the vehicle speed time sequence, identifying the resistance vehicle speed value in the vehicle speed value according to the preset resistance speed threshold, determining the resistance vehicle quantity corresponding to the resistance vehicle speed value, and combining the resistance vehicle quantity, the time sequence length and the vehicle quantity to calculate the regional resistance coefficient of the traffic area by the following formula:

[0067]

[0068] Wherein, D represents the regional resistance coefficient of the traffic area, E represents the resistance vehicle quantity, T l represents the time sequence length, and F represents the vehicle quantity.

[0069] Wherein, the time sequence length is the length corresponding to the vehicle speed time sequence, for example, the time sequence length is 10 seconds, which means that the vehicle speed time sequence is between 0 and 10 seconds, the resistance speed threshold represents the speed defined by the congestion vehicle, and the resistance speed threshold can be calculated by analyzing a large amount of vehicle speed data to determine the speed characteristics in congestion, so as to calculate the corresponding resistance speed threshold. Generally, the resistance speed threshold is set to 10 km / h, and the vehicle below the speed is called a congestion vehicle. Further, the calculation of the time sequence length corresponding to the vehicle speed time sequence can be obtained by subtracting the initial time sequence from the final time sequence.

[0070] S2, frame processing is performed on the vehicle video data to obtain a target video frame, a frame sequence corresponding to the target video frame is detected, and a vehicle acceleration corresponding to the driving vehicle is calculated according to the frame sequence and the vehicle speed value.

[0071] The vehicle video data can be converted into corresponding individual frames by frame processing the vehicle video data, so that each frame can be processed individually, and the calculation complexity is reduced, wherein the target video frame is an optimized frame obtained by processing a single picture corresponding to the vehicle video data.

[0072] As an embodiment of the present application, the frame processing of the vehicle video data to obtain a target video frame comprises: using a preset frame processing tool to perform frame processing on the vehicle video data to obtain an initial video frame, performing noise reduction processing on the initial video frame to obtain a noise reduction video frame, performing deblurring processing on the noise reduction video frame to obtain a clear video frame, and performing color enhancement processing on the clear video frame to obtain a target video frame.

[0073] The frame splitting tool is a tool for splitting the vehicle video data into individual frames, such as an FFmpeg tool, the initial video frame is an individual frame obtained after the vehicle video data is split, the noise-removed video frame is a frame obtained after noise interference in the initial video frame is removed, and the clear video frame is a frame obtained after a blurred area in the noise-removed video frame is optimized.

[0074] Further, the noise-removed processing of the initial video frame can be realized by a mean filtering method, the noise in the video frame is reduced by performing an average operation on a pixel neighborhood of the initial video frame, the deblurring processing of the noise-removed video frame can be realized by a deblurring algorithm, such as a Rich a ardson-Lucy algorithm, and the color-enhanced processing of the clear video frame can be realized by a Retinex algorithm.

[0075] The application can obtain the time sequence and time point of the target video frame by detecting the frame sequence corresponding to the target video frame, and can obtain the motion state of the running vehicle by calculating the vehicle acceleration corresponding to the running vehicle according to the frame sequence and the vehicle speed value, thereby facilitating the calculation of the brake stroke, and further, the detection of the frame sequence corresponding to the target video frame can be realized by an optical flow method.

[0076] As an embodiment of the application, the calculation of the vehicle acceleration corresponding to the running vehicle according to the frame sequence and the vehicle speed value comprises: performing mapping processing on the vehicle speed value and the frame sequence to obtain a frame speed value, calculating the vehicle speed interval corresponding to the frame speed value according to the frame sequence, combining the frame speed value to calculate the vehicle speed amplitude corresponding to the frame sequence, and combining the vehicle speed amplitude and the vehicle speed interval to calculate the vehicle acceleration corresponding to the running vehicle.

[0077] The frame speed value is a vehicle speed value obtained by sorting the vehicle speed value and the frame sequence, the vehicle speed interval represents a time interval or time difference between two vehicle speeds, and the vehicle speed amplitude represents a vehicle speed change amount between the frame sequences.

[0078] Further, the vehicle speed interval corresponding to the frame speed value can be obtained by calculating the sequence difference value between the frame sequences, the vehicle speed amplitude corresponding to the frame sequence can be obtained by calculating the difference value between the frame speed values, and the vehicle acceleration corresponding to the running vehicle can be obtained by calculating the ratio of the vehicle speed amplitude and the vehicle speed interval.

