A vehicle entering a service area detection method and terminal based on ETC data

Abstract

A kind of vehicle entry service area detection method based on ETC data, according to ETC transaction data, the driving trajectory of vehicle in a service area located road section is obtained;Vehicle data entering the service area is obtained, and according to the driving trajectory, whether the vehicle enters the service area is taken as a feature to build a first data set;The driving speed of vehicle in the section before service area, service area section and the section after service area is calculated respectively, and the driving speed of each section of vehicle is taken as a feature to build a second data set;According to the first data set and the second data set, a classification model is established and trained, and after training is completed, whether the vehicle enters the service area is judged using the classification model.The present application judges whether the vehicle enters the service area according to ETC transaction data, greatly reduces the cost of service area informatization construction, and provides a constructive data source for digital management and optimization of highway service area.

Description

Technical Field

[0001] This invention relates to the field of intelligent transportation, and in particular to a method and terminal for detecting vehicles entering service areas based on ETC data. Background Technology

[0002] To date, my country's expressway mileage has reached 177,000 kilometers, ranking first in the world, and the number of vehicles has been increasing year by year. To further improve the operational efficiency of my country's expressways, most provinces have achieved network connectivity of ETC gantry systems, while correspondingly eliminating provincial toll stations on expressways, building ETC gantry systems, and upgrading ETC lanes accordingly. Today, the number of ETC users is repeatedly reaching new highs.

[0003] The massive amounts of data accumulated by the ETC system have laid the data foundation for the construction of smart highways. Highway service areas are crucial facilities for ensuring the safe operation of vehicles on highways. Drivers are required to rest in service areas after long periods of driving on highways, according to regulations. Therefore, the informatization of highway service areas is essential to better serve highway travel.

[0004] Currently, traffic flow data in highway service areas is generally collected using checkpoint systems at the service area entrances and exits. These systems capture images through cameras and identify license plates. While this method meets the management and analysis needs of highway service areas to some extent, it suffers from high costs and errors in image capture, preventing its widespread adoption in service areas. Summary of the Invention

[0005] The technical problem to be solved by the present invention is to provide a method and terminal for detecting vehicles entering service areas based on ETC data. The method uses vehicle driving data in ETC to establish a feature model and train it, and uses the trained model to determine whether a vehicle has entered a service area.

[0006] To solve the above-mentioned technical problems, the technical solution adopted by the present invention is as follows:

[0007] A method for detecting vehicles entering service areas based on ETC data, comprising the following steps:

[0008] S1. Obtain the vehicle's travel trajectory on a road segment within a certain service area based on ETC transaction data;

[0009] S2. Obtain vehicle data that has entered the service area, and construct a first dataset based on the driving trajectory, using whether the vehicle has entered the service area as a feature;

[0010] S3. Calculate the vehicle speed in the section before the service area, the service area section, and the service area section respectively, and construct the second dataset with the vehicle speed in each section as the feature.

[0011] S4. Establish and train a classification model based on the first and second datasets. After training, use the classification model to determine whether a vehicle has entered the service area.

[0012] To solve the above-mentioned technical problems, another technical solution adopted by the present invention is as follows:

[0013] A vehicle entry into a service area detection terminal based on ETC data includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the aforementioned vehicle entry into a service area detection method based on ETC data.

[0014] The beneficial effects of this invention are as follows: It provides a vehicle entry service area detection method and terminal based on ETC data. By utilizing vehicle driving data in ETC, a dataset of vehicle driving speeds on various road segments is formed. A classification model is established based on this dataset. After training, the model is used to determine whether a vehicle has entered the service area. This allows for accurate judgment of whether a vehicle has entered the service area based on ETC transaction data, greatly reducing the cost of information construction in service areas and providing a constructive data source for the digital management and optimization of highway service areas. Attached Figure Description

[0015] Figure 1 This is a flowchart of a vehicle entry into a service area detection method based on ETC data according to a certain embodiment of the present invention;

[0016] Figure 2 This is a schematic diagram of the SMOTE algorithm interpolation according to a certain embodiment of the present invention:

[0017] Figure 3 This is a schematic diagram of a vehicle entry into a service area detection terminal based on ETC data according to a certain embodiment of the present invention;

[0018] Label Explanation:

[0019] 1. A vehicle entry detection terminal based on ETC data; 2. A processor;

[0020] 3. Storage. Detailed Implementation

[0021] To explain in detail the technical content, objectives, and effects of the present invention, the following description is provided in conjunction with the embodiments and accompanying drawings.

