Illegal vehicle identification method, device, storage medium and electronic equipment

Abstract

The present application discloses a method, device, storage medium and electronic device for identifying illegally operated vehicles, the method comprising: obtaining multiple historical traffic data of vehicles to be identified that enter and exit the expressway within a preset time period, each piece of historical traffic data comprising expressway checkpoint data and GIS data corresponding to the checkpoint location to which the expressway checkpoint data belongs; constructing the operation characteristics of the vehicle to be identified based on the multiple historical traffic data of the vehicle to be identified, the operation characteristics comprising the frequency of entering and exiting the preset site, the vehicle operation intensity characteristics, the operation discreteness characteristics, the checkpoint coverage and the regional label to which the vehicle belongs; inputting the operation characteristics of the vehicle to be identified into a pre-trained illegal operation vehicle identification model, and outputting the probability that the vehicle to be identified is an online-hailing vehicle; when the probability is greater than a preset threshold, pushing the vehicle information of the vehicle to be identified to the supervision platform to verify the legality of the vehicle to be identified. Therefore, by adopting the embodiments of the present application, illegally operated vehicles can be accurately identified.

Description

Technical Field

[0001] The present application relates to the field of computer technology, and in particular to a method, device, storage medium and electronic device for identifying illegally operated vehicles. Background Art

[0002] In modern urban traffic management, accurate identification and supervision of online-hailing vehicles is the key to maintaining traffic order and urban safety. A typical application scenario is that urban traffic management departments need to identify online-hailing vehicles that frequently enter and exit hot spots such as preset venues in the city in order to conduct compliance inspections and law enforcement.

[0003] In related technologies, the trajectory of passenger cars is tracked and the collision of mobile phone signaling data and highway checkpoint information is used to identify suspected online ride-hailing vehicles. This method relies on the coverage and strength of mobile phone signals. The collision of mobile phone signaling data and checkpoint information often results in insufficient data accuracy due to signal interference, coverage limitations and other factors, making it impossible to accurately identify illegally operated vehicles. Summary of the invention

[0004] The embodiments of the present application provide a method, device, storage medium and electronic device for identifying illegally operated vehicles. In order to have a basic understanding of some aspects of the disclosed embodiments, a simple summary is given below. This summary is not a general review, nor is it intended to identify key / important components or describe the scope of protection of these embodiments. Its only purpose is to present some concepts in a simple form as a preface to the detailed description that follows.

[0005] In a first aspect, an embodiment of the present application provides a method for identifying illegally operated vehicles, the method comprising:

[0006] Acquire multiple historical traffic data of vehicles to be identified entering and exiting the expressway within a preset time period, each historical traffic data includes expressway checkpoint data and GIS data corresponding to the checkpoint location to which the expressway checkpoint data belongs;

[0007] Based on multiple historical traffic data of the vehicle to be identified, the operation characteristics of the vehicle to be identified are constructed. The operation characteristics include the frequency of entering and exiting the preset site, the vehicle operation intensity characteristics, the operation dispersion characteristics, the checkpoint coverage and the regional label of the vehicle;

[0008] Input the operating characteristics of the vehicle to be identified into the pre-trained illegal vehicle identification model, and output the probability that the vehicle to be identified is an online-hailing vehicle;

[0009] When the probability is greater than a preset threshold, the vehicle information of the vehicle to be identified will be pushed to the supervision platform to verify the legality of the vehicle to be identified.

[0010] Optionally, follow the steps below to generate a pre-trained illegal vehicle identification model, including:

[0011] Collect multiple historical traffic data of each target vehicle in a vehicle set within a preset period, where the vehicle set includes multiple online-hailing vehicles and multiple private cars;

[0012] According to multiple historical traffic data of each target vehicle, the operation characteristics of each target vehicle are constructed, and the operation characteristics of each target vehicle are the operation characteristics of an online car-hailing vehicle or the operation characteristics of a private car;

[0013] The operating characteristics of all online ride-hailing vehicles are labeled as positive samples, and the operating characteristics of all private cars are labeled as negative samples to obtain model training samples;

[0014] A random forest network was used to create an illegal vehicle identification model;

[0015] According to the model training samples, the illegal operation vehicle identification model is machine-learned to obtain a pre-trained illegal operation vehicle identification model.

[0016] Optionally, based on multiple pieces of historical traffic data of the vehicle to be identified or each target vehicle, the operation characteristics of the vehicle to be identified or each target vehicle are constructed, including:

[0017] According to the GIS data included in each piece of historical traffic data of the first vehicle, the frequency of the vehicle entering and exiting the preset site is counted, and the first vehicle is the vehicle to be identified or each target vehicle;

[0018] Extracting the first highway entry and exit time farthest from the current moment and the second highway entry and exit time closest to the current moment from the highway checkpoint data included in each piece of historical traffic data of the first vehicle;

[0019] Determine a time period according to the first high-speed entry and exit time and the second high-speed entry and exit time;

[0020] Based on the high-speed checkpoint data and time period included in each historical traffic data, the operation intensity characteristics are constructed;

[0021] Based on the high-speed checkpoint data included in each historical traffic data, the operational dispersion characteristics are constructed;

[0022] According to the high-speed checkpoint data included in each historical traffic data, the checkpoint coverage and the regional label of the vehicle are counted respectively;

[0023] The frequency of the vehicle entering and exiting the preset site, the operation intensity characteristics, the operation dispersion characteristics, the checkpoint coverage and the regional label to which the vehicle belongs are used as the operation characteristics of the first vehicle.

[0024] Optionally, based on the high-speed checkpoint data and time period included in each piece of historical traffic data, construct operation intensity characteristics, including:

[0025] According to the high-speed checkpoint data included in each historical traffic data, the total number of all high-speed records within the time period is counted to obtain the total number of operations;

[0026] According to the highway checkpoint data included in each historical traffic data, the sum of the toll mileage of all highway records within the statistical time period is calculated to obtain the total highway mileage;

[0027] According to the high-speed checkpoint data included in each historical traffic data, the sum of the duration of all high-speed records within the time period is calculated to obtain the total high-speed duration;

[0028] According to the high-speed checkpoint data included in each historical traffic data, the sum of all days with high-speed records in the statistical time period is calculated to obtain the total operating days;

[0029] According to the high-speed checkpoint data included in each historical traffic data, the sum of the high-speed records with the same starting and ending points within the time period is counted to determine the number of times the frequently-run route is traveled;

[0030] Determine the number of trips on high-frequency routes based on the number of trips on the frequently-run routes;

[0031] The operation intensity characteristics are calculated based on the total number of operations, total highway mileage, total highway time, total operation days, number of trips on frequently traveled routes, and number of trips on high-frequency operation routes.

