ETC Data-based Service Area Congestion Prediction Method and Information Release and Induction System

Abstract

The present invention provides a congestion prediction method and information release and guidance system based on ETC data, including standardizing key features extracted from camera capture images at ramp entrances and exits, mapping them to corresponding discrete data ranges to form discrete data points, treating the data points as mass points and allowing them to move under the action of gravity towards a set initial center position to form a circular structure, determining the radius of the circle by calculating the average distance and dividing the circle according to the distribution of data points, comparing the data point distribution patterns in the corresponding circular areas of the capture data at the entrance and exit, and generating a first ID code. The present invention accurately predicts the congestion trend by integrating ETC gantry data and multi-source information, which helps to improve the traffic efficiency and management level of the service area and alleviate the traffic congestion situation.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence technology, and particularly to a congestion prediction method and information release and guidance system for service areas based on ETC data. Background Art

[0002] As an important part of expressways, service areas not only provide basic rest services for drivers and passengers, effectively relieve long-distance driving fatigue and improve driving safety, but also undertake functions of information release and traffic guidance, assisting in relieving road congestion and optimizing road network traffic efficiency. They are key nodes for ensuring the safe and efficient operation of expressways.

[0003] However, at the current stage, the monitoring of vehicle carrying capacity in expressway service areas mainly relies on visual solutions, that is, vehicle data is collected by capturing images with cameras. Although this solution is feasible in service areas with complete hardware equipment, its high hardware investment and maintenance costs limit the popularization scope and it is difficult to be widely applied in all service areas. In addition, the visual solution is also limited by weather and lighting environment factors, and the stability and accuracy of data collection are difficult to fully guarantee. Summary of the Invention

[0004] The technical problem to be solved by the present invention is to provide a congestion prediction method and information release and guidance system for service areas based on ETC data. By integrating ETC gantry data with multi-source information, the congestion trend can be accurately predicted, which helps to improve the traffic efficiency and management level of service areas and relieve traffic congestion.

[0005] To solve the above technical problem, the technical solution of the present invention is as follows:

[0006] In the first aspect, a congestion prediction method for service areas based on ETC data, the method includes:

[0007] Standardize the key features extracted from the captured images of the ramp entrance and exit checkpoints, map them to the corresponding discrete data ranges to form discrete data points, regard the data points as mass points and let them move under the gravitational force towards the set initial center position to form a circular structure, determine the radius of the circle by calculating the average distance and divide the circle according to the data point distribution, compare the data point distribution patterns in the corresponding circular areas of the captured data at the entrance and exit, and generate a first ID code;

[0008] Based on the first ID code, associate the captured data of the front and rear gantries of the service area, generate a second ID code, screen and compare vehicle information, and calculate the interval passing speed according to the time difference and distance between gantries to obtain the predicted vehicles entering the service area;

[0009] Capture and collect the real-time data of vehicles entering the service area through the checkpoint camera, and dynamically judge the current real-time congestion index of the service area according to the predicted vehicles entering the service area;

[0010] Calculate the estimated arrival time based on the remaining journey of the vehicle to the service area entrance, and calculate the traffic flow contribution value for different types of vehicles at the same section of the service area entrance according to the real-time congestion index;

[0011] Based on the historical passing behavior data of vehicles, establish a vehicle portrait library, and predict the passing behavior of vehicles on the road section through a neural network model to generate the prediction results of passing behavior in the future time period.

[0012] Furthermore, the key features include the appearance color, vehicle type, license plate number, and vehicle contour of the vehicle; standardize the key features extracted from the captured images of the ramp entrance and exit checkpoints, and map them to the corresponding discrete data range to form discrete data points. Treat the data points as mass points and let them move under the gravitational force to a set initial center position to form a circular structure. Determine the radius of the circle by calculating the average distance, and divide the circle according to the distribution of data points. Compare the data point distribution patterns in the corresponding circular areas of the captured data at the entrance and exit, and generate the first ID code, including:

[0013] Standardize each key feature to obtain the standardized key features, specifically including: convert the appearance color to a fixed color classification code, convert the vehicle type to the corresponding vehicle type number, convert the license plate number to a digital code, and convert the vehicle contour to a numerical feature through shape feature extraction;

[0014] Based on the standardized key features, determine the discrete data range corresponding to each key feature, map the specific value of each key feature to the corresponding discrete data range to form the corresponding discrete data points, so that each piece of captured data at the entrance and exit can be represented as a set of discrete data points;

[0015] Regard the discrete data points as mass points with gravitational force, set an initial center position, and let each discrete data point move towards the center position under the gravitational force to finally form a circular structure;

[0016] Determine the radius of the circle by calculating the average distance from all data points to the center of the circle, and divide the circle into corresponding parts according to the distribution of data points of different key features;

[0017] Compare the data point distribution in the circular area corresponding to each piece of captured data at the entrance with the data point distribution in the circular area corresponding to all captured data at the exit. If the data point distribution patterns in the divided areas of the circle are the same, that is, the number and position characteristics of the data points in the corresponding areas are similar, then it is determined that the captured data at the entrance and exit are the same vehicle, and generate the corresponding first ID code.

[0018] Further, based on the first ID code, the captured data of the front and rear gantries of the service area are associated to generate a second ID code. The vehicle information is compared and screened, and the interval passing speed is calculated according to the time difference and distance between the gantries, so as to obtain the vehicles predicted to enter the service area, including:

[0019] Associate and match the captured data of the front and rear gantries of the service area, and generate a second ID code according to the capture time, position and characteristic information of the vehicle at the front and rear gantries of the service area to identify the vehicles passing through the front and rear gantries of the service area;

[0020] Compare the first ID code with the second ID code to obtain the screened target vehicle information;

[0021] Based on the screened target vehicle information, calculate the passing time difference and gantry distance between the front and rear gantries of the service area; calculate the interval passing speed of the vehicle according to the passing time difference and gantry distance;

[0022] According to the interval passing speed of the vehicle and the preset threshold, predict the list of vehicles entering the service area.