[0022] Please refer to Figure 1 as well as Figure 2 A method for detecting vehicles entering service areas based on ETC data, comprising the following steps:

[0023] S1. Obtain the vehicle's travel trajectory on a road segment within a certain service area based on ETC transaction data;

[0024] S2. Obtain vehicle data that has entered the service area, and construct a first dataset based on the driving trajectory, using whether the vehicle has entered the service area as a feature;

[0025] S3. Calculate the vehicle speed in the section before the service area, the service area section, and the service area section respectively, and construct the second dataset with the vehicle speed in each section as the feature.

[0026] S4. Establish and train a classification model based on the first and second datasets. After training, use the classification model to determine whether a vehicle has entered the service area.

[0027] The working principle of this invention is as follows: a vehicle entry service area detection method based on ETC data. By utilizing vehicle driving data in ETC, a dataset of vehicle speeds on various road segments is formed. A classification model is established based on this dataset. After training, the model is used to determine whether a vehicle has entered the service area. This allows for accurate judgment of whether a vehicle has entered the service area based on ETC transaction data, greatly reducing the cost of information technology construction in service areas and providing a constructive data source for the digital management and optimization of highway service areas.

[0028] Furthermore, step S40 is included between steps S3 and S4:

[0029] S40. Use the SMOTE algorithm to balance the first and second datasets.

[0030] As described above, there are generally two solutions to address the problem of imbalanced data: oversampling and undersampling. Oversampling expands the sample size of the minority class by copying sample data. However, this method leads to the repetition of identical sample data, which may cause overfitting during model training. Conversely, undersampling randomly deletes a portion of the data from the majority class samples, or selects a certain proportion of samples as the sample data for that class. This method may cause the model to only learn a part of the pattern of the sample data, failing to effectively reflect the complete pattern of the sample data for that class. To preserve the overall pattern of the sample data of vehicles that did not enter the service area, avoid the overfitting problem caused by directly copying minority class samples, and eliminate the impact of imbalanced data on the model, the SMOTE (Synthetic Minority Over-sampling Technique) algorithm is adopted. The SMOTE algorithm utilizes the similarity of minority class samples in the feature space, analyzes the minority class samples and generates new samples, adding simulated minority samples to the dataset to expand the number of minority class samples in the original dataset. This helps reduce the intra-class scatter between different categories of whether a vehicle enters a service area in the vehicle trajectory sample data, thereby restoring the balance between the categories in the dataset.

[0031] Further, step S40 includes:

[0032] S401. Based on the first dataset and the second dataset, establish a data set D(x) i y i ), x i The features of sample i, y i It is the label of sample i;

[0033] S402. Select the samples of vehicles labeled as having entered the service area, calculate the Euclidean spatial distance between all the samples, and select the k nearest neighbor samples for each sample; the formula for calculating the Euclidean spatial distance is as follows:

[0034]

[0035] Where, x i and x j These represent the i-th and j-th samples of vehicles whose labels are used to enter the road segment where the service area is located, respectively; N is the number of features; k is a positive integer; and N is a positive integer.

[0036] S403. Randomly select s sample points belonging to the data set from the k nearest neighbor sample points. Perform random linear interpolation on the line connecting this sample point and the selected s nearest neighbor sample points to generate a new feature vector sample, where s is a positive integer less than or equal to k.

[0037] S404. Repeat steps S402 and S403 until the number of samples labeled as vehicles entering the service area reaches a preset balance with the number of samples of other categories in the data set.

[0038] S405, Integrate the samples by replacing the first dataset and the second dataset.

[0039] As described above, in this step, a data set is first established based on the existing dataset, and samples are provided simultaneously. In one embodiment of the present invention, the vehicle's driving speed in the service area section is used as a sample feature, that is, x, which corresponds to the feature of sample i in step S401. i Vehicles entering the service area can be selected as the sample labels, corresponding to the label y of sample i in steps S401 and S402. i Then, to prevent the number of samples labeled as vehicles entering the service area from being too small, the SMOTE algorithm is used for balance adjustment. First, the Euclidean spatial distance between the samples is calculated, and k nearest neighbor sample points are selected for each sample. Then, s sample points belonging to the dataset are randomly selected from the k nearest neighbor sample points. Random linear interpolation is performed on the line connecting this sample point and the selected s nearest neighbor sample points to generate new feature vector samples. This process is repeated until the capacity of the dataset of such a small number of samples is balanced among the categories.