[0032] Optionally, the operation intensity characteristics are calculated based on the total number of operations, total highway mileage, total highway time, total number of operating days, number of trips on frequently-run routes, and number of trips on high-frequency routes, including:

[0033] Calculate the ratio of total operating days to the time period to get the high-speed operating rate;

[0034] Calculate the ratio of total highway mileage to total highway time to get the average vehicle speed;

[0035] According to the highway checkpoint data included in each historical traffic data, the sum of the toll mileage in the highway records on weekdays is counted; the ratio of the sum of the toll mileage in the highway records on weekdays to the total highway mileage is calculated to obtain the mileage ratio on weekdays;

[0036] Calculate the ratio of total highway mileage to total operating days to get the average daily highway mileage;

[0037] Calculate the ratio of total operation times to total operation days to get the average daily high-speed times;

[0038] Calculate the ratio of the number of trips on the frequently-used routes to the number of trips on the high-frequency operating routes to obtain the proportion of the number of trips on the high-frequency operating routes;

[0039] According to the highway checkpoint data included in each historical traffic data, the sum of the number of times the checkpoints for entering and exiting the highway are not in the same district or county is counted; the ratio of the sum of the number of times the checkpoints for entering and exiting the highway are not in the same district or county to the total number of operations is calculated to obtain the proportion of inter-county driving times;

[0040] According to the highway checkpoint data included in each historical traffic data, the operation time in the second half of the night and the operation time in the day are counted; the ratio of the operation time in the second half of the night and the operation time in the day to the total highway time is calculated respectively, and the proportion of the time in the second half of the night and the proportion of the time in the day are obtained;

[0041] The highway operation rate, average vehicle speed, proportion of weekday mileage, average daily highway mileage, average daily highway times, proportion of high-frequency route trips, proportion of inter-county trips, proportion of late night time and proportion of daytime time are taken as characteristics of operation intensity.

[0042] Optionally, based on the high-speed checkpoint data included in each piece of historical traffic data, operational dispersion features are constructed, including:

[0043] According to the high-speed checkpoint data included in each historical traffic data, the high-speed operation period is counted;

[0044] Divide the high-speed operation period into multiple slices and count the number of operations on each slice;

[0045] Calculate the ratio of the number of operations on each slice to the total number of operations to obtain the proportion of the number of operations on each slice;

[0046] According to the proportion of operations on each slice, the average and standard deviation of the high-speed operation period are calculated;

[0047] The ratio of the standard deviation to the mean is calculated to obtain the operational dispersion characteristics.

[0048] Optionally, based on the model training samples, machine learning is performed on the illegal operation vehicle identification model to obtain a pre-trained illegal operation vehicle identification model, including:

[0049] Divide the model training samples into training set, test set and validation set by presetting the division ratio;

[0050] Input the training set into the illegal vehicle identification model for machine learning to generate an initial training model;

[0051] Input the test set into the initial model and output the probability that the vehicle in the test set is an online ride-hailing vehicle;

[0052] According to the probability that the vehicles in the test set are online-hailing vehicles, determine the number of vehicles in the test set that are predicted to be real online-hailing vehicles and are in the positive samples of the test set;

[0053] Calculate the ratio between the number of online ride-hailing vehicles in the test set that are predicted to be true and in the positive samples of the test set and the total number of vehicles in the positive samples of the test set to obtain the recall rate;

[0054] Calculate the number of vehicles in the test set that are predicted to be real online-hailing vehicles and in the positive samples of the test set and the number of vehicles in the test set that are predicted to be online-hailing vehicles, and get the accuracy rate;

[0055] The model is tuned according to the recall rate and accuracy, and the tuned model is used to predict the validation set. When the prediction results meet the preset conditions, the tuned model is used as a pre-trained illegal vehicle identification model.

[0056] In a second aspect, an embodiment of the present application provides a device for identifying illegally operated vehicles, the device comprising:

[0057] An acquisition module is used to acquire multiple historical traffic data of vehicles to be identified entering and exiting the expressway within a preset time period, each of which includes expressway checkpoint data and GIS data corresponding to the checkpoint location to which the expressway checkpoint data belongs;

[0058] A construction module is used to construct the operation characteristics of the vehicle to be identified based on multiple historical traffic data of the vehicle to be identified. The operation characteristics include the frequency of entering and exiting the preset site, the vehicle operation intensity characteristics, the operation dispersion characteristics, the checkpoint coverage and the regional label of the vehicle;

[0059] An output module is used to input the operating characteristics of the vehicle to be identified into a pre-trained illegal operation vehicle identification model, and output the probability that the vehicle to be identified is an online-hailing vehicle;

[0060] The push module is used to push the vehicle information of the vehicle to be identified to the supervision platform when the probability is greater than a preset threshold, so as to verify the legality of the vehicle to be identified.

[0061] In a third aspect, an embodiment of the present application provides a computer storage medium, which stores a plurality of instructions, and the instructions are suitable for being loaded by a processor and executing the above-mentioned method steps.

[0062] In a fourth aspect, an embodiment of the present application provides a device, which may include: a processor and a memory; wherein the memory stores a computer program, and the computer program is suitable for being loaded by the processor and executing the above-mentioned method steps.

[0063] The technical solution provided by the embodiments of the present application may have the following beneficial effects:

[0064] In an embodiment of the present application, the operating characteristics of the vehicle are constructed through multiple historical traffic data of the vehicle. The historical traffic data include highway checkpoint data and GIS data corresponding to the checkpoint location to which the highway checkpoint data belongs. This data provides comprehensive operating information of the vehicle, so that the constructed operating characteristics (frequency of entering and exiting preset venues, vehicle operation intensity characteristics, operation dispersion characteristics, checkpoint coverage, and vehicle regional label) can comprehensively reflect the vehicle's operating mode and behavioral habits. Combined with the pre-trained illegal operation vehicle identification model, it can accurately identify the probability that the vehicle is an online car-hailing vehicle, and then determine the illegal online car-hailing vehicle based on the probability of the online car-hailing vehicle, while improving the coverage and continuity of supervision.

[0065] It should be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0066] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0067] Figure 1 It is a flowchart of a method for identifying illegally operated vehicles provided in an embodiment of the present application;

[0068] Figure 2 This is an overall flow chart of the identification of illegally operated vehicles provided by this application;

[0069] Figure 3 It is a flowchart of a model training method for an illegal operation vehicle identification model provided by this application;

[0070] Figure 4 It is a structural schematic diagram of an illegal vehicle identification device provided in an embodiment of the present application;

[0071] Figure 5 It is a structural schematic diagram of a device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0072] The following description and the accompanying drawings sufficiently illustrate specific embodiments of the present application to enable those skilled in the art to practice them.

[0073] It should be clear that the described embodiments are only part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in the field without creative work are within the scope of protection of the present application.

[0074] When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present application. Instead, they are only examples of devices and methods consistent with some aspects of the present application as detailed in the attached claims.

[0075] In the description of the present application, it should be understood that the terms "first", "second", etc. are used for descriptive purposes only and should not be understood as indicating or implying relative importance. For those of ordinary skill in the art, the specific meanings of the above terms in the present application can be understood according to the specific circumstances. In addition, in the description of the present application, unless otherwise specified, "multiple" refers to two or more. "And / or" describes the association relationship of associated objects, indicating that three relationships may exist. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. The character " / " generally indicates that the objects associated before and after are in an "or" relationship.