[0023] Further, capture and collect the real-time data of vehicles entering the service area through the bayonet camera, and dynamically judge the current real-time congestion index of the service area according to the vehicles predicted to enter the service area, including:

[0024] According to the vehicle data captured by the bayonet camera, record the entrance capture data and the exit capture data;

[0025] Calculate the number of vehicles entering the service area and the number of vehicles leaving the service area per unit time;

[0026] Calculate the number of vehicles in the service area, and judge the real-time congestion index of the service area according to the number of vehicles in the area and the carrying capacity of the service area.

[0027] Further, according to the remaining journey of the vehicle to the service area entrance, calculate the estimated arrival time, and calculate the flow contribution value for different types of vehicles at the same section of the service area entrance according to the real-time congestion index, including:

[0028] According to the journey of the vehicle from the node to the service area entrance, calculate the time when the vehicle arrives at the service area entrance;

[0029] At the same section of the service area entrance, obtain all different types of vehicles;

[0030] Obtain the equivalent number of each type of vehicle in terms of flow compared to the standard vehicle and the time interval of the section;

[0031] Combine the equivalent number of each type of vehicle in terms of traffic flow relative to the standard vehicle with the time interval of the cross-section to obtain the contribution of each type of vehicle relative to the standard vehicle to the traffic flow per unit time, where the contribution represents vehicle density;

[0032] Adjust the contribution of each type of vehicle relative to the standard vehicle to the traffic flow according to historical actual data to obtain the traffic flow contribution value;

[0033] Add up the traffic flow contribution values of each type of vehicle to obtain the total traffic flow of all vehicles arriving at the same cross-section at the same moment at the service area entrance.

[0034] Furthermore, based on the historical passing behavior data of vehicles, establish a vehicle portrait library, and predict the passing behavior of vehicles on the road section through a neural network model to generate the prediction results of passing behavior in the future time period, including:

[0035] Integrate the basic information and passing records of vehicles into the static database and the dynamic database;

[0036] Construct portrait tags according to the historical passing behavior of vehicles, and dynamically update the vehicle portrait library when new passing data is received;

[0037] Extract key features from the vehicle portrait and determine the influence weights of different features on the passing behavior;

[0038] When the vehicle enters the highway again, predict the service area it passes through according to historical data and output the prediction results of passing behavior.

[0039] Furthermore, extract key features from the vehicle portrait and determine the influence weights of different features on the passing behavior, including:

[0040] Extract key features from the vehicle portrait, including vehicle type, vehicle size, vehicle weight, driving speed, driving mileage, driver behavior pattern, time period, weather condition, and historical passing records;

[0041] For the key features, perform quantization processing to obtain numerical features;

[0042] Construct a random forest according to the extracted and quantized features. For each tree in the random forest, traverse each node in the tree. For each node, determine the number of samples contained in the node and calculate the Gini impurity of the node;

[0043] Split the features, divide the samples in the node into the left child node and the right child node, and calculate the Gini impurity of the left child node and the right child node respectively;

[0044] Calculate the reduction in Gini impurity according to the Gini impurity of the node and the Gini impurities of the left child node and the right child node;

[0045] For each feature, sum up the reduction in Gini impurity brought about by splitting in all the trees to obtain the importance score of the feature.

[0046] Average the importance scores of each feature to obtain the final importance score of the feature in the random forest.

[0047] Based on the final importance score of the feature in the random forest, determine the influence weights of different features on the passing behavior.

[0048] In a second aspect, an information release and induction system for service area congestion based on ETC data includes:

[0049] A checkpoint information module, which is used to standardize the key features extracted from the captured images of the ramp entrance and exit checkpoints, map them to the corresponding discrete data ranges to form discrete data points, regard the data points as mass points and let them move under the gravitational force towards the set initial center position to form a circular structure, determine the radius of the circle by calculating the average distance and divide the circle according to the distribution of the data points, compare the data point distribution patterns within the corresponding circular areas of the captured data at the entrance and exit, and generate a first ID code; based on the first ID code, associate the captured data of the front and rear gantries of the service area, generate a second ID code, screen the vehicle information through comparison, and calculate the interval passing speed according to the time difference and spacing between the gantries to obtain the vehicles predicted to enter the service area.

[0050] A multi-source recognition module, which is used to capture and collect the real-time data of vehicles entering the service area through the checkpoint cameras, and dynamically judge the current real-time congestion index of the service area according to the vehicles predicted to enter the service area; calculate the estimated arrival time according to the remaining journey of the vehicle to the service area entrance, and calculate the flow contribution value for different types of vehicles at the same section of the service area entrance according to the real-time congestion index.

[0051] A vehicle trajectory prediction module, which is used to establish a vehicle portrait library based on the historical passing behavior data of the vehicle, and predict the passing behavior of the vehicle on the road section through a neural network model to generate the prediction results of the passing behavior in the future time period.

[0052] In a third aspect, a computing device includes:

[0053] One or more processors;

[0054] A storage device for storing one or more programs, which when executed by the one or more processors, cause the one or more processors to implement the method.

[0055] In a fourth aspect, a computer-readable storage medium stores a program that implements the method when executed by a processor.

[0056] The above solution of the present invention has at least the following beneficial effects:

[0057] By standardizing the key features extracted from the captured images of the ramp entrance and exit checkpoints, mapping them to a discrete data range to form data points, and using the gravitational effect to form a circular structure for comparison, it is possible to more accurately identify and track vehicles, generate a unique first ID code, and effectively solve the problems of error and missed detection in traditional vehicle identification methods. Based on the first ID code, the captured data of the front and rear gantries in the service area are associated to generate a second ID code. By comparing and screening vehicle information, and combining the time difference and distance between gantries to calculate the interval passing speed, it is possible to accurately predict the vehicles entering the service area. By capturing and collecting the real-time data of vehicles entering the service area through the checkpoint cameras, and combining the predicted vehicle information entering the service area, the current real-time congestion index of the service area can be dynamically judged, which helps to take effective dredging measures. By calculating the remaining journey of the vehicle to the service area entrance to estimate the expected arrival time, and combining the real-time congestion index to calculate the traffic flow contribution value of different types of vehicles at the same section at the service area entrance, it is possible to more scientifically evaluate the impact of different vehicles on traffic flow. Based on the historical passing behavior data of vehicles, a vehicle portrait library is established, and the passing behavior of vehicles on the road section is predicted through a neural network model to generate the predicted results of passing behavior in the future period, which helps to plan traffic flow in advance and reduce congestion and traffic accidents.