[0040] Further, step S4 specifically includes:

[0041] S41. Divide the first dataset and the second dataset into a training set and a test set according to a preset ratio;

[0042] S42. Use the classification model to train the model. After training, use the classification model to determine whether the vehicle has entered the service area.

[0043] As described above, the dataset is divided into a training set and a test set according to a preset ratio. Specifically, the preset ratio can be selected as training set: test set = 4:1 or 7:3 to meet the requirements of model training and validation. After the model training is completed, it is possible to determine whether a vehicle has entered the service area based on the ETC vehicle data.

[0044] Furthermore, the classification model is the XGBoost model.

[0045] As described above, the classification model specifically adopts the XGBoost model, which has the advantages of being highly efficient and scalable, fast and effective when processing large-scale datasets, and has low requirements for hardware resources such as memory.

[0046] Further, step S2 specifically includes:

[0047] S21. Based on the ETC gantry data, extract vehicle information that has entered the road section where the service area is located;

[0048] S22. Obtain the time point when the vehicle passes the previous gantry of the service area and the time point when the vehicle enters the service area, and calculate the difference between the time point when the vehicle passes the previous gantry of the service area and the time point when the vehicle enters the service area.

[0049] S23. If the difference is less than the first preset time, it is determined that the vehicle has entered the service area; otherwise, it is determined that the vehicle has not entered the service area.

[0050] S24. Construct a first dataset characterized by whether a vehicle enters the service area.

[0051] As described above, the determination of whether a vehicle has entered a service area can be made by combining the ETC gantry system and the service area's entry records. First, ETC data and service area data are used to determine whether a vehicle has entered the service area. Vehicles with overlapping data are identified, and these vehicles are included in the service area's general data. Then, to prevent vehicles from entering the service area multiple times and causing duplicate data collection (as there may be instances where a vehicle merely passes through a service area without actually stopping), a preset time interval is used to compare the difference between the time a vehicle enters the service area and the time it passes the previous gantry. If the difference is less than the preset time interval, the vehicle is considered to have entered the service area; otherwise, the vehicle is considered to have either passed through the service area or not entered, and this is uniformly recorded in the dataset as not entering the service area.

[0052] Further, step S3 specifically includes:

[0053] S31. Based on the ETC gantry data, obtain the gantry position of the road segment where the service area is located, and divide the service area road segment into the front service area segment, the service area segment, and the rear service area segment according to the gantry position.

[0054] S32. Obtain the time taken for the vehicle to pass through the front section, the service area section, and the rear section of the service area, and calculate the vehicle's speed in the front section, the service area section, and the rear section of the service area respectively;

[0055] S33. Construct a second dataset featuring the driving speed of vehicles in different sections.

[0056] As described above, the driving speed of a vehicle in each section of the service area is calculated based on the ETC gantry system. The gantry position of the road segment where the service area is located is obtained, and the road segment of the service area is divided into the front section, the service area section, and the rear section of the service area according to the gantry position. The average driving speed of the vehicle in each section is calculated based on the time it takes for the vehicle to pass through the gantry recorded in the gantry system, and thus recorded as a feature in the dataset.

[0057] Please refer to Figure 3 A vehicle entry into a service area detection terminal 1 based on ETC data includes a memory 3, a processor 2, and a computer program stored in the memory 3 and executable on the processor 2. When the processor executes the computer program, it implements the aforementioned vehicle entry into a service area detection method based on ETC data.

[0058] As described above, a terminal is provided as the main body for the operation and management of the above method, so as to realize the controllable operation of the method.

[0059] This invention provides a vehicle entry service area detection method and terminal based on ETC data, mainly used for intelligent judgment of whether a vehicle has entered a service area. The following is a description with reference to specific embodiments:

[0060] Embodiment 1 of the present invention is as follows: Please refer to Figures 1 to 2 A method for detecting vehicles entering service areas based on ETC data, comprising the following steps:

[0061] S1. Obtain the vehicle's travel trajectory on a road segment within a certain service area based on ETC transaction data;

[0062] S2. Obtain vehicle data entering the service area, and construct the first dataset based on the driving trajectory, using whether the vehicle entered the service area as a feature;

[0063] S3. Calculate the vehicle speed in the section before the service area, the service area section, and the service area section respectively, and construct the second dataset with the vehicle speed in each section as the feature.