[0076] The present application provides a method, device, storage medium and electronic device for identifying illegally operated vehicles to solve the problems existing in the above-mentioned related technical problems. In an embodiment of the present application, the operation characteristics of the vehicle are constructed through multiple historical traffic data of the vehicle. The historical traffic data includes high-speed checkpoint data and GIS data corresponding to the checkpoint location to which the high-speed checkpoint data belongs. The data provides comprehensive operation information of the vehicle, so that the constructed operation characteristics (frequency of entering and exiting the preset venue, vehicle operation intensity characteristics, operation dispersion characteristics, checkpoint coverage and vehicle regional label) can fully reflect the vehicle's operation mode and behavior habits. Combined with the pre-trained illegal operation vehicle identification model, it can accurately identify the probability that the vehicle is an online car-hailing vehicle, and then determine the illegal online car-hailing vehicle based on the probability of the online car-hailing vehicle, while improving the coverage and continuity of supervision. The following uses an exemplary embodiment for detailed description.

[0077] The following will be combined with the attached Figure 1 -Attached Figure 3 , the illegal vehicle identification method provided in the embodiment of the present application is introduced in detail. The method can be implemented by a computer program and can be run on an illegal vehicle identification device based on the von Neumann system. The computer program can be integrated into an application or run as an independent tool application.

[0078] See also Figure 1 , which is a flow chart of a method for identifying an illegally operated vehicle according to an embodiment of the present application. Figure 1 As shown, the method of the embodiment of the present application may include the following steps:

[0079] S101, obtaining a plurality of historical traffic data of vehicles to be identified entering and exiting the expressway within a preset time period, each piece of historical traffic data including expressway checkpoint data and GIS data corresponding to the checkpoint location to which the expressway checkpoint data belongs;

[0080] Among them, the vehicle to be identified is the vehicle currently entering or exiting the highway, and it is necessary to identify whether the vehicle is an illegally operated vehicle. The preset time period refers to a time range set for a specific analysis purpose. Highway checkpoint data refers to vehicle traffic data collected at toll stations or specific monitoring points (checkpoints) on highways, including license plate numbers, entry and exit times, toll mileage, checkpoint ID, and the administrative region to which the checkpoint belongs. GIS data (geographic information system data) refers to data related to geographic location, including whether the checkpoint is connected to a preset site tag.

[0081] The data structure of each piece of historical traffic data is the license plate number, time of entering the highway, time of exiting the highway, toll mileage, checkpoint ID, administrative region to which the checkpoint belongs, and whether the checkpoint is connected to a preset site label.

[0082] In some embodiments of the present application, for example, a vehicle leaves a highway exit on December 5, 2024, and the traffic management department needs to determine whether the vehicle is suspected of illegal operation. The traffic management department sets a preset time period, such as the past 6 months (June 1, 2024 to December 1, 2024), for collecting and analyzing the historical traffic data of the vehicle. The system automatically retrieves all traffic records of the vehicle within the preset time period from the highway checkpoint database.

[0083] S102, constructing the operation characteristics of the vehicle to be identified based on multiple historical traffic data of the vehicle to be identified, the operation characteristics including the frequency of entering and exiting the preset site, the vehicle operation intensity characteristics, the operation dispersion characteristics, the checkpoint coverage and the regional label of the vehicle;

[0084] The operation characteristics refer to the feature set extracted from historical traffic data that can describe the vehicle operation mode. Preset venues include airports and high-speed railway stations.

[0085] In some embodiments of the present application, the specific process of constructing the operation characteristics of the vehicle to be identified based on multiple historical passage data of the vehicle to be identified includes: according to the GIS data included in each historical passage data of the first vehicle, counting the frequency of the vehicle entering and exiting the preset venue, the first vehicle is the vehicle to be identified; from the highway checkpoint data included in each historical passage data of the first vehicle, extracting the first highway entry and exit time farthest from the current moment and the second highway entry and exit time closest; determining the time period based on the first highway entry and exit time and the second highway entry and exit time; constructing the operation intensity characteristics based on the highway checkpoint data and the time period included in each historical passage data; constructing the operation dispersion characteristics based on the highway checkpoint data included in each historical passage data; according to the highway checkpoint data included in each historical passage data, respectively counting the checkpoint coverage and the regional label to which the vehicle belongs; the frequency of the vehicle entering and exiting the preset venue, the operation intensity characteristics, the operation dispersion characteristics, the checkpoint coverage and the regional label to which the vehicle belongs are used as the operation characteristics of the first vehicle.

[0086] For example, the time period TD may be the sum of all days from the month in which the first high-speed record occurs to the month in which the last high-speed record occurs.

[0087] Among them, GIS data includes whether the checkpoint is connected to a preset venue tag (such as an airport tag); whether the checkpoint is connected to a preset venue tag, a passenger car can pass through this checkpoint and drive to a nearby preset venue (such as an airport), and there is a high probability that it is heading to the airport.

[0088] In some embodiments of the present application, the specific process of counting the frequency of the vehicle entering and exiting the preset venue according to the GIS data included in each piece of historical passage data of the first vehicle includes: determining the number of times n1 of entering and exiting the airport per day according to the GIS data included in each piece of historical passage data of the first vehicle, counting the number of days d1 when the number of times of entering and exiting the airport reaches the set threshold n2, and calculating the proportion to the total operating days: T1 = d1 / D. D is the total operating days.

[0089] In some embodiments of the present application, according to the highway checkpoint data and time period included in each historical passage data, the specific process of constructing the operation intensity characteristics includes: according to the highway checkpoint data included in each historical passage data, the total number of all highway records in the time period is counted to obtain the total number of operations; according to the highway checkpoint data included in each historical passage data, the sum of the toll mileage of all highway records in the time period is counted to obtain the total highway mileage; according to the highway checkpoint data included in each historical passage data, the sum of the duration of all highway records in the time period is counted to obtain the total highway duration; according to the highway checkpoint data included in each historical passage data, the sum of all days with highway records in the time period is counted to obtain the total operation days; according to the highway checkpoint data included in each historical passage data, the sum of the number of highway records with the same starting and ending points in the time period is counted to determine the number of trips on frequently run routes; according to the determined number of trips on frequently run routes, the number of trips on high-frequency operating routes is determined; according to the total number of operations, the total highway mileage, the total highway duration, the total number of operation days, the number of trips on frequently run routes, and the number of trips on high-frequency operating routes, the operation intensity characteristics are calculated.

[0090] Specifically, since each piece of historical traffic data includes the license plate number, time of entering the highway, time of exiting the highway, toll mileage, and checkpoint ID, each piece of historical traffic data can be directly analyzed to count the total number of operations C, total highway mileage M, total highway duration T, total number of operating days D, number of trips on frequently-run routes L, and number of trips on high-frequency operating routes 3L. Among them, the number of trips on frequently-run routes L is the sum of the number of highway records with the same starting and ending points within the statistical time period, with the district or county where the checkpoint is located as the starting point or end point of a route. After determining the number of trips on frequently-run routes for each long-distance route, the routes can be sorted from high to low according to the number of trips on frequently-run routes, and the sum of the number of trips corresponding to the first three frequently-run routes is taken as the number of trips on high-frequency operating routes.