[0058] The invention can predict the congestion situation of the service area in real time and accurately, provide scientific decision-making support for the service area management, and help improve the management efficiency and service quality of the service area. By calculating the traffic flow contribution value and predicting the passing behavior, the invention can optimize the distribution of traffic flow, reduce congestion during peak hours, and improve the road passing capacity. The provision of the real-time congestion index and the expected arrival time helps drivers plan their trips in advance, avoid congested hours and sections, and enhance the driving experience. By reducing congestion and traffic accidents, it is possible to reduce the economic losses and social costs brought by traffic congestion, improve the utilization efficiency of road resources. Accurate passing behavior prediction and congestion warning help to improve the public safety level, reduce secondary disasters and casualties caused by congestion and traffic accidents, help to optimize the distribution of traffic flow, reduce energy consumption and environmental pollution caused by vehicle idling and frequent start-stop, and promote the sustainable development of the transportation field. Brief Description of the Drawings

[0059] Figure 1 is a schematic flow chart of the service area congestion prediction method based on ETC data provided by an embodiment of the present invention.

[0060] Figure 2 is a schematic diagram of the information release and induction system for service area congestion based on ETC data provided by an embodiment of the present invention. Detailed implementation manners

[0061] Exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although the exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be implemented in forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided so that the present disclosure can be more thoroughly understood and the scope of the present disclosure can be completely conveyed to those skilled in the art.

[0062] As Figure 1 shown, an embodiment of the present invention provides a method for predicting service area congestion based on ETC data, and the method includes the following steps:

[0063] Step 11: Standardize the key features extracted from the captured images of the ramp entrance and exit checkpoints, map them to the corresponding discrete data ranges to form discrete data points, regard the data points as mass points and let them move under the gravitational force towards the set initial center position to form a circular structure, determine the radius of the circle by calculating the average distance, and divide the circle according to the distribution of the data points. Compare the data point distribution patterns in the corresponding circular areas of the captured data at the entrance and exit, and generate a first ID code;

[0064] Step 12: Based on the first ID code, associate the captured data of the front and rear gantries of the service area, generate a second ID code, compare and screen vehicle information, and calculate the interval passing speed according to the time difference and distance between the gantries to obtain the vehicles predicted to enter the service area;

[0065] Step 13: Collect the real-time data of vehicles entering the service area through the checkpoint camera, and dynamically judge the current real-time congestion index of the service area according to the vehicles predicted to enter the service area;

[0066] Step 14: Calculate the estimated arrival time according to the remaining journey of the vehicle to the service area entrance, and calculate the flow contribution value for different types of vehicles in the same section at the service area entrance according to the real-time congestion index;

[0067] Step 15: Establish a vehicle portrait library based on the historical passing behavior data of vehicles, and predict the passing behavior of vehicles on the road section through a neural network model to generate the prediction result of the passing behavior in the future time period.

[0068] In the embodiments of the present invention, standardizing the key features extracted from the captured images of the ramp entrance and exit checkpoints and mapping them to a discrete data range to form discrete data points helps to eliminate the dimensional differences between data and improve the accuracy and efficiency of data processing. Regarding the data points as mass points and forming a circular structure further simplifies the data representation. By comparing the distribution patterns of data points within the corresponding circular regions of the captured data at the entrance and exit, a first ID code is generated, enabling precise identification and tracking of vehicles. Based on the first ID code, the captured data of the front and rear gantries in the service area are associated to generate a second ID code, and the vehicle information is screened by comparison. Calculating the interval passing speed based on the time difference and distance between gantries helps to understand the driving state of vehicles on the road section. Real-time data of vehicles entering the service area are collected through the captured images of the checkpoints, and the current real-time congestion index of the service area is judged based on the predicted dynamics of vehicles entering the service area, realizing real-time monitoring and early warning of the congestion status in the service area. Calculating the estimated arrival time based on the remaining journey of the vehicle to the service area entrance helps to understand the specific time when the vehicle arrives at the service area. Calculating the flow contribution value for different types of vehicles at the same section of the service area entrance based on the real-time congestion index helps to understand the impact degree of different types of vehicles on the service area flow. Establishing a vehicle portrait library based on the historical passing behavior data of vehicles helps to understand the passing habits and preferences of vehicles.

[0069] In a preferred embodiment of the present invention, step 11 may include:

[0070] Step 111, standardizing each key feature to obtain the standardized key features, specifically including: converting the appearance color into a fixed color classification code, converting the vehicle type into the corresponding vehicle type number, converting the license plate number into a digital code, and converting the vehicle contour into a numerical feature through shape feature extraction;

[0071] Step 112, based on the standardized key features, determining the discrete data range corresponding to each key feature, mapping the specific value of each key feature into the corresponding discrete data range to form the corresponding discrete data points, so that each piece of captured data at the entrance and exit can be represented as a set of discrete data points;

[0072] Step 113, regarding the discrete data points as mass points with gravity, setting an initial central position, and allowing each discrete data point to move towards the central position under the action of gravity, finally forming a circular structure;

[0073] Step 114, determining the radius of the circle by calculating the average distance from all data points to the center of the circle, and dividing the circle into corresponding parts according to the distribution of data points of different key features;

[0074] Step 115: Compare the data point distribution within the circular area corresponding to each piece of entrance capture data with the data point distribution within the circular area corresponding to all exit capture data. If the data point distribution patterns within the divided areas of the circles are the same, that is, the number and position characteristics of the data points in the corresponding areas are similar, then determine that the entrance and exit capture data are for the same vehicle, and generate a corresponding first ID code.