[0079] S3, extract the region feature corresponding to the target video frame, locate the vehicle position of the running vehicle in the traffic area according to the region feature, calculate the follow-up brake distance corresponding to each vehicle in the running vehicle in combination with the vehicle acceleration, the vehicle speed value and the vehicle position, and calculate the front brake distance corresponding to each vehicle in the running vehicle in combination with the vehicle reaction time lag, the vehicle speed value and the vehicle acceleration.

[0080] The application can obtain the region representation in the target video frame by extracting the region feature corresponding to the target video frame, thereby providing a basis for subsequent positioning of the vehicle position of the running vehicle in the traffic area.

[0081] As an embodiment of the application, the extraction of the region feature corresponding to the target video frame comprises: vehicle recognition of the target video frame to obtain an identified vehicle, determination of a vehicle surrounding region corresponding to the identified vehicle, image segmentation processing of the target video frame according to the vehicle surrounding region and the identified vehicle to obtain a segmented region image, extraction of a region color feature corresponding to the segmented region image, gray processing of the segmented region image to obtain a gray region image, extraction of a region texture feature corresponding to the gray region image, and determination of the region feature corresponding to the target video frame in combination with the region color feature and the region texture feature.

[0082] The segmented region image is an image of the surrounding region corresponding to the identified vehicle and does not contain the corresponding identified vehicle, the region color feature is a color representation corresponding to the segmented region image, the gray region image is an image of the segmented region image expressed by a single color, and the region texture feature is a representation of texture information corresponding to the segmented region image.

[0083] Further, the vehicle recognition of the target video frame can be realized by a vehicle detection algorithm such as a YOLO algorithm, the determination of the vehicle surrounding region corresponding to the identified vehicle can be realized by taking a rectangular frame of a fixed size as a surrounding region marker frame with the position of the identified vehicle as the center, thereby obtaining the vehicle surrounding region, the segmented region image can be obtained by removing the image of the identified vehicle in the vehicle surrounding region, the extraction of the region color feature corresponding to the segmented region image can be realized by an HSV model, the gray processing of the segmented region image can be realized by an average value method, that is, the average value of the RGB value of each pixel of the segmented region image is taken to obtain a gray value, and each pixel is replaced by the gray value, thereby obtaining the gray region image, and the extraction of the region texture feature corresponding to the gray region image can be realized by a gray level co-occurrence matrix method.

[0084] Optionally, as an optional embodiment of the present application, the combination of the region color feature and the region texture feature to determine the region feature corresponding to the target video frame comprises: calculating a color feature value corresponding to the region color feature, calculating a color dispersion of each feature in the region color feature according to the color feature value, filtering the region color feature according to the color dispersion to obtain a first region feature, quantizing the region texture feature to obtain a texture feature value, calculating a probability density corresponding to the texture feature value, calculating a texture entropy corresponding to the region texture feature according to the probability density, screening the region texture feature according to the texture entropy to obtain a second region feature, and fusing the first region feature and the second region feature to obtain the region feature corresponding to the target video frame.

[0085] In the formula, the color feature value is an expression value corresponding to the region color feature, the color dispersion represents a dispersion degree corresponding to the region color feature, the texture feature value is an expression value corresponding to the region texture feature, the probability density represents an occurrence probability of each feature value in the texture feature value, and the texture entropy represents a confusion degree corresponding to the region texture feature. The higher the confusion degree is, the more important the corresponding texture feature is.

[0086] Further, the calculation of the color feature value corresponding to the region color feature can be implemented through an RGB color space model; the color dispersion can be obtained according to a color standard deviation of each feature in the region color feature calculated from the color feature value; the color dispersion is compared with a dispersion threshold value, when the color dispersion is less than the dispersion threshold value, the corresponding region color feature is filtered, features with the color dispersion greater than the dispersion threshold value are reserved to obtain the first region feature, the dispersion threshold value can be set to 0.8, or can be set according to an actual application scenario; the quantization of the region texture feature can be implemented through an encoding method, such as a Huffman encoding method; the probability density can be obtained by calculating a frequency of occurrence of each feature value in the texture feature value; the texture entropy corresponding to the region texture feature can be calculated through a Shannon entropy algorithm; the screening principle of the second region feature and the first region feature is the same, and will not be described in detail here; the feature fusion of the first region feature and the second region feature can be implemented through a series fusion method, such as a direct series method, the vectors corresponding to the first region feature and the second region feature are connected together to form a new vector, and the region feature corresponding to the new vector is generated according to the new vector.