[0091] In some embodiments of the present application, the specific process of calculating the operation intensity characteristics according to the total number of operations, total highway mileage, total highway time, total operating days, number of trips on frequently traveled routes, and number of trips on high-frequency operating routes includes: calculating the ratio of the total number of operating days to the time period to obtain the highway operation rate; calculating the ratio of the total highway mileage to the total highway time to obtain the average vehicle speed; according to the highway checkpoint data included in each historical traffic data, counting the sum of the toll mileage in the highway records on weekdays; calculating the ratio of the sum of the toll mileage in the highway records on weekdays to the total highway mileage to obtain the proportion of mileage on weekdays; calculating the ratio of the total highway mileage to the total number of operating days to obtain the average daily highway mileage; calculating the ratio of the total number of operations to the total number of operating days to obtain the average daily number of highway trips; calculating the ratio of the number of trips on frequently traveled routes to the number of trips on high-frequency operating routes the ratio of the number of trips to the total number of trips per day to obtain the proportion of trips on high-frequency operating routes; according to the highway checkpoint data included in each historical traffic data, the sum of the number of times the checkpoints for entering and exiting the highway are not in the same district or county is counted; the ratio of the sum of the number of times the checkpoints for entering and exiting the highway are not in the same district or county to the total number of operations is calculated to obtain the proportion of inter-county trips; according to the highway checkpoint data included in each historical traffic data, the operating hours in the second half of the night and the operating hours in the day are counted; the ratios of the operating hours in the second half of the night and the operating hours in the day to the total highway time are calculated respectively to obtain the proportion of the second half of the night time and the proportion of the daytime time; the highway operation rate, average vehicle speed, proportion of mileage on weekdays, average daily highway mileage, average daily number of highways, proportion of trips on high-frequency operating routes, proportion of inter-county trips, proportion of the second half of the night time and proportion of the daytime time are taken as operation intensity characteristics.

[0092] For example, after calculating the total number of operations C, total highway mileage M, total highway time T, total operating days D, number of trips on frequently run routes L, and number of trips on high-frequency operating routes 3L, the highway operation rate = D / TD; average vehicle speed = M / T; proportion of weekday mileage = sum of toll mileage in weekday highway records / M; average daily highway mileage = M / D; average daily number of highway trips = C / D; proportion of trips on high-frequency operating routes = L / 3L; proportion of inter-county trips = sum of the number of times the checkpoints for entering and exiting the highway are not in the same district or county / C; proportion of time in the second half of the night = time of operation in the second half of the night / T; proportion of time during the day = time of operation during the day / T.

[0093] In some embodiments of the present application, the specific process of constructing an operation discreteness feature based on the high-speed checkpoint data included in each historical traffic data includes: counting the high-speed operation period based on the high-speed checkpoint data included in each historical traffic data; dividing the high-speed operation period into multiple slices, and counting the number of operations on each slice; calculating the ratio of the number of operations on each slice to the total number of operations to obtain the proportion of the number of operations on each slice; calculating the average and standard deviation of the high-speed operation period based on the proportion of the number of operations on each slice; calculating the ratio of the standard deviation to the average value to obtain the operation discreteness feature.

[0094] For example, divide the high-speed operation period into n slices, count the number of operations on the i-th slice, and calculate the proportion of operations on each slice x i , operation times account for x i = Number of operations on the i-th slice / total number of operations C, where the total number of operations C is obtained by counting the total number of all high-speed records within the time period based on the high-speed checkpoint data included in each historical traffic data.

[0095] Therefore, the average Standard Deviation Operational dispersion V = S / x.

[0096] In some embodiments of the present application, according to the high-speed checkpoint data included in each piece of historical traffic data, the specific process of counting the checkpoint coverage and the vehicle's regional label includes: determining the checkpoint ID corresponding to the checkpoint where the vehicle enters and exits the highway every day, counting the number of IDs n3 that the vehicle passes through every day based on the counted checkpoint ID, recording the number of days d2 when the number of IDs n3 reaches the set threshold n4, and calculating the proportion to the total number of operating days: T2 = d2 / D. Based on the vehicle's license plate number, determine whether the vehicle's regional label is a Beijing, Shanghai, or Guangzhou vehicle label, and set it to 1 if it is, and set it to 0 if it is not.

[0097] S103, inputting the operating characteristics of the vehicle to be identified into a pre-trained illegal operation vehicle identification model, and outputting the probability that the vehicle to be identified is an online-hailing vehicle;

[0098] In some embodiments of the present application, after obtaining the operating characteristics of the vehicle to be identified, the operating characteristics of the vehicle to be identified can be input into a pre-trained illegal vehicle identification model, and after model processing, the probability that the vehicle to be identified is an online-hailing vehicle is output.

[0099] In some embodiments of the present application, the specific process of generating a pre-trained illegal vehicle identification model includes: collecting multiple historical traffic data of each target vehicle in a vehicle set within a preset period, the vehicle set including multiple online-hailing vehicles and multiple private cars; constructing the operating characteristics of each target vehicle based on the multiple historical traffic data of each target vehicle, the operating characteristics of each target vehicle being the operating characteristics of an online-hailing vehicle or the operating characteristics of a private car; marking the operating characteristics of all online-hailing vehicles as positive samples, and marking the operating characteristics of all private cars as negative samples, to obtain model training samples; using a random forest network to create an illegal vehicle identification model; performing machine learning on the illegal vehicle identification model based on the model training samples to obtain a pre-trained illegal vehicle identification model.

[0100] Specifically, according to multiple historical passage data of each target vehicle, the specific process of constructing the operation characteristics of the vehicle to be identified or each target vehicle includes: according to the GIS data included in each historical passage data of the first vehicle, counting the frequency of the vehicle entering and exiting the preset site, the first vehicle is each target vehicle; from the highway checkpoint data included in each historical passage data of the first vehicle, extracting the first highway entry and exit time farthest from the current moment and the second highway entry and exit time closest; determining the time period according to the first highway entry and exit time and the second highway entry and exit time; constructing the operation intensity characteristics according to the highway checkpoint data and the time period included in each historical passage data; constructing the operation dispersion characteristics according to the highway checkpoint data included in each historical passage data; according to the highway checkpoint data included in each historical passage data, respectively counting the checkpoint coverage and the regional label to which the vehicle belongs; taking the frequency of the vehicle entering and exiting the preset site, the operation intensity characteristics, the operation dispersion characteristics, the checkpoint coverage and the regional label to which the vehicle belongs as the operation characteristics of the first vehicle.

[0101] Specifically, based on the model training samples, the illegal operation vehicle identification model is subjected to machine learning, and the specific process of obtaining the pre-trained illegal operation vehicle identification model includes: dividing the model training samples into a training set, a test set, and a validation set according to a preset division ratio; inputting the training set into the illegal operation vehicle identification model for machine learning to generate an initial model for training; inputting the test set into the initial model, and outputting the probability that the vehicle in the test set is an online-hailing vehicle; based on the probability that the vehicle in the test set is an online-hailing vehicle, determining the number of online-hailing vehicles in the test set that are predicted to be true and in the positive samples of the test set; calculating the ratio between the number of online-hailing vehicles in the test set that are predicted to be true and in the positive samples of the test set and the total number of vehicles in the positive samples of the test set to obtain the recall rate; calculating the number of online-hailing vehicles in the test set that are predicted to be true and in the positive samples of the test set and the number of vehicles predicted to be online-hailing vehicles in the test set to obtain the accuracy rate; performing model tuning based on the recall rate and the accuracy rate, and using the tuned model to predict the validation set, and when the prediction result meets the preset conditions, using the tuned model as the pre-trained illegal operation vehicle identification model.