[0075] In the embodiment of the present invention, key features such as appearance color, vehicle type, license plate number, and vehicle contour are converted into a unified code or numerical form, eliminating differences between different data sources or formats, improving data consistency and comparability. The standardized data is more easily processed and analyzed by computer programs, improving the efficiency and accuracy of data processing. Mapping continuous or complex feature values to discrete numerical values or categories simplifies the data representation method. Representing data points as a circular structure provides an intuitive data visualization method, helping to observe and analyze the distribution and characteristics of the data. By calculating the radius of the circle, the distribution range of the data points can be quantified. By comparing the data point distribution patterns of the entrance and exit capture data within the circular area, it is possible to accurately identify whether it is the same vehicle, improving the accuracy and reliability of vehicle identification. Generating a corresponding first ID code for the data determined to be the same vehicle provides a unique identifier for the vehicle. By accurately identifying the vehicle identity, traffic flow statistics, violation behavior monitoring, traffic incident handling, etc. can be carried out more effectively, improving the efficiency and level of traffic management.

[0076] In the embodiment of the present invention, the specific steps include:

[0077] Step 111: Convert the appearance color of the vehicle into a fixed color classification code. For example, red may correspond to the code "R", blue corresponds to the code "B", and so on. Convert the vehicle type into the corresponding vehicle type number. For example, a sedan may correspond to the number "1", an SUV corresponds to the number "2", a truck corresponds to the number "3", etc. This numbering system helps to unify the representation of vehicle type information. Convert the license plate number into a digital code. This can be achieved by mapping each character in the license plate number to a number, such as mapping letters A-Z to 1-26 and keeping numbers 0-9 unchanged, and then combining these numbers into a long integer or coded string. By identifying key points or features of the vehicle contour, shape feature extraction is performed to convert the vehicle contour into numerical features.

[0078] Step 112: For each standardized key feature, determine the value range and divide this range into multiple discrete intervals or categories. For example, for the vehicle model number, the value range is from 1 to 10, which can be divided into 10 discrete categories, each category corresponding to a vehicle model number. Map the specific value of each key feature to its corresponding discrete data range to form a discrete data point. For example, if the vehicle model number of a car is "3", it is mapped to the category corresponding to the discrete data point "3". For each piece of entrance capture data and exit capture data, map the values of all its key features to the corresponding discrete data points to form a set of discrete data points.

[0079] Step 113: Treat each discrete data point as a mass point with gravity. A mass point has no size and shape, only mass. Here, the mass point represents an abstract representation of the vehicle feature information. Set an initial central position, which is the geometric center of the data point set or a position determined according to a certain rule. Under the action of gravity, let each discrete data point move towards the central position until the data points form a stable circular structure.

[0080] Step 114: Determine the radius of the circle by calculating the average distance from all data points to the center of the circle. The radius represents the distribution range of the data point set. According to the distribution of data points of different key features, divide the circle into corresponding parts. For example, if a certain key feature has 10 possible values, the circle can be divided into 10 sectors, each sector corresponding to a value.

[0081] Step 115: Compare the distribution of data points within the circular area corresponding to each piece of entrance capture data with the distribution of data points within the circular area corresponding to all exit capture data. If the distribution patterns of data points within the divided areas of the circle are the same, that is, the number and position characteristics of data points in the corresponding areas are similar, it is determined that the entrance and exit capture data are of the same vehicle. For the data determined to be of the same vehicle, generate a corresponding first ID code.

[0082] In a preferred embodiment of the present invention, the above step 12 may include:

[0083] Step 121: Correlate and match the capture data of the front gantry and the rear gantry of the service area, and generate a second ID code based on the capture time, position, and feature information of the vehicle at the front and rear gantries of the service area to identify the vehicle passing through the front and rear gantries of the service area;

[0084] Step 122: Compare the first ID code with the second ID code to obtain the filtered target vehicle information;

[0085] Step 123: Based on the filtered target vehicle information, calculate the passing time difference and the gantry distance between the front and rear gantries of the service area for the vehicle; calculate the passing speed of the vehicle between the gantries according to the passing time difference and the gantry distance.

[0086] Step 124: Predict the list of vehicles entering the service area according to the passing speed of the vehicle between the gantries and a preset threshold.

[0087] In the embodiment of the present invention, by associating and matching the entrance capture data with the exit capture data to generate a first ID code, and associating and matching the capture data of the front gantry and the rear gantry of the service area to generate a second ID code, it is possible to accurately track and identify the vehicles passing through the ramp checkpoints and the front and rear gantries of the service area. By comparing the first ID code with the second ID code, it is possible to screen out the vehicle information that is not recorded in the ramp checkpoint capture data but appears in the capture data of the front and rear gantries of the service area, indicating abnormal passing behavior. By calculating the passing time difference and the gantry distance of the vehicle between the front and rear gantries of the service area, it is possible to accurately obtain the passing speed of the vehicle between the gantries, and then predict whether the vehicle will enter the service area, which helps the service area management department better understand the traffic conditions of the service area, optimize the resource allocation of the service area, and improve the overall management efficiency and service quality of the service area.

[0088] In the embodiment of the present invention, the specific steps include:

[0089] Step 121: Preprocess the collected image data, and use the license plate recognition technology to extract the license plate number of the vehicle from the entrance and exit capture images as the key feature. By Associating and matching the entrance capture data with the exit capture data to generate a first ID code to identify the vehicle passing through the ramp checkpoint, where is the association matching function, is the entrance capture data, is the exit capture data, and a unique identification code is generated according to the vehicle feature information; the association matching function will extract the vehicle feature information in E and X. Among them, the association matching function is the feature vector similarity calculation formula. Through the comparison and analysis of the feature vector similarity calculation, the capture records of the same vehicle at the entrance and the exit are found, and a unique first ID code is generated to realize the association and matching of the license plate number in the entrance capture image with the license plate number in the exit capture image. For the successfully matched vehicles, a unique first ID code is generated to identify the vehicles passing through the ramp checkpoint, and the information is associated and stored.