[0087] The application combines the vehicle acceleration, the vehicle speed value and the vehicle position to calculate the following vehicle braking distance corresponding to each vehicle in the traveling vehicle, so as to obtain the braking distance of each vehicle in the traveling vehicle when following the vehicle, thereby providing a basis for subsequent calculation of the rear-end risk coefficient, wherein the vehicle position is a specific area position of the traveling vehicle in the traffic area, and further, the area feature is matched with the feature in the traffic area, and the vehicle position of the traveling vehicle in the traffic area is located according to the matching result.

[0088] As an embodiment of the application, the combination of the vehicle acceleration, the vehicle speed value and the vehicle position to calculate the front vehicle braking distance corresponding to each vehicle in the traveling vehicle comprises: querying the body length parameter of the traveling vehicle, determining the head distance between the traveling vehicles according to the vehicle position, combining the head distance, the body length parameter, the vehicle acceleration and the vehicle speed value, and calculating the front vehicle braking distance corresponding to each vehicle in the traveling vehicle through the following formula:

[0089]

[0090] Wherein, M represents the front vehicle braking distance corresponding to each vehicle in the traveling vehicle, B i+1 represents the vehicle speed value corresponding to the i+1th vehicle in the traveling vehicle, s i，i+1 represents the head distance between the i th vehicle and the i+1th vehicle in the traveling vehicle, a i represents the vehicle acceleration corresponding to the i th vehicle in the traveling vehicle, G i represents the body length parameter corresponding to the i th vehicle in the traveling vehicle, and i and i+1 represent the serial numbers of the traveling vehicles.

[0091] Wherein, the body length parameter represents the specific length value of the body of the traveling vehicle, the head distance represents the distance between the head of the following vehicle and the tail of the corresponding front vehicle, further, the area position corresponding to the traveling vehicle is determined according to the vehicle position, the distance between the tail of the front vehicle and the head of the rear vehicle in the traveling vehicle is measured by a measuring tool, so as to obtain the head distance, wherein the measuring tool comprises a laser range finder, i and i+1 in the above formula correspond to the serial numbers of the following vehicle and the front vehicle corresponding to the following vehicle in the traveling vehicle, and the front vehicle braking distance calculated by the above formula is the braking distance of the previous vehicle of the i th vehicle as the reference vehicle.

[0092] The application calculates the follow-up braking distance of each vehicle in the traveling vehicle by combining the vehicle reaction time lag, the vehicle speed value and the vehicle acceleration, so as to improve the calculation accuracy of the subsequent rear-end risk coefficient, wherein the vehicle reaction time lag represents the time interval required for the driver of the following vehicle in the traveling vehicle to perceive and react to the state change (such as speed, acceleration, etc.) of the front vehicle, and the vehicle reaction time lag corresponding to the traveling vehicle can be obtained by querying the pre-constructed reaction time table, and the reaction time table is constructed by a natural driving research method, which records the data in the actual driving process by the sensors and cameras installed on the vehicle, and analyzes the corresponding perception reaction time of the driver.

[0093] As an embodiment of the application, the combination of the vehicle reaction time lag, the vehicle speed value and the vehicle acceleration to calculate the follow-up braking distance of each vehicle in the traveling vehicle comprises:

[0094]

[0095] wherein H represents the follow-up braking distance of each vehicle in the traveling vehicle, B i represents the vehicle speed value corresponding to the i-th vehicle in the traveling vehicle, K i represents the vehicle reaction time lag of the i-th vehicle in the traveling vehicle, a i represents the vehicle acceleration corresponding to the i-th vehicle in the traveling vehicle, and i represents the serial number of the traveling vehicle.

[0096] S4, combine the follow-up braking distance and the front vehicle braking distance to calculate the rear-end risk coefficient corresponding to each vehicle in the traveling vehicle, and combine the rear-end risk coefficient, the speed variation value and the regional resistance coefficient to analyze the traffic flow stability corresponding to the traffic area.