[0102] The preset division ratio is: training set (positive z1: negative z2), test set (positive z3: negative z4), validation set (positive z5: negative z6), the positive and negative sample ratio of the training set and the test set is 1:1, and the positive and negative sample ratio of the validation set should be consistent with the distribution in the market. There are two main parameters for tuning the random forest model: the number of trees and the depth.

[0103] S104: When the probability is greater than a preset threshold, the vehicle information of the vehicle to be identified is pushed to the supervision platform to verify the legality of the vehicle to be identified.

[0104] In some embodiments of the present application, these operating characteristics are input into the trained machine learning model, and the model calculates the probability that the vehicle is an illegally operated vehicle, assuming that the probability output by the model is 85%. A preset threshold can be set, such as 75%. This means that only when the probability predicted by the model exceeds 75%, the vehicle will be marked as high risk and require further verification. Since the probability predicted by the model (85%) is greater than the preset threshold (75%), the system can automatically push the vehicle information of the vehicle (such as license plate number, traffic record, predicted probability, etc.) to the supervision platform. After the supervision platform receives the pushed vehicle information, the staff of the traffic management department can view this information and decide whether it is necessary to conduct further legality verification on the vehicle. Specifically, the "Online Taxi Transport Certificate" and the "Online Taxi Driver Certificate" can be compared to confirm whether there is illegal operation.

[0105] For example Figure 2 As shown, Figure 2 This is an overall flow chart for the identification of illegally operated vehicles provided by this application, including model training, model prediction, and result application. Model training includes data collection and feature analysis. Data collection specifically collects high-speed checkpoint data for the past 12 months, uses online ride-hailing vehicles as positive samples, and private cars as negative samples. Feature analysis specifically analyzes operation intensity, operation dispersion, airport frequency, checkpoint coverage, and operation area. Model prediction can use the trained random forest model for prediction, and push the predicted abnormal results to the supervision platform.

[0106] In an embodiment of the present application, the operation characteristics of the vehicle are constructed through multiple historical traffic data of the vehicle. The historical traffic data includes highway checkpoint data and GIS data corresponding to the checkpoint location to which the highway checkpoint data belongs. This data provides comprehensive operation information of the vehicle, so that the constructed operation characteristics (frequency of entering and exiting preset venues, vehicle operation intensity characteristics, operation dispersion characteristics, checkpoint coverage, and vehicle regional label) can comprehensively reflect the vehicle's operation mode and behavior habits. Combined with the pre-trained illegal operation vehicle identification model, it can accurately identify the probability that the vehicle is an online car-hailing vehicle, and then determine the illegal online car-hailing vehicle based on the probability of the online car-hailing vehicle, while improving the coverage and continuity of supervision.

[0107] See also Figure 3 , which is a flow chart of a model training method for an illegal vehicle operation identification model provided in the embodiment of the present application. Figure 3 As shown, the method of the embodiment of the present application may include the following steps:

[0108] S201, collecting multiple historical traffic data of each target vehicle in a vehicle set within a preset period, where the vehicle set includes multiple online-hailing vehicles and multiple private cars;

[0109] S202, constructing an operation characteristic of each target vehicle according to multiple historical traffic data of each target vehicle, where the operation characteristic of each target vehicle is an operation characteristic of an online car-hailing vehicle or an operation characteristic of a private car;

[0110] S203, marking the operating characteristics of all online-hailing vehicles as positive samples, and marking the operating characteristics of all private cars as negative samples, to obtain model training samples;

[0111] S204, uses random forest network to create an illegal operation vehicle identification model;

[0112] S205, performing machine learning on the illegal operation vehicle identification model according to the model training samples to obtain a pre-trained illegal operation vehicle identification model.

[0113] In an embodiment of the present application, the vehicle's operating characteristics are constructed through multiple historical traffic data of the vehicle. The historical traffic data includes highway checkpoint data and GIS data corresponding to the checkpoint location to which the highway checkpoint data belongs. This data provides comprehensive vehicle operation information, so that the constructed operation characteristics (frequency of entering and exiting preset venues, vehicle operation intensity characteristics, operation dispersion characteristics, checkpoint coverage, and vehicle regional label) can comprehensively reflect the vehicle's operation mode and behavior habits. Combined with the pre-trained illegal operation vehicle identification model, it can accurately identify the probability that the vehicle is an online car-hailing vehicle, and then determine the illegal online car-hailing vehicle based on the probability of the online car-hailing vehicle, while improving the coverage and continuity of supervision.

[0114] The following is an embodiment of the device of the present application, which can be used to execute the embodiment of the method of the present application. For details not disclosed in the embodiment of the device of the present application, please refer to the embodiment of the method of the present application.

[0115] See also Figure 4 , which shows a schematic diagram of the structure of an illegal vehicle identification device provided by an exemplary embodiment of the present application. The illegal vehicle identification device can be implemented as all or part of a device through software, hardware, or a combination of both. The device 1 includes an acquisition module 10, a construction module 20, an output module 30, and a push module 40.

[0116] The acquisition module 10 is used to acquire a plurality of historical traffic data of the to-be-identified vehicles entering and exiting the expressway within a preset time period, each of which includes expressway checkpoint data and GIS data corresponding to the checkpoint location to which the expressway checkpoint data belongs;

[0117] A construction module 20 is used to construct the operation characteristics of the vehicle to be identified based on multiple historical traffic data of the vehicle to be identified, and the operation characteristics include the frequency of entering and exiting the preset site, the vehicle operation intensity characteristics, the operation dispersion characteristics, the checkpoint coverage and the regional label of the vehicle;

[0118] The output module 30 is used to input the operating characteristics of the vehicle to be identified into the pre-trained illegal operation vehicle identification model, and output the probability that the vehicle to be identified is an online-hailing vehicle;

[0119] The push module 40 is used to push the vehicle information of the vehicle to be identified to the supervision platform when the probability is greater than a preset threshold, so as to verify the legality of the vehicle to be identified.

[0120] It should be noted that the illegal vehicle identification device provided in the above embodiment only uses the division of the above functional modules as an example when executing the illegal vehicle identification method. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above. In addition, the illegal vehicle identification device provided in the above embodiment and the illegal vehicle identification method embodiment belong to the same concept, and the implementation process thereof is detailed in the method embodiment, which will not be repeated here.

[0121] The serial numbers of the above-mentioned embodiments of the present application are for description only and do not represent the advantages or disadvantages of the embodiments.

[0122] In an embodiment of the present application, the vehicle's operating characteristics are constructed through multiple historical traffic data of the vehicle. The historical traffic data includes highway checkpoint data and GIS data corresponding to the checkpoint location to which the highway checkpoint data belongs. This data provides comprehensive vehicle operation information, so that the constructed operation characteristics (frequency of entering and exiting preset venues, vehicle operation intensity characteristics, operation dispersion characteristics, checkpoint coverage, and vehicle regional label) can comprehensively reflect the vehicle's operation mode and behavior habits. Combined with the pre-trained illegal operation vehicle identification model, it can accurately identify the probability that the vehicle is an online car-hailing vehicle, and then determine the illegal online car-hailing vehicle based on the probability of the online car-hailing vehicle, while improving the coverage and continuity of supervision.