[0090] Step 122: Collect vehicle image data from the capture systems of the front gantry and the rear gantry of the service area, preprocess the collected image data to ensure that the images are clear and readable, and also use the license plate recognition technology to extract the license plate number of the vehicle from the front gantry and rear gantry capture images. By Associate and match the capture data of the front gantry in the service area with the capture data of the rear gantry to generate a second ID code to identify the vehicles passing through the front and rear gantries in the service area. Among them, is the capture data of the front gantry, is the capture data of the rear gantry, represents the association and matching function; the association and matching function will extract the vehicle feature information in A and B, compare and analyze through the calculation of the feature vector similarity, find the capture records of the same vehicle at the front and rear gantries, and generate a unique second ID code. Associate and match the license plate number in the capture image of the front gantry with the license plate number in the capture image of the rear gantry. For the successfully matched vehicles, generate a unique second ID code to identify the vehicles passing through the front and rear gantries in the service area and store the associated information.

[0091] Step 123, through Compare the first ID code with the second ID code, and filter out the vehicle information that is not recorded in the capture data of the ramp checkpoint but appears in the capture data of the front and rear gantries in the service area. Among them, represents the set of vehicle information filtered out; compare the first ID code with the second ID code to find the vehicle information that does not exist in the first ID code list but appears in the second ID code list.

[0092] Step 124, for each vehicle in the second ID code list, calculate the passing time difference of the vehicle at the front and rear gantries in the service area, and use the passing time difference and the gantry spacing to calculate the interval passing speed of the vehicle through Set a threshold, and according to the threshold, predict whether the vehicle enters the service area, and the vehicle list. Among them, is the interval passing speed, is the passing time difference between the front and rear gantries, is the gantry spacing.

[0093] In a preferred embodiment of the present invention, the above step 13 may include:

[0094] Step 131, record the entrance capture data and the exit capture data according to the vehicle data captured by the checkpoint camera;

[0095] Step 132, calculate the number of vehicles entering the service area and the number of vehicles leaving the service area per unit time;

[0096] Step 133, calculate the number of vehicles in the service area, and judge the real-time congestion index of the service area according to the number of vehicles in the area and the carrying capacity of the service area.

[0097] In the embodiments of the present invention, vehicle data is captured by a bayonet camera, and the captured data at the entrance and exit is recorded in detail, which can ensure the accurate tracking and recording of each vehicle entering and leaving the service area. By counting the number of vehicles entering and leaving the service area per unit time, the traffic flow situation in the service area can be grasped in real time, which helps to timely detect changes in traffic flow and provide timely and effective information support for traffic management. By calculating the number of vehicles in the service area, the current vehicle occupancy in the service area can be accurately understood, providing a scientific basis for judging the congestion status of the service area.

[0098] In the embodiments of the present invention, the specific steps include:

[0099] Step 131: Organize the captured vehicle data, classify and store it according to the entrance and exit, and add a unique identifier to each record for subsequent data association and query.

[0100] Step 132: By Count the number of vehicles leaving the service area per unit time. By Count the number of vehicles leaving the service area per unit time, where is at a certain moment within the unit time , and the number of vehicles leaving the service area is counted through the entrance capture data , is the entrance capture data, is at a certain moment within the unit time , and the number of vehicles leaving the service area is counted through the exit capture data , is the exit capture data, is the index, is the index, is the number of vehicles leaving the service area per unit time, is the number of vehicles leaving the service area per unit time.

[0101] According to the above formula, count the number of vehicles entering and leaving the service area per unit time. In each time window, screen out the vehicle records entering the service area during this period from the entrance capture data and count the quantity. Similarly, screen out the vehicle records leaving the service area during this period from the exit capture data and count the quantity.

[0102] Step 133: According to the number of entering and leaving vehicles within each time window, calculate the number of vehicles in the service area at the end of this time period according to the above formula. The carrying capacity of the service area (such as the maximum number of parking spaces, road traffic capacity) is preset, and compare the calculated number of vehicles in the area with the carrying capacity of the service area. If the number of vehicles in the area is close to or exceeds the carrying capacity of the service area, it is determined that the service area is in a congested state. The congested state is divided into the following three categories: , where is the real-time congestion index of the service area.

[0103] In a preferred embodiment of the present invention, the above step 14 may include:

[0104] According to the remaining journey of the vehicle to the service area entrance, calculate the estimated arrival time;

[0105] According to the real-time congestion index, for different types of vehicles at the same section of the service area entrance, calculate the flow contribution value, where is the total amount, is all vehicle types, is the th type of vehicle's converted flow number, is the th type of vehicle's time interval passing through the section, is the weight factor, is the flow adjustment coefficient, is the exponential decay factor, is the decay constant, is the index.

[0106] In the embodiment of the present invention, by accurately calculating the estimated arrival time of the vehicle, the service area can make preparations in advance for reception, such as arranging parking spaces, preparing catering services, etc., so as to improve service efficiency and reduce vehicle waiting time. By calculating the flow contribution values of different types of vehicles, the composition and changing trend of traffic flow can be understood more accurately, providing decision-making support for traffic management departments, such as adjusting signal light duration, optimizing lane settings, etc. By analyzing the congestion index and flow contribution value in real time, the service area can timely discover the bottleneck points of traffic congestion and take corresponding measures for dredging, such as adding temporary lanes, guiding vehicles to detour, etc., thus effectively alleviating traffic congestion. Considering multiple factors such as vehicle remaining journey, real-time congestion index, vehicle type, etc., the analysis result is more comprehensive and accurate. By introducing parameters such as weight factor, flow adjustment coefficient and exponential decay factor, the calculation model can be flexibly adjusted according to the actual situation to adapt to different traffic scenarios and requirements.

[0107] In a preferred embodiment of the present invention, the above step 15 may include:

[0108] Step 151: Integrate the basic information of the vehicle and the passing records into the static database and the dynamic database;

[0109] Step 152: Construct portrait tags based on the historical passing behaviors of the vehicle, and dynamically update the vehicle portrait library when new passing data is received;

[0110] Step 153: Extract key features from the vehicle portrait, and determine the influence weights of different features on the passing behavior;

[0111] Step 154: When the vehicle enters the highway again, predict the service areas it will pass through based on historical data, and output the prediction results of the passing behavior.