[0064] S4. Build and train a classification model based on the first and second datasets. After training, use the classification model to determine whether a vehicle has entered a service area.

[0065] In this embodiment, a vehicle entry into a service area detection method based on ETC data is provided. By utilizing vehicle driving data in ETC, a dataset of vehicle speeds on various road segments is formed. A classification model is built based on this dataset. After training, the model is used to determine whether a vehicle has entered the service area. This allows for accurate determination of whether a vehicle has entered the service area based on ETC transaction data, greatly reducing the cost of information technology construction for service areas and providing a constructive data source for the digital management and optimization of highway service areas.

[0066] Embodiment 2 of the present invention is as follows: Based on Embodiment 1, please refer to... Figures 1 to 2 Step S2 is as follows:

[0067] S21. Based on the ETC gantry data, extract vehicle information for the road section where the service area is located;

[0068] S22. Obtain the time point when the vehicle passes the previous gantry of the service area and the time point when the vehicle enters the service area, and calculate the difference between the time point when the vehicle passes the previous gantry of the service area and the time point when the vehicle enters the service area.

[0069] S23. If the difference is less than the first preset time, it is determined that the vehicle has entered the service area; otherwise, it is determined that the vehicle has not entered the service area.

[0070] S24. Construct the first dataset based on whether a vehicle enters a service area.

[0071] Step S3 is as follows:

[0072] S31. Based on the ETC gantry data, obtain the gantry position of the road segment where the service area is located, and divide the service area road segment into the front service area segment, the service area segment, and the rear service area segment according to the gantry position.

[0073] S32. Obtain the time taken for the vehicle to pass through the pre-service area section, the service area section, and the service area section, and calculate the vehicle's speed in the pre-service area section, the service area section, and the service area section respectively.

[0074] S33. Construct a second dataset featuring the driving speed of vehicles in different sections.

[0075] In this embodiment, the determination of whether a vehicle has entered a service area can be made by combining the ETC gantry system and the service area's entry records. First, the ETC data and service area data are used to determine whether a vehicle has entered the service area. Vehicles with overlapping data are identified, as they exist in the general data for this service area. Then, to prevent duplicate data collection due to vehicles entering the service area multiple times, a preset time interval is used to compare the difference between the time a vehicle enters the service area and the time it passes the previous gantry. If the difference is less than the preset time interval, the vehicle is considered to have entered the service area; otherwise, it is considered either a vehicle that passed through the service area or did not enter, and this is recorded in the dataset as "not entering the service area." In addition, the vehicle's speed in each section of the service area is calculated using the ETC gantry system. The gantry position of the service area is obtained, and the service area section is divided into a pre-service area section, a service area section, and a post-service area section based on the gantry position. The average speed of the vehicle in each section is calculated based on the time recorded in the gantry system for passing through the gantry, and this average speed is recorded as a feature in the dataset.

[0076] Embodiment 3 of the present invention is as follows: Based on Embodiment 2, please refer to... Figures 1 to 2 The step between steps S3 and S4 includes step S40, which specifically involves:

[0077] S401. Based on the first dataset and the second dataset, establish a data union D(x) i y i ), x i The features of sample i, y i It is the label of sample i;

[0078] S402. Select samples of vehicles labeled as having entered the service area, calculate the Euclidean spatial distance between all samples, and select the k nearest neighbor samples for each sample; the formula for calculating the Euclidean spatial distance is as follows:

[0079]

[0080] Where, x i and x j Let i and j represent the i-th and j-th samples of vehicles labeled as entering the service area, respectively; N is the number of features; k is a positive integer; and N is a positive integer.

[0081] S403. Randomly select s sample points from the k nearest neighbor sample points that belong to the dataset. Perform random linear interpolation on the line connecting this sample point and the selected s nearest neighbor sample points to generate a new feature vector sample; s is a positive integer less than or equal to k.

[0082] S404. Repeat steps S402 and S403 until the number of samples labeled as vehicles entering the service area reaches a preset balance with the number of samples of other categories in the data set.

[0083] S405, Replace the first and second datasets with integrated samples.