[0123] The present application also provides a computer-readable medium having program instructions stored thereon, which, when executed by a processor, implement the illegal vehicle identification method provided by the above-mentioned various method embodiments.

[0124] The present application also provides a computer program product comprising instructions, which, when executed on a computer, enables the computer to execute the illegal vehicle identification method of each of the above method embodiments.

[0125] See also Figure 5 , which is a schematic diagram of the structure of a device provided in the embodiment of the present application. Figure 5 As shown, the device 1000 may include: at least one processor 1001 , at least one network interface 1004 , a user interface 1003 , a memory 1005 , and at least one communication bus 1002 .

[0126] The communication bus 1002 is used to realize the connection and communication between these components.

[0127] The user interface 1003 may include a display screen (Display) and a camera (Camera), and the optional user interface 1003 may also include a standard wired interface and a wireless interface.

[0128] The network interface 1004 may optionally include a standard wired interface or a wireless interface (such as a WI-FI interface).

[0129] Among them, the processor 1001 may include one or more processing cores. The processor 1001 uses various interfaces and lines to connect the various parts of the entire device 1000, and executes various functions and processes data of the device 1000 by running or executing instructions, programs, code sets or instruction sets stored in the memory 1005, and calling data stored in the memory 1005. Optionally, the processor 1001 can be implemented in at least one hardware form of digital signal processing (Digital Signal Processing, DSP), field programmable gate array (Field-Programmable Gate Array, FPGA), and programmable logic array (Programmable Logic Array, PLA). The processor 1001 can integrate one or a combination of a central processing unit (Central Processing Unit, CPU), a graphics processing unit (Graphics Processing Unit, GPU) and a modem. Among them, the CPU mainly processes operating devices, user interfaces and applications, etc.; the GPU is responsible for rendering and drawing the content to be displayed on the display screen; the modem is used to process wireless communications. It can be understood that the above-mentioned modem may not be integrated into the processor 1001, and it can be implemented separately through a chip.

[0130] Among them, the memory 1005 may include a random access memory (Random Access Memory, RAM) and may also include a read-only memory (Read-Only Memory). Optionally, the memory 1005 includes a non-transitory computer-readable storage medium. The memory 1005 can be used to store instructions, programs, codes, code sets or instruction sets. The memory 1005 may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing the operating device, instructions for at least one function (such as a touch function, a sound playback function, an image playback function, etc.), instructions for implementing the above-mentioned various method embodiments, etc.; the data storage area may store data involved in the above-mentioned various method embodiments, etc. The memory 1005 may also be optionally at least one storage device located away from the aforementioned processor 1001. As Figure 5 As shown, the memory 1005 as a computer storage medium may include an operating device, a network communication module, a user interface module, and an illegal vehicle operation identification application.

[0131] exist Figure 5 In the device 1000 shown, the user interface 1003 is mainly used to provide an input interface for the user and obtain the data input by the user; and the processor 1001 can be used to call the illegal vehicle identification application stored in the memory 1005, and specifically perform the following operations:

[0132] Obtain multiple historical traffic data of vehicles to be identified entering and exiting the expressway within a preset time period, each historical traffic data includes expressway checkpoint data and GIS data corresponding to the checkpoint location to which the expressway checkpoint data belongs;

[0133] Based on multiple historical traffic data of the vehicle to be identified, the operation characteristics of the vehicle to be identified are constructed. The operation characteristics include the frequency of entering and exiting the preset site, the vehicle operation intensity characteristics, the operation dispersion characteristics, the checkpoint coverage and the regional label of the vehicle;

[0134] Input the operating characteristics of the vehicle to be identified into the pre-trained illegal vehicle identification model, and output the probability that the vehicle to be identified is an online-hailing vehicle;

[0135] When the probability is greater than a preset threshold, the vehicle information of the vehicle to be identified will be pushed to the supervision platform to verify the legality of the vehicle to be identified.

[0136] In one embodiment, when the processor 1001 generates a pre-trained illegal vehicle identification model, the processor 1001 specifically performs the following operations:

[0137] Collect multiple historical traffic data of each target vehicle in a vehicle set within a preset period, where the vehicle set includes multiple online-hailing vehicles and multiple private cars;

[0138] According to multiple historical traffic data of each target vehicle, the operation characteristics of each target vehicle are constructed, and the operation characteristics of each target vehicle are the operation characteristics of an online car-hailing vehicle or the operation characteristics of a private car;

[0139] The operating characteristics of all online ride-hailing vehicles are labeled as positive samples, and the operating characteristics of all private cars are labeled as negative samples to obtain model training samples;

[0140] A random forest network was used to create an illegal vehicle identification model;

[0141] According to the model training samples, the illegal operation vehicle identification model is machine-learned to obtain a pre-trained illegal operation vehicle identification model.

[0142] In one embodiment, when the processor 1001 constructs the operation characteristics of the vehicle to be identified or each target vehicle according to multiple historical traffic data of the vehicle to be identified or each target vehicle, the processor 1001 specifically performs the following operations:

[0143] According to the GIS data included in each piece of historical traffic data of the first vehicle, the frequency of the vehicle entering and exiting the preset site is counted, and the first vehicle is the vehicle to be identified or each target vehicle;

[0144] Extracting the first highway entry and exit time farthest from the current moment and the second highway entry and exit time closest to the current moment from the highway checkpoint data included in each piece of historical traffic data of the first vehicle;

[0145] Determine a time period according to the first high-speed entry and exit time and the second high-speed entry and exit time;

[0146] Based on the high-speed checkpoint data and time period included in each historical traffic data, the operation intensity characteristics are constructed;

[0147] Based on the high-speed checkpoint data included in each historical traffic data, the operational dispersion characteristics are constructed;

[0148] According to the high-speed checkpoint data included in each historical traffic data, the checkpoint coverage and the regional label of the vehicle are counted respectively;

[0149] The frequency of the vehicle entering and exiting the preset site, the operation intensity characteristics, the operation dispersion characteristics, the checkpoint coverage and the regional label to which the vehicle belongs are used as the operation characteristics of the first vehicle.

[0150] In one embodiment, when the processor 1001 constructs the operation intensity feature according to the high-speed checkpoint data and time period included in each piece of historical traffic data, the processor 1001 specifically performs the following operations:

[0151] According to the high-speed checkpoint data included in each historical traffic data, the total number of all high-speed records within the time period is counted to obtain the total number of operations;

[0152] According to the highway checkpoint data included in each historical traffic data, the sum of the toll mileage of all highway records within the statistical time period is calculated to obtain the total highway mileage;

[0153] According to the high-speed checkpoint data included in each historical traffic data, the sum of the duration of all high-speed records within the time period is calculated to obtain the total high-speed duration;

[0154] According to the high-speed checkpoint data included in each historical traffic data, the sum of all days with high-speed records in the statistical time period is calculated to obtain the total operating days;

[0155] According to the high-speed checkpoint data included in each historical traffic data, the sum of the high-speed records with the same starting and ending points within the time period is counted to determine the number of times the frequently-run route is traveled;

[0156] Determine the number of trips on high-frequency routes based on the number of trips on the frequently-run routes;

[0157] The operation intensity characteristics are calculated based on the total number of operations, total highway mileage, total highway time, total operation days, number of trips on frequently traveled routes, and number of trips on high-frequency operation routes.