[0112] In the embodiment of the present invention, integrating the basic information of the vehicle and the passing records into the static database and the dynamic database realizes the centralized management of data. The division of the static database and the dynamic database makes the storage and query of data more efficient. The static database stores the basic information of the vehicle, and the dynamic database stores the passing records of the vehicle. The combination of the two can quickly obtain the historical passing information of the vehicle. The integrated data supports multi-dimensional analysis, such as vehicle type, passing time, and passing section. By constructing portrait tags, the characteristics of the vehicle can be accurately described, such as the frequently traveled section, passing time preference, and habit of staying at service areas. When new passing data is received, the vehicle portrait library is dynamically updated to ensure the real-time and accuracy of the vehicle portrait, making the prediction results more reliable. The construction and update of the vehicle portrait help to provide personalized services for the vehicle, such as customized passing suggestions and preferential information, improving the user experience. Extracting key features from the vehicle portrait helps to discover the key factors affecting the passing behavior of the vehicle. By training the prediction model with historical data and determining the influence weights of different features on the passing behavior, by predicting the service areas that the vehicle may pass through, the passing route can be planned in advance to avoid congestion and delays, improving the passing efficiency. According to the prediction results, personalized service suggestions can be provided for the vehicle, such as recommending service areas and providing preferential information, improving the user experience and satisfaction.

[0113] In the embodiment of the present invention, the specific steps include:

[0114] Step 151: Collect the basic information and passing records of the vehicle. The basic vehicle information includes, but is not limited to, license plate number, vehicle model, vehicle type, and owner information; the passing records include the time, location, speed, and driving direction of the vehicle each time it passes through an ETC gantry or a bayonet camera. Integrate the collected basic vehicle information and passing records. For the static database, mainly store the basic attributes of the vehicle, and for the dynamic database, record the passing situation of the vehicle in real time, including the specific information of each pass through an ETC gantry or a bayonet camera. By establishing a unique identifier for each vehicle, such as the license plate number, associate the information in the static database and the dynamic database, and store the integrated data in the static database and the dynamic database respectively.

[0115] Step 152: Based on the historical passing behavior of the vehicle, construct portrait tags. Portrait tags are abstract descriptions of the vehicle's passing behavior, including driving habits, frequently visited locations, and vehicle type preferences, and construct unique portrait tags for each vehicle. When the system receives new passing data, dynamically update the vehicle portrait library in a timely manner.

[0116] Step 153: Extract key features from the vehicle portrait. These features can reflect the passing behavior patterns and preferences of the vehicle. Through random forest, determine the influence weights of different features on the passing behavior.

[0117] Step 154: When the vehicle enters the highway again, the system identifies the vehicle's identity information through an ETC gantry or a bayonet camera. By comparing this information with the records in the vehicle portrait library, determine the vehicle's historical passing behavior and portrait tags. Based on the vehicle's historical passing behavior and portrait tags, use the trained prediction model to calculate the service areas that the vehicle may pass through. During the prediction process, the system will comprehensively consider factors such as the current traffic conditions and the facilities of the service areas to improve the accuracy and practicality of the prediction. Output the prediction results to the driver or the traffic management department. The prediction results can include the names of the service areas that the vehicle may pass through, the arrival time, and the driving route information. Through this information, the driver can plan the driving route in advance to avoid congestion and delays; the traffic management department can also optimize the resource allocation and management strategies of the service areas according to the prediction results.

[0118] In a preferred embodiment of the present invention, the above step 153 may include:

[0119] Step 1531: Extract key features from the vehicle portrait, including vehicle type, vehicle size, vehicle weight, driving speed, driving mileage, driver behavior pattern, time period, weather condition, and historical passing records;

[0120] Step 1532: For the key features, perform quantization processing to obtain numerical features;

[0121] Step 1533: Construct a random forest based on the extracted and quantified features. For each tree in the random forest, traverse each node in the tree. If the node is split by a feature, calculate the reduction in Gini impurity of the node, where

[0122] is the number of classes, is the node, is the node is the proportion of samples in node that belong to class ; is the feature, is the feature When splitting, is the reduction in Gini impurity of node is the number of samples in node ; and are the numbers of samples in the left and right child nodes after splitting, and are the proportions of samples in the left and right child nodes after splitting that belong to class ; is the index;

[0123] Step 1534: For each feature, count the sum of the reductions in Gini impurity brought about by splitting in all trees to obtain the importance score of the feature;

[0124] Step 1535: Average the importance scores of each feature to obtain the final importance score of the feature in the random forest;

[0125] Step 1536: Determine the influence weights of different features on the passing behavior based on the final importance scores of the features in the random forest.

[0126] In the embodiments of the present invention, by extracting key features such as vehicle type, vehicle size, vehicle weight, driving speed, driving mileage, driver behavior pattern, time period, weather condition, and historical passing records, the state and driving environment of the vehicle can be comprehensively reflected. These features cover multiple aspects such as the vehicle's own attributes, driving state, external environment, and historical behavior, which helps to analyze the vehicle passing behavior more accurately. Converting non-numerical features into numerical features enables all features to be input into the model in a unified form. The quantization process helps to eliminate the dimensional differences between features and improve the convergence speed and prediction accuracy of the model. Random forest can effectively process high-dimensional data. By calculating the reduction in Gini impurity of nodes, the contribution of each feature to the model's prediction ability can be evaluated. By statistically summing up the reduction in Gini impurity brought by each feature through splitting in all trees and calculating the average value, the key features that have the greatest impact on the passing behavior can be identified. By determining the influence weights of different features on the passing behavior, the specific influence degree of each feature on the vehicle passing behavior can be quantified.

[0127] In the embodiments of the present invention, the specific steps include:

[0128] Step 1531: Collect detailed information of the vehicle from the traffic management system, ETC system, or other data sources, including vehicle type, vehicle size, vehicle weight, driving speed, driving mileage, driver behavior pattern, time period, weather condition, and historical passing records. Extract the key features listed above from the collected vehicle portrait data.