[0084] As described above, to address the problem of imbalanced data samples, the SMOTE algorithm is used for data balancing. Its principle lies in leveraging the similarity of minority class samples in the feature space. By analyzing minority class samples and generating new samples, these simulated minority samples are added to the dataset, thereby expanding the number of minority class samples in the original dataset. This helps reduce the intra-class scatter between different categories of whether a vehicle trajectory sample has entered a service area, thus restoring the balance between the categories in the dataset.

[0085] The specific execution principle is as follows: First, a data set is established based on the existing dataset, and samples are provided. Specifically, the driving speed of vehicles on service area road sections is used as the sample feature, that is, the feature x corresponding to sample i in step S401. i The vehicles that enter the service area are selected as the labels for the samples, that is, the labels y of sample i in steps S401 and S402. i Then, the SMOTE algorithm is used for balance adjustment. First, the Euclidean spatial distance between samples is calculated, and k nearest neighbor sample points are selected for each sample. Then, s sample points belonging to the dataset are randomly selected from the k nearest neighbor sample points. Random linear interpolation is performed on the line connecting this sample point and the selected s nearest neighbor sample points to generate new feature vector samples. This process is repeated until the capacity of the dataset of such minority samples is balanced among the categories.

[0086] Embodiment four of the present invention is as follows: Based on embodiment three, please refer to... Figures 1 to 2 Step S41 is as follows:

[0087] S41. Divide the first dataset and the second dataset into a training set and a test set according to a preset ratio;

[0088] S42. Use the XGBoost model to train the model. After training, use the classification model to determine whether a vehicle has entered a service area.

[0089] As described above, the dataset is divided into a training set and a test set according to a preset ratio. Specifically, the preset ratio can be selected as training set: test set = 4:1 or 7:3 to meet the requirements of model training and validation. After the model training is completed, it is possible to determine whether a vehicle has entered the service area based on the ETC vehicle data.

[0090] Specifically, the XGBoost model operates as follows:

[0091] Suppose we have a dataset D(x) i y i ), where x i The feature of sample i, x i It can be represented as a vector (v1, v2, v3, t).

[0092] Where v1 represents the speed of the vehicle as it passes through the section before the service area;

[0093] v2 represents the speed of the vehicle as it passes through the service area section;

[0094] v3 indicates the speed of the vehicle in the section after passing through the service area;

[0095] t represents the time period during which a vehicle passes through the ETC gantry of the service section. The time period is divided into hours, and there are 24 time periods in a day. The value of t ranges from 0 to 23.

[0096] y i These are the labels for the samples; 0 indicates that the vehicle has not entered the service area, and 1 indicates that the vehicle has entered the service area.

[0097] Step 1: Read the dataset and divide it into training and validation sets according to the ratio. Initialize the model parameters, including tree depth, learning rate, regularization parameters, etc.

[0098] Step 2: Initialize the model: Input the training set data into XGBoost and initialize a weak classifier, i.e., a single decision tree. Define the objective function (loss function) as follows:

[0099]

[0100]

[0101] Where l is a differentiable convex loss function, representing the predicted value. and target value y i The difference between them is taken by the squared loss function in this invention. K represents the k-th decision tree. The second term ∑Ω(f) represents the regularization term, where γ is the penalty coefficient, representing the weight of the decision tree, taking a value between 0 and 1, which penalizes the complexity of the model, T is the number of leaf nodes of the tree, λ is the regularization coefficient, and ω represents the model's prediction value for the leaf nodes.

[0102] Step 3: Iteratively train multiple weak classifiers, each optimized based on the previous classifier to minimize the objective function value. In each iteration, calculate the gradient (first derivative of the loss function) and second derivative of each sample to generate the output of each weak classifier.

[0103] The formulas for calculating the gradient (first derivative) and second derivative of each sample are as follows:

[0104]

[0105]

[0106] Where y i f represents the result of the t-th tree. t-1 (x i ) represents the result of the (t-1)th tree.

[0107] Step 4: Update the model: Repeat steps 2 and 3 until the stopping condition is met, i.e., the number of iterations is completed or the loss value of the objective function is lower than the target value.

[0108] Step 5: Obtain the model for classification. The XGBoost model trained through the preceding steps is the model for determining whether a vehicle has entered a service area. The vehicle's speed in the service area segment and the segments before and after it, as well as the time taken to pass through the service area segment, are input into the model. Based on the rate of change of the vehicle's speed in each segment, the result of whether the vehicle has entered the service area is obtained.