[0097] The application combines the following: the rear vehicle braking distance and the front vehicle braking distance, calculates the rear-end risk coefficient corresponding to each vehicle in the traveling vehicle, can understand the high and low of the rear-end risk of each vehicle in the traveling vehicle, provides a basis for improving the analysis accuracy of traffic flow stability, further, the rear-end risk coefficient corresponding to each vehicle in the traveling vehicle can be calculated by the ratio between the rear vehicle braking distance and the front vehicle braking distance, the larger the ratio of the rear vehicle braking distance to the front vehicle braking distance, the higher the rear-end risk; the analysis of the traffic flow stability corresponding to the traffic area can be realized by matrix analysis method, the rear-end risk coefficient, the speed variation value and the area resistance coefficient are constructed into a stability analysis matrix, the matrix row represents the corresponding evaluation index, such as the rear-end risk, the speed variation index and the area resistance index, the matrix list represents the corresponding numerical value, that is, the rear-end risk coefficient, the speed variation value and the area resistance coefficient, according to the numerical value corresponding to the evaluation index, the corresponding index level is evaluated, such as the rear-end risk coefficient, the speed variation value and the area resistance coefficient are 0.8, 0.9 and 0.8 respectively, which indicates that the rear-end risk is high risk, the speed fluctuation is large, and the congestion degree is large, so the traffic flow stability corresponding to the traffic area is poor.

[0098] The application calculates the speed variation value and the area resistance coefficient of the traffic area by combining the vehicle speed value and the number of vehicles, can understand the speed variation degree of the traffic area and the congestion degree of the area, so as to analyze the traffic flow stability of the traffic area subsequently, the application can convert the vehicle video data into corresponding individual frames by frame processing, so that each frame can be processed individually, reducing the calculation complexity, the application can obtain the area representation in the target video frame by extracting the area features corresponding to the target video frame, providing a basis for subsequent positioning of the vehicle position of the traveling vehicle in the traffic area, the application combines the rear vehicle braking distance and the front vehicle braking distance, calculates the rear-end risk coefficient corresponding to each vehicle in the traveling vehicle, can understand the high and low of the rear-end risk of each vehicle in the traveling vehicle, provides a basis for improving the analysis accuracy of traffic flow stability. Therefore, the intelligent networked traffic flow stability analysis method provided by the embodiment of the application can improve the accuracy of intelligent networked traffic flow stability analysis.

[0099] As Figure 2 shown, is a functional module diagram of an intelligent networked traffic flow stability analysis system provided by an embodiment of the application.

[0100] The intelligent networked traffic flow stability analysis system 100 can be installed in an electronic device. According to the functions implemented, the intelligent networked traffic flow stability analysis system 100 can include a regional hysteresis coefficient calculation module 101, a vehicle acceleration calculation module 102, a brake stroke calculation module 103, and a stability analysis module 104. The modules in the present application can also be referred to as units, which refer to a series of computer program segments that can be executed by an electronic device processor and can complete a fixed function, which are stored in the memory of the electronic device.

[0101] In the present embodiment, the functions of each module / unit are as follows:

[0102] The regional hysteresis coefficient calculation module 101 is configured to obtain a traffic area to be analyzed, collect vehicle video data of a vehicle driving in the traffic area using a preset camera device, and use a preset intelligent networked system to count a vehicle speed value and a vehicle quantity corresponding to the vehicle driving in the traffic area. The speed variation value and the regional hysteresis coefficient of the traffic area are calculated in combination with the vehicle speed value and the vehicle quantity.

[0103] The vehicle acceleration calculation module 102 is configured to perform frame processing on the vehicle video data to obtain a target video frame, detect a frame sequence corresponding to the target video frame, and calculate a vehicle acceleration corresponding to the vehicle driving in the traffic area according to the frame sequence and the vehicle speed value.

[0104] The brake stroke calculation module 103 is configured to extract a regional feature corresponding to the target video frame, locate a vehicle position of the vehicle driving in the traffic area according to the regional feature, calculate a following vehicle brake stroke corresponding to each vehicle in the vehicle driving in the traffic area in combination with the vehicle acceleration, the vehicle speed value, and the vehicle position, query a vehicle reaction time lag corresponding to the vehicle driving in the traffic area, and calculate a preceding vehicle brake stroke corresponding to each vehicle in the vehicle driving in the traffic area in combination with the vehicle reaction time lag, the vehicle speed value, and the vehicle acceleration.