[0129] Step 1532: For non-numerical features, quantization processing is required. For example, the vehicle type can be encoded as a numerical value, such as sedan = 1, SUV = 2; the weather condition can be encoded as a numerical value, such as sunny = 1, rainy = 2, snowy = 3; for some features that can be measured numerically, directly retain their numerical form and perform data normalization.

[0130] Step 1533: Use the quantized features to construct a random forest. The random forest consists of multiple decision trees, and each tree is constructed by randomly sampling samples and features from the original dataset. For each tree in the random forest, traverse all its nodes. For each node, calculate its Gini impurity. Gini impurity is an index to measure the unevenness of the sample category distribution in the node. For each node, try to split it using each feature, and divide the samples in the node into a left child node and a right child node. Calculate the Gini impurity of the left child node and the right child node respectively. According to the Gini impurity of the node and the Gini impurities of the left child node and the right child node, calculate the reduction in Gini impurity through the above formula.

[0131] Step 1534: For each feature, calculate the sum of the Gini impurity reduction brought by splitting in all trees. This sum reflects the overall contribution of the feature to the classification task in the random forest. Take the sum of the Gini impurity reduction as the importance score of the feature.

[0132] Step 1535: Average the importance scores of each feature to obtain the final importance score of the feature in the random forest. This score reflects the average importance of the feature in the random forest.

[0133] Step 1536: According to the final importance score of the feature in the random forest, normalize the importance scores so that the sum of the weights of all features is 1. Determine the influence weights of different features on the passing behavior by dividing the importance score of each feature by the sum of the importance scores of all features. The higher the weight, the greater the influence of the feature on the passing behavior.

[0134] As Figure 2 shown, the embodiment of the present invention also provides an information release and induction system 20 for service area congestion based on ETC data, including:

[0135] The checkpoint information module 21 is used to standardize the key features extracted from the captured images of the ramp entrance and exit checkpoints, map them to the corresponding discrete data ranges to form discrete data points, regard the data points as mass points and let them move under the gravitational force towards the set initial center position to form a circular structure, determine the radius of the circle by calculating the average distance and divide the circle according to the distribution of the data points, compare the data point distribution patterns in the corresponding circular areas of the captured data at the entrance and exit, and generate the first ID code; based on the first ID code, associate the captured data of the front and rear gantries of the service area, generate the second ID code, screen the vehicle information by comparison, and calculate the interval passing speed according to the time difference and distance between the gantries to obtain the vehicles predicted to enter the service area.

[0136] The multi-source recognition module 22 is used to capture and collect the real-time data of vehicles entering the service area through the checkpoint cameras, and dynamically judge the current real-time congestion index of the service area according to the vehicles predicted to enter the service area; calculate the estimated arrival time according to the remaining journey of the vehicle to the service area entrance, and calculate the flow contribution value for different types of vehicles at the same section of the service area entrance according to the real-time congestion index.

[0137] The vehicle trajectory prediction module 23 is used to establish a vehicle portrait library based on the vehicle historical passing behavior data, and predict the passing behavior of the vehicle on the road section through a neural network model to generate the passing behavior prediction results for future time periods.

[0138] The above are the preferred embodiments of the present invention. It should be noted that for those of ordinary skill in the art, without departing from the principle of the present invention, several improvements and modifications can be made, and these improvements and modifications should also be regarded as the protection scope of the present invention.

Claims

1. A method for predicting service area congestion based on ETC data, characterized in that The method includes: Standardize the key features extracted from the captured images of the ramp entrance and exit bayonet cameras, map them to the corresponding discrete data ranges to form discrete data points, regard the data points as mass points, and let them move under the gravitational force towards the set initial center position to form a circular structure. Determine the radius of the circle by calculating the average distance and divide the circle according to the distribution of the data points. Compare the data point distribution patterns in the corresponding circular areas of the captured data at the entrance and exit, and generate a first ID code; Based on the first ID code, associate the captured data of the front and rear gantries of the service area, generate a second ID code, compare and screen vehicle information, and calculate the interval passing speed based on the time difference and distance between the gantries to obtain the vehicles predicted to enter the service area; Capture and collect the real-time data of vehicles entering the service area through the bayonet camera, and dynamically judge the current real-time congestion index of the service area according to the vehicles predicted to enter the service area; Calculate the estimated arrival time according to the remaining journey of the vehicle to the service area entrance, and calculate the flow contribution value for different types of vehicles in the same section at the service area entrance according to the real-time congestion index; Based on the historical passing behavior data of the vehicle, establish a vehicle portrait library, and predict the passing behavior of the vehicle on the road section through a neural network model to generate the prediction result of the passing behavior in the future time period.

2. The method for predicting service area congestion based on ETC data according to claim 1, wherein, The key features include the appearance color, vehicle type, license plate number, and vehicle contour of the vehicle; standardize the key features extracted from the captured images of the ramp entrance and exit bayonet cameras, map them to the corresponding discrete data ranges to form discrete data points, regard the data points as mass points, and let them move under the gravitational force towards the set initial center position to form a circular structure. Determine the radius of the circle by calculating the average distance and divide the circle according to the distribution of the data points. Compare the data point distribution patterns in the corresponding circular areas of the captured data at the entrance and exit, and generate a first ID code, including: Standardize each key feature to obtain the standardized key features, specifically including: converting the appearance color into a fixed color classification code, converting the vehicle type into the corresponding vehicle type number, converting the license plate number into a digital code, and converting the vehicle contour into a numerical feature through shape feature extraction; Based on the standardized key features, determine the discrete data range corresponding to each key feature, map the specific value of each key feature to the corresponding discrete data range to form the corresponding discrete data points, so that each piece of captured data at the entrance and exit can be represented as a set of discrete data points; Regard the discrete data points as mass points with gravity, set an initial center position, and let each discrete data point move towards the center position under the gravitational force to finally form a circular structure; Determine the radius of the circle by calculating the average distance from all data points to the center of the circle, and divide the circle into the corresponding number of parts according to the distribution of data points of different key features; For the data point distribution within the circular area corresponding to each piece of entrance capture data, compare it with the data point distribution within the circular area corresponding to all exit capture data. If the data point distribution patterns within the divided areas of the circle are the same, that is, the number and position characteristics of the data points in the corresponding areas are similar, then determine that the entrance and exit capture data are for the same vehicle, and generate a corresponding first ID code.