[0109] Embodiment five of the present invention is as follows: Please refer to Figure 3 A vehicle entry into a service area detection terminal 1 based on ETC data includes a memory 3, a processor 2, and a computer program stored in the memory 3 and executable on the processor 2. When the processor executes the computer program, it implements a vehicle entry into a service area detection method based on ETC data according to any of the above embodiments.

[0110] In summary, this invention provides a vehicle entry service area detection method and terminal based on ETC data. By utilizing vehicle driving data in ETC, a dataset of vehicle speeds on various road segments is formed. A classification model is established based on this dataset. After training, the model is used to determine whether a vehicle has entered the service area. This allows for accurate judgment of whether a vehicle has entered the service area based on ETC transaction data, significantly reducing the cost of service area information construction and providing a constructive data source for the digital management and optimization of highway service areas.

[0111] The above description is merely an embodiment of the present invention and does not limit the patent scope of the present invention. Any equivalent modifications made based on the content of the present invention specification and drawings, or direct or indirect applications in related technical fields, are similarly included within the patent protection scope of the present invention.

Claims

1. A method for detecting vehicles entering service areas based on ETC data, characterized in that: Including the following steps: S1. Obtain the vehicle's travel trajectory on a road segment within a certain service area based on ETC transaction data; S2. Obtain vehicle data that has entered the service area, and construct a first dataset based on the driving trajectory, using whether the vehicle has entered the service area as a feature; S3. Calculate the vehicle speed in the section before the service area, the service area section, and the service area section respectively, and construct the second dataset with the vehicle speed in each section as the feature. S4. Build and train a classification model based on the first and second datasets. After training, use the classification model to determine whether a vehicle has entered the service area. Step S40 is also included between steps S3 and S4: S40. Use the SMOTE algorithm to balance the first and second datasets; Step S40 includes the following steps: S401. Establish a data set based on the first dataset and the second dataset. , These are the features of sample i. It is the label of sample i; S402. Select the samples of vehicles whose labels are used to enter the road segment where the service area is located, calculate the Euclidean spatial distance between all the samples, and select each sample. The nearest neighbor sample points; the Euclidean spatial distance is calculated using the following formula: in, and These represent the i-th and j-th samples of vehicles whose labels are used to enter the road segment where the service area is located, respectively, and N is the number of features; N is a positive integer; S403, in Randomly select from the nearest neighbor sample points A number of sample points belonging to the data set, and in this sample point and the selected Random linear interpolation is performed on the lines connecting the nearest neighbor sample points to generate new feature vector samples. less than or equal to Positive integers; S404. Repeat steps S402 and S403 until the number of samples labeled as vehicles entering the service area reaches a preset balance with the number of samples of other categories in the data set. S405, Integrate the samples by replacing the first dataset and the second dataset; Step S3 specifically involves: S31. Based on the ETC gantry data, obtain the gantry position of the road segment where the service area is located, and divide the service area road segment into the front service area segment, the service area segment, and the rear service area segment according to the gantry position. S32. Obtain the time taken for the vehicle to pass through the front section, the service area section, and the rear section of the service area, and calculate the vehicle's speed in the front section, the service area section, and the rear section of the service area respectively. S33. Construct a second dataset featuring the driving speed of vehicles in different sections.

2. The method for detecting vehicles entering service areas based on ETC data according to claim 1, characterized in that: Step S4 includes: S41. Divide the first dataset and the second dataset into a training set and a test set according to a preset ratio; S42. Use the classification model to train the model. After training, use the classification model to determine whether the vehicle has entered the service area.

3. The method for detecting vehicles entering service areas based on ETC data according to claim 2, characterized in that: The classification model is the XGBoost model.

4. The method for detecting vehicles entering service areas based on ETC data according to claim 1, characterized in that: Step S2 specifically involves: S21. Based on the ETC gantry data, extract vehicle information that has entered the road section where the service area is located; S22. Obtain the time point when the vehicle passes the previous gantry of the service area and the time point when the vehicle enters the service area, and calculate the difference between the time point when the vehicle passes the previous gantry of the service area and the time point when the vehicle enters the service area. S23. If the difference is less than the first preset time, it is determined that the vehicle has entered the service area; otherwise, it is determined that the vehicle has not entered the service area. S24. Construct a first dataset characterized by whether a vehicle enters the service area.

5. A vehicle entry service area detection terminal based on ETC data, characterized in that: The device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the vehicle entry into service area detection method based on ETC data as described in any one of claims 1-4.

Images