[0105] The stability analysis module 104 is configured to calculate a rear-end collision risk coefficient corresponding to each vehicle in the vehicle driving in the traffic area in combination with the following vehicle brake stroke and the preceding vehicle brake stroke, and analyze the traffic flow stability of the traffic area in combination with the rear-end collision risk coefficient, the speed variation value, and the regional hysteresis coefficient.

[0106] In detail, each module in the intelligent networked traffic flow stability analysis system 100 in the present embodiment uses the same technical means as the intelligent networked traffic flow stability analysis method described in the above Figure 1 , and can produce the same technical effects, which will not be described here again.

[0107] In several embodiments provided by the present application, it should be understood that the provided method and system can be implemented in other manners. For example, the embodiments of the method described above are merely schematic. For example, the division of the modules is merely a logical function division. In actual implementation, there can be another division manner.

[0108] Finally, it should be noted that the above embodiments are merely used for describing, but not limiting the present application. Although the present application is described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present application can be modified or equivalent replaced without departing from the spirit and scope of the present application.

Claims

1. A method for intelligent connected traffic flow stability analysis, characterized in that, The method comprises: acquiring a traffic area to be analyzed, collecting vehicle video data of a vehicle driving in the traffic area by using a preset camera device, and counting a vehicle speed value and a vehicle quantity corresponding to the vehicle driving by using a preset intelligent network system, and combining the vehicle speed value and the vehicle quantity to calculate a speed variation value and a regional resistance coefficient of the traffic area; frame processing is performed on the vehicle video data to obtain a target video frame, a frame sequence corresponding to the target video frame is detected, and a vehicle acceleration corresponding to the vehicle driving is calculated according to the frame sequence and the vehicle speed value; regional features corresponding to the target video frame are extracted, a vehicle position of the vehicle driving in the traffic area is located according to the regional features, and a following brake travel corresponding to each vehicle in the vehicle driving is calculated in combination with the vehicle acceleration, the vehicle speed value and the vehicle position, a vehicle reaction time lag corresponding to the vehicle driving is queried, and a front vehicle brake travel corresponding to each vehicle in the vehicle driving is calculated in combination with the vehicle reaction time lag, the vehicle speed value and the vehicle acceleration; the following brake travel and the front vehicle brake travel are combined to calculate a rear-end collision risk coefficient corresponding to each vehicle in the vehicle driving, and traffic flow stability corresponding to the traffic area is analyzed in combination with the rear-end collision risk coefficient, the speed variation value and the regional resistance coefficient. 2.The intelligent network traffic flow stability analysis method of claim 1, wherein, The combination of the vehicle speed value and the vehicle quantity to calculate the speed variation value and the regional resistance coefficient of the traffic area comprises: a vehicle speed time sequence corresponding to the vehicle speed value is identified, and a time sequence average speed corresponding to the vehicle speed time sequence is calculated according to the vehicle speed value; the speed variation value of the traffic area is calculated in combination with the time sequence average speed and the vehicle speed value; the regional resistance coefficient of the traffic area is calculated in combination with the vehicle speed time sequence, the vehicle speed value and the vehicle quantity. 3.The intelligent network traffic flow stability analysis method of claim 2, wherein, The combination of the vehicle speed time sequence, the vehicle speed value and the vehicle quantity to calculate the regional resistance coefficient of the traffic area comprises: a time sequence length corresponding to the vehicle speed time sequence is calculated; a resistance vehicle speed value in the vehicle speed value is identified according to a preset resistance speed threshold value; a resistance vehicle number corresponding to the resistance vehicle speed value is determined; the regional resistance coefficient of the traffic area is calculated in combination with the resistance vehicle number, the time sequence length and the vehicle quantity. 4.The intelligent network traffic flow stability analysis method of claim 1, wherein, The frame processing on the vehicle video data to obtain a target video frame comprises: the vehicle video data is frame processed by using a preset frame tool to obtain an initial video frame; noise reduction processing is performed on the initial video frame to obtain a noise reduction video frame; deblurring processing is performed on the noise reduction video frame to obtain a clear video frame; color enhancement processing is performed on the clear video frame to obtain a target video frame. 5.The intelligent network-connected traffic flow stability analysis method of claim 1, wherein, The calculation of the vehicle acceleration corresponding to the vehicle driving according to the frame sequence and the vehicle speed value comprises: mapping processing is performed on the vehicle speed value and the frame sequence to obtain a frame speed value; a vehicle speed interval corresponding to the frame speed value is calculated according to the frame sequence; a vehicle speed variation corresponding to the frame sequence is calculated in combination with the frame speed value; The vehicle acceleration corresponding to the driving vehicle is calculated in combination with the vehicle speed variation and the vehicle speed interval. 6.The intelligent network-connected traffic flow stability analysis method of claim 1, wherein, The region feature corresponding to the target video frame is extracted, including: The target video frame is subjected to vehicle identification to obtain an identified vehicle, and a vehicle surrounding region corresponding to the identified vehicle is determined; The target video frame is subjected to image segmentation processing according to the vehicle surrounding region and the identified vehicle to obtain a segmented region image; The region color feature corresponding to the segmented region image is extracted, and the segmented region image is subjected to gray processing to obtain a gray region image; The region texture feature corresponding to the gray region image is extracted; The region feature corresponding to the target video frame is determined in combination with the region color feature and the region texture feature.