3. The method for predicting service area congestion based on ETC data according to claim 2, wherein Based on the first ID code, associate the capture data of the front and rear gantries of the service area, generate a second ID code, screen vehicle information through comparison, and calculate the interval passing speed according to the gantry passing time difference and distance to obtain the predicted vehicles entering the service area, including: Associate and match the capture data of the front and rear gantries of the service area, and generate a second ID code based on the capture time, position, and characteristic information of the vehicle at the front and rear gantries of the service area to identify the vehicles passing through the front and rear gantries of the service area; Compare the first ID code with the second ID code to obtain the screened target vehicle information; Based on the screened target vehicle information, calculate the passing time difference and gantry distance between the front and rear gantries of the vehicle in the service area; calculate the interval passing speed of the vehicle according to the passing time difference and gantry distance; According to the interval passing speed of the vehicle and the preset threshold, predict the list of vehicles entering the service area.

4. The method for predicting service area congestion based on ETC data according to claim 3, wherein Capture and collect the real-time data of vehicles entering the service area through the bayonet camera, and dynamically judge the current real-time congestion index of the service area according to the predicted vehicles entering the service area, including: According to the vehicle data captured by the bayonet camera, record the entrance capture data and exit capture data; Calculate the number of vehicles entering the service area and the number of vehicles leaving the service area per unit time; Calculate the number of vehicles in the service area, and judge the real-time congestion index of the service area according to the number of vehicles in the area and the carrying capacity of the service area.

5. The method for predicting service area congestion based on ETC data according to claim 4, wherein, According to the remaining journey of the vehicle to the service area entrance, calculate the estimated arrival time, and according to the real-time congestion index, calculate the flow contribution value for different types of vehicles at the same section of the service area entrance, including: According to the journey of the vehicle from the node to the service area entrance, calculate the time when the vehicle arrives at the service area entrance; At the same section of the service area entrance, obtain all different types of vehicles; Obtain the equivalent number of each type of vehicle in terms of flow relative to the standard vehicle and the time interval of the section; Combine the equivalent number of each type of vehicle in terms of flow relative to the standard vehicle with the time interval of the section to obtain the contribution of each type of vehicle to the traffic flow relative to the standard vehicle per unit time, where the contribution represents the vehicle density; According to the historical actual data, adjust the contribution of each type of vehicle to the traffic flow relative to the standard vehicle to obtain the flow contribution value; Add up the flow contribution values of each type of vehicle to obtain the total flow of all vehicles arriving at the same section of the service area entrance at the same moment.

6. The method for predicting service area congestion based on ETC data according to claim 5, wherein Based on the historical passing behavior data of vehicles, establish a vehicle portrait library, and predict the passing behavior of vehicles on the road section through a neural network model to generate the predicted results of passing behavior in the future period, including: Integrate the basic information and passing records of vehicles into the static database and dynamic database; Construct portrait tags based on the historical traffic behavior of the vehicle, and dynamically update the vehicle portrait library when new traffic data is received; Extract key features from the vehicle portrait and determine the influence weights of different features on traffic behavior; When the vehicle enters the highway again, predict the service areas it will pass through based on historical data and output the prediction results of traffic behavior.

7. The method for predicting service area congestion based on ETC data according to claim 6, wherein Extract key features from the vehicle portrait and determine the influence weights of different features on traffic behavior, including: Extract key features from the vehicle portrait, including vehicle type, vehicle size, vehicle weight, driving speed, driving mileage, driver behavior pattern, time period, weather condition, and historical traffic records; For the key features, perform quantization processing to obtain numerical features; Construct a random forest based on the extracted and quantized features. For each tree in the random forest, traverse each node in the tree. For each node, determine the number of samples contained in the node and calculate the Gini impurity of the node; Split the features, divide the samples in the node into a left child node and a right child node, and calculate the Gini impurity of the left child node and the right child node respectively; Calculate the reduction in Gini impurity based on the Gini impurity of the node and the Gini impurities of the left child node and the right child node; For each feature, count the total reduction in Gini impurity brought about by splitting in all trees to obtain the importance score of the feature; Average the importance scores of each feature to obtain the final importance score of the feature in the random forest; Determine the influence weights of different features on traffic behavior based on the final importance score of the feature in the random forest.

8. An information release and induction system for service area congestion based on ETC data, which implements the method described in any one of claims 1 to 7, characterized in that, Including: The bayonet information module is used to standardize the key features extracted from the captured images of the ramp entrance and exit bayonet cameras, map them to the corresponding discrete data ranges to form discrete data points, regard the data points as mass points, let them move under the action of gravity towards the set initial center position to form a circular structure, determine the radius of the circle by calculating the average distance, and divide the circle according to the distribution of the data points, compare the data point distribution patterns in the corresponding circular areas of the entrance and exit captured data, and generate the first ID code; Based on the first ID code, associate the captured data of the front and rear gantries of the service area, generate the second ID code, screen the vehicle information by comparison, and calculate the interval passing speed according to the time difference and spacing of the gantry passes to obtain the vehicles predicted to enter the service area; The multi-source recognition module is used to capture and collect the real-time data of the vehicle entering the service area through the bayonet camera, and dynamically judge the current real-time congestion index of the service area according to the vehicles predicted to enter the service area; calculate the estimated arrival time according to the remaining journey of the vehicle to the service area entrance, and calculate the traffic contribution value for different types of vehicles in the same section of the service area entrance according to the real-time congestion index; The vehicle trajectory prediction module is used to establish a vehicle portrait library based on the vehicle historical traffic behavior data, and predict the traffic behavior of the vehicle on the road section through a neural network model to generate the prediction results of traffic behavior in the future time period.

9. A computing device, characterized in that, Including: One or more processors; A storage device for storing one or more programs, which when executed by the one or more processors cause the one or more processors to implement the method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, A program is stored in the computer-readable storage medium, and when the program is executed by a processor, the method according to any one of claims 1 to 7 is implemented.

Images