[0158] In one embodiment, when the processor 1001 calculates the operation intensity characteristics according to the total number of operations, the total mileage of highways, the total duration of highways, the total number of operation days, the number of trips on frequently-run routes, and the number of trips on high-frequency operation routes, the processor 1001 specifically performs the following operations:

[0159] Calculate the ratio of total operating days to the time period to get the high-speed operating rate;

[0160] Calculate the ratio of total highway mileage to total highway time to get the average vehicle speed;

[0161] According to the highway checkpoint data included in each historical traffic data, the sum of the toll mileage in the highway records on weekdays is counted; the ratio of the sum of the toll mileage in the highway records on weekdays to the total highway mileage is calculated to obtain the mileage ratio on weekdays;

[0162] Calculate the ratio of total highway mileage to total operating days to get the average daily highway mileage;

[0163] Calculate the ratio of total operation times to total operation days to get the average daily high-speed times;

[0164] Calculate the ratio of the number of trips on the frequently-used routes to the number of trips on the high-frequency operating routes to obtain the proportion of the number of trips on the high-frequency operating routes;

[0165] According to the highway checkpoint data included in each historical traffic data, the sum of the number of times the checkpoints for entering and exiting the highway are not in the same district or county is counted; the ratio of the sum of the number of times the checkpoints for entering and exiting the highway are not in the same district or county to the total number of operations is calculated to obtain the proportion of inter-county driving times;

[0166] According to the highway checkpoint data included in each historical traffic data, the operation time in the second half of the night and the operation time in the day are counted; the ratio of the operation time in the second half of the night and the operation time in the day to the total highway time is calculated respectively, and the proportion of the operation time in the second half of the night and the proportion of the operation time in the day are obtained;

[0167] The highway operation rate, average vehicle speed, proportion of weekday mileage, average daily highway mileage, average daily highway times, proportion of high-frequency route trips, proportion of inter-county trips, proportion of late night time and proportion of daytime time are taken as characteristics of operation intensity.

[0168] In one embodiment, when the processor 1001 constructs the operational dispersion feature according to the high-speed checkpoint data included in each piece of historical traffic data, the processor 1001 specifically performs the following operations:

[0169] According to the high-speed checkpoint data included in each historical traffic data, the high-speed operation period is counted;

[0170] Divide the high-speed operation period into multiple slices and count the number of operations on each slice;

[0171] Calculate the ratio of the number of operations on each slice to the total number of operations to obtain the proportion of the number of operations on each slice;

[0172] According to the proportion of operations on each slice, the average and standard deviation of the high-speed operation period are calculated;

[0173] The ratio of the standard deviation to the mean is calculated to obtain the operational dispersion characteristics.

[0174] In one embodiment, when the processor 1001 performs machine learning on the illegal operation vehicle identification model according to the model training sample to obtain the pre-trained illegal operation vehicle identification model, the following operations are specifically performed:

[0175] Divide the model training samples into training set, test set and validation set by presetting the division ratio;

[0176] Input the training set into the illegal vehicle identification model for machine learning to generate an initial training model;

[0177] Input the test set into the initial model and output the probability that the vehicle in the test set is an online ride-hailing vehicle;

[0178] According to the probability that the vehicles in the test set are online-hailing vehicles, determine the number of vehicles in the test set that are predicted to be real online-hailing vehicles and are in the positive samples of the test set;

[0179] Calculate the ratio between the number of online ride-hailing vehicles in the test set that are predicted to be true and in the positive samples of the test set and the total number of vehicles in the positive samples of the test set to obtain the recall rate;

[0180] Calculate the number of vehicles in the test set that are predicted to be real online-hailing vehicles and in the positive samples of the test set and the number of vehicles in the test set that are predicted to be online-hailing vehicles, and get the accuracy rate;

[0181] The model is tuned according to the recall rate and accuracy, and the tuned model is used to predict the validation set. When the prediction results meet the preset conditions, the tuned model is used as a pre-trained illegal vehicle identification model.

[0182] In an embodiment of the present application, the vehicle's operating characteristics are constructed through multiple historical traffic data of the vehicle. The historical traffic data includes highway checkpoint data and GIS data corresponding to the checkpoint location to which the highway checkpoint data belongs. This data provides comprehensive vehicle operation information, so that the constructed operation characteristics (frequency of entering and exiting preset venues, vehicle operation intensity characteristics, operation dispersion characteristics, checkpoint coverage, and vehicle regional label) can comprehensively reflect the vehicle's operation mode and behavior habits. Combined with the pre-trained illegal operation vehicle identification model, it can accurately identify the probability that the vehicle is an online car-hailing vehicle, and then determine the illegal online car-hailing vehicle based on the probability of the online car-hailing vehicle, while improving the coverage and continuity of supervision.

[0183] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiments can be implemented by instructing related hardware through a computer program, and the program for identifying illegal vehicles can be stored in a computer-readable storage medium, and when the program is executed, it can include the processes of the embodiments of the above-mentioned methods. The storage medium of the program for identifying illegal vehicles can be a disk, an optical disk, a read-only storage memory, or a random access memory, etc.

[0184] The above disclosure is only the preferred embodiment of the present application, which certainly cannot be used to limit the scope of rights of the present application. Therefore, equivalent changes made according to the claims of the present application are still within the scope covered by the present application.

Claims

1. A method for identifying illegally operated vehicles, characterized in that: The method comprises: Acquire multiple historical traffic data of vehicles to be identified entering and exiting the expressway within a preset time period, each historical traffic data including expressway checkpoint data and GIS data corresponding to the checkpoint location to which the expressway checkpoint data belongs; According to the multiple historical traffic data of the vehicle to be identified, the operation characteristics of the vehicle to be identified are constructed, and the operation characteristics include the frequency of entering and exiting the preset venue, the vehicle operation intensity characteristics, the operation dispersion characteristics, the checkpoint coverage and the regional label of the vehicle; Inputting the operating characteristics of the vehicle to be identified into a pre-trained illegal operation vehicle identification model, and outputting the probability that the vehicle to be identified is an online-hailing vehicle; When the probability is greater than a preset threshold, the vehicle information of the vehicle to be identified is pushed to the supervision platform to verify the legality of the vehicle to be identified.

2. The method according to claim 1, characterized in that Follow these steps to generate a pre-trained illegal vehicle identification model, including: Collect multiple historical traffic data of each target vehicle in a vehicle set within a preset period, where the vehicle set includes multiple online-hailing vehicles and multiple private cars; Constructing the operation characteristics of each target vehicle according to the multiple historical traffic data of each target vehicle, wherein the operation characteristics of each target vehicle are the operation characteristics of the online car-hailing vehicle or the operation characteristics of the private car; The operating characteristics of all online ride-hailing vehicles are labeled as positive samples, and the operating characteristics of all private cars are labeled as negative samples to obtain model training samples; A random forest network was used to create an illegal vehicle identification model; The illegal operation vehicle identification model is machine-learned according to the model training samples to obtain a pre-trained illegal operation vehicle identification model.