7. The intelligent network traffic flow stability analysis method of claim 6, wherein, The region feature corresponding to the target video frame is determined in combination with the region color feature and the region texture feature, including: A color feature value corresponding to the region color feature is calculated, and a color dispersion degree of each feature in the region color feature is calculated according to the color feature value; The region color feature is subjected to filtering processing according to the color dispersion degree to obtain a first region feature; The region texture feature is subjected to quantization processing to obtain a texture feature value, and a probability density corresponding to the texture feature value is calculated; A texture entropy corresponding to the region texture feature is calculated according to the probability density; The region texture feature is subjected to screening processing according to the texture entropy to obtain a second region feature; The first region feature and the second region feature are subjected to feature fusion to obtain the region feature corresponding to the target video frame. 8.The intelligent network-connected traffic flow stability analysis method of claim 1, wherein, The front vehicle brake stroke corresponding to each vehicle in the driving vehicle is calculated in combination with the vehicle acceleration, the vehicle speed value and the vehicle position, including: A vehicle body length parameter corresponding to the driving vehicle is queried; A vehicle head distance between the driving vehicles is determined according to the vehicle position; The front vehicle brake stroke corresponding to each vehicle in the driving vehicle is calculated in combination with the vehicle head distance, the vehicle body length parameter, the vehicle acceleration and the vehicle speed value.

9. An intelligent connected vehicle flow stability analysis system, characterized in that, The system includes: A region drag coefficient calculation module is configured to acquire a traffic region to be analyzed, collect vehicle video data of driving vehicles in the traffic region by using a preset camera device, and calculate a speed variation value and a region drag coefficient of the traffic region in combination with a vehicle speed value and a vehicle number of the driving vehicles by using a preset intelligent network system; A vehicle acceleration calculation module is configured to perform frame processing on the vehicle video data to obtain a target video frame, detect a frame sequence corresponding to the target video frame, and calculate a vehicle acceleration corresponding to the driving vehicles according to the frame sequence and the vehicle speed value. The brake stroke calculation module is configured to extract the region feature corresponding to the target video frame, locate the vehicle position of the running vehicle in the traffic region according to the region feature, calculate the following vehicle brake stroke corresponding to each vehicle in the running vehicle in combination with the vehicle acceleration, the vehicle speed value and the vehicle position, query the vehicle reaction time lag corresponding to the running vehicle, and calculate the preceding vehicle brake stroke corresponding to each vehicle in the running vehicle in combination with the vehicle reaction time lag, the vehicle speed value and the vehicle acceleration. The stability analysis module is configured to calculate the rear-end risk coefficient corresponding to each vehicle in the running vehicle in combination with the following vehicle brake stroke and the preceding vehicle brake stroke, and analyze the traffic flow stability corresponding to the traffic region in combination with the rear-end risk coefficient, the speed variation value and the region resistance coefficient.

Images