3. The method according to claim 1 or 2, characterized in that: According to the plurality of historical traffic data of the vehicle to be identified or each target vehicle, constructing the operation characteristics of the vehicle to be identified or each target vehicle, including: According to the GIS data included in each piece of historical traffic data of the first vehicle, counting the frequency of the vehicle entering and exiting the preset site, the first vehicle being the vehicle to be identified or each of the target vehicles; Extracting the first highway entry and exit time farthest from the current moment and the second highway entry and exit time closest to the current moment from the highway checkpoint data included in each piece of historical traffic data of the first vehicle; Determining a time period according to the first high-speed entry and exit time and the second high-speed entry and exit time; Constructing an operation intensity feature according to the high-speed checkpoint data and the time period included in each piece of historical traffic data; Constructing an operational dispersion feature according to the high-speed checkpoint data included in each piece of historical traffic data; According to the high-speed checkpoint data included in each piece of historical traffic data, the checkpoint coverage and the regional label to which the vehicle belongs are respectively counted; The frequency of the vehicle entering and exiting the preset site, the operation intensity characteristics, the operation dispersion characteristics, the checkpoint coverage and the regional label to which the vehicle belongs are used as the operation characteristics of the first vehicle.

4. The method according to claim 3, characterized in that The constructing of the operation intensity characteristics according to the high-speed checkpoint data and the time period included in each piece of historical traffic data includes: According to the high-speed checkpoint data included in each piece of historical traffic data, the total number of all high-speed records in the time period is counted to obtain the total number of operations; According to the highway checkpoint data included in each piece of historical traffic data, the sum of the toll mileage recorded on all highways within the time period is counted to obtain the total highway mileage; According to the high-speed checkpoint data included in each piece of historical traffic data, the sum of the durations of all high-speed records within the time period is counted to obtain the total high-speed duration; According to the high-speed checkpoint data included in each piece of historical traffic data, the sum of all days with high-speed records in the time period is counted to obtain the total number of operating days; According to the highway checkpoint data included in each piece of historical traffic data, the sum of the number of highway records with the same starting and ending points within the time period is counted to determine the number of times the frequently-run route is traveled; Determine the number of trips on high-frequency routes based on the number of trips on the frequently-run routes; The operation intensity characteristics are calculated based on the total number of operations, total highway mileage, total highway time, total operation days, number of trips on frequently traveled routes, and number of operations on high-frequency routes.

5. The method according to claim 4, characterized in that The operation intensity characteristics are calculated based on the total number of operations, total highway mileage, total highway time, total number of operating days, number of trips on frequently-run routes, and number of trips on high-frequency operating routes, including: Calculating the ratio of the total number of operating days to the time period to obtain a high-speed operating rate; Calculate the ratio of the total highway mileage to the total highway time to obtain an average vehicle speed; According to the highway checkpoint data included in each piece of historical traffic data, the sum of the toll mileage in the highway record on weekdays is counted; the ratio of the sum of the toll mileage in the highway record on weekdays to the total highway mileage is calculated to obtain the mileage ratio on weekdays; Calculate the ratio of the total expressway mileage to the total number of operating days to obtain the average daily expressway mileage; Calculate the ratio of the total number of operations to the total number of operating days to obtain the average daily high-speed number of operations; Calculate the ratio of the number of times the frequently-used route is traveled to the number of times the high-frequency operating route is traveled, and obtain the proportion of the number of times the high-frequency operating route is traveled; According to the highway checkpoint data included in each piece of historical traffic data, the sum of the number of times the checkpoints for entering and exiting the highway are not in the same district or county is counted; the ratio of the sum of the number of times the checkpoints for entering and exiting the highway are not in the same district or county to the total number of operations is calculated to obtain the proportion of inter-county driving times; According to the highway checkpoint data included in each piece of historical traffic data, the operation time in the second half of the night and the operation time in the day are counted; the ratio of the operation time in the second half of the night and the operation time in the day to the total highway time is calculated to obtain the proportion of the operation time in the second half of the night and the proportion of the operation time in the day; The highway operation rate, average vehicle speed, proportion of weekday mileage, average daily highway mileage, average daily highway times, proportion of high-frequency route trips, proportion of inter-county trips, proportion of late night time and proportion of daytime time are taken as operation intensity characteristics.

6. The method according to claim 3, characterized in that: The constructing of the operational dispersion feature according to the high-speed checkpoint data included in each piece of historical traffic data includes: According to the highway checkpoint data included in each piece of historical traffic data, statistics are collected on the highway operation period; Divide the high-speed operation period into multiple slices, and count the number of operations on each slice; Calculate the ratio of the number of operations on each slice to the total number of operations to obtain the proportion of the number of operations on each slice; Calculate the average value and standard deviation of the high-speed operation period according to the proportion of the number of operations on each slice; The ratio of the standard deviation to the mean value is calculated to obtain the operational dispersion characteristic.

7. The method according to claim 2, characterized in that The step of performing machine learning on the illegal operation vehicle identification model according to the model training sample to obtain a pre-trained illegal operation vehicle identification model includes: Dividing the model training samples into a training set, a test set and a validation set according to a preset division ratio; Inputting the training set into the illegal operation vehicle identification model for machine learning to generate an initial training model; Input the test set into the initial model, and output the probability that the vehicle in the test set is an online-hailing vehicle; According to the probability that the vehicles in the test set are online-hailing vehicles, determine the number of vehicles in the test set that are predicted to be real online-hailing vehicles and are in the positive samples of the test set; Calculate the ratio between the number of online-hailing vehicles in the test set that are predicted to be true and in the positive samples of the test set and the total number of vehicles in the positive samples of the test set to obtain the recall rate; Calculate the number of vehicles in the test set that are predicted to be real online-hailing vehicles and in the positive samples of the test set and the number of vehicles in the test set that are predicted to be online-hailing vehicles, and obtain the accuracy rate; The model is tuned according to the recall rate and the accuracy rate, and the tuned model is used to predict the verification set. When the prediction result meets the preset conditions, the tuned model is used as a pre-trained illegal vehicle identification model.

8. An illegal vehicle identification device, characterized in that: The device comprises: An acquisition module is used to acquire multiple historical traffic data of vehicles to be identified entering and exiting the expressway within a preset time period, each historical traffic data includes expressway checkpoint data and GIS data corresponding to the checkpoint location to which the expressway checkpoint data belongs; A construction module, used to construct the operation characteristics of the vehicle to be identified according to the multiple historical traffic data of the vehicle to be identified, wherein the operation characteristics include the frequency of entering and exiting the preset site, the vehicle operation intensity characteristics, the operation dispersion characteristics, the checkpoint coverage and the vehicle region label; An output module, used to input the operating characteristics of the vehicle to be identified into a pre-trained illegal operation vehicle identification model, and output the probability that the vehicle to be identified is an online-hailing vehicle; The push module is used to push the vehicle information of the vehicle to be identified to the supervision platform when the probability is greater than a preset threshold, so as to verify the legality of the vehicle to be identified.

9. A computer storage medium, characterized in that The computer storage medium stores a plurality of instructions, and the instructions are suitable for being loaded by a processor and executing the method according to any one of claims 1 to 7.

10. An electronic device, characterized in that: include: A processor and a memory; wherein the memory stores a computer program, and the computer program is suitable for being loaded by the processor and executing the method according to any one of claims 1 to 7.

Images