Shore smuggling vehicle inspection method, device, equipment and storage medium

Abstract

The present application discloses a shore smuggling vehicle inspection method, device, equipment and storage medium. It includes: extracting vehicle behavior characteristic factors based on vehicle trajectory data within a preset time period; inputting the vehicle behavior characteristic factors and corresponding preset weights into a pre-trained smuggling vehicle identification model, and outputting the suspicion score of each vehicle; when the suspicion score of the vehicle is greater than or equal to the preset score threshold, determining that the vehicle is a shore smuggling vehicle. The smuggling vehicle inspection method of the present application analyzes the characteristics of smuggling behavior of vehicle-ship docking, extracts multiple smuggling behavior characteristic factors of the vehicle, and then analyzes and identifies based on the trained machine learning model to obtain identified smuggled vehicles. Provide accurate clues for the customs anti-smuggling department and improve the success rate of investigation.

Description

Technical Field

[0001] The present application relates to the field of traffic big data technology, and more specifically, to a shore smuggling vehicle inspection method, device, equipment and storage medium. Background Art

[0002] Due to the tariff system, price differences between countries and trade restrictions, some criminals engage in smuggling. Cargo smuggling is centered on loading and unloading points, operates across provinces, and radiates to multiple provinces and cities. The scope is wide, and the long distance and large span make it very difficult to obtain evidence and arrest suspects. Therefore, it is urgent to use advanced technical means to effectively monitor and accurately combat smuggling.

[0003] Traditional means of monitoring and combating smuggling often rely on human patrols and simple monitoring equipment, which are inadequate in the vast sea areas and complex terrain. In addition, traditional monitoring methods lack intelligent analysis capabilities and are difficult to quickly identify vehicle smuggling from massive amounts of data. Summary of the invention

[0004] The embodiments of the present application provide a shore smuggling vehicle inspection method, device, equipment and storage medium to at least solve the technical problem in the related art that it is difficult to quickly identify smuggled vehicles.

[0005] According to one aspect of an embodiment of the present application, a shore smuggling vehicle inspection method is provided, comprising:

[0006] Extracting vehicle behavior characteristic factors based on vehicle trajectory data within a preset time period;

[0007] Input the vehicle behavior characteristic factors and corresponding preset weights into a pre-trained smuggled vehicle identification model, and output a suspicion score for each vehicle;

[0008] When the suspicion score of the vehicle is greater than or equal to a preset score threshold, the vehicle is determined to be a shore smuggling vehicle.

[0009] In one embodiment, the vehicle behavior characteristic factors include the stop time in the shore tidal flat area, the stop duration in the shore tidal flat area, the stop POI type in the shore tidal flat area, the number of aggregated parked vehicles, the number of vehicles with similar trajectories in the aggregated vehicles, the number of the same stop points for the aggregated vehicles, whether there is a waiting pick-up point, the terminal POI type, the number of days docked at the shore, the length of time traveling at sea, the landing terminal, the landing time, the angle of the sea driving route, the docking time at sea, the docking duration at sea, and the longitude and latitude of the sea docking point.

[0010] In one embodiment, vehicle parking data is obtained based on vehicle trajectory data within a preset time period;

[0011] Based on the vehicle stop data matching the shore tidal flat area fence, the vehicle's shore tidal flat area stop time, shore tidal flat area stop duration, shore tidal flat area stop POI type, and shore stop days are calculated.

[0012] In one implementation, calculating the number of aggregated parked vehicles includes:

[0013] Based on the improved DBSCAN algorithm, clustering is performed on the vehicles parked in the shore tidal flat area; the improved DBSCAN algorithm includes a set parking distance threshold, a parking time interval threshold and a density threshold;

[0014] Calculate the number of clustered parked vehicles in each cluster of the cluster;

[0015] The aggregated number of parked vehicles for each cluster is taken as the aggregated number of parked vehicles for each vehicle in the cluster.

[0016] In one implementation, calculating the number of similar vehicles in the clustered vehicle trajectories includes:

[0017] For each cluster, obtain the driving trajectory of each vehicle in the cluster within a preset time period before docking at the shore, and compare it with the driving trajectory of other vehicles in the same cluster to obtain the mileage of the same trajectory;

[0018] Based on the ratio of the same trajectory mileage of the two vehicles and the total mileage, the trajectory similarity is obtained;

[0019] When the trajectory similarity is greater than a preset similarity threshold, determining that the trajectories of the two vehicles are similar;

[0020] The number of vehicles in the cluster that are similar to each vehicle's trajectory is calculated separately, and the maximum value is used as the number of vehicles with similar aggregated vehicle trajectories for each vehicle in the cluster.

[0021] In one implementation, calculating the number of identical stop points of the aggregated vehicles includes:

[0022] Obtain the docking points of each vehicle in each cluster within a preset time period before and after docking at the shore;

[0023] Based on the stop distance and time, the stop points of all vehicles in each cluster are clustered to obtain the clustered stop area;

[0024] When the ratio of the number of vehicles in the aggregated parking area to the total number of vehicles in the cluster is greater than a preset threshold, determining that the aggregated parking area is a same parking point;

[0025] The number of clustered parking areas that meet the threshold condition is summarized to obtain the number of common parking points of the clustered vehicles.

[0026] In one implementation, determining whether there is a waiting docking point includes:

[0027] Obtain road stop points with a stop time greater than a preset threshold within a preset time period before the vehicle stops at the shore;

[0028] Calculating the distance between the road stop and the shore stop;

[0029] When the distance is less than a preset threshold, it is determined that there is a waiting docking point.

[0030] In one embodiment, when the vehicle track passes through the ferry terminal fence and the inspection sea area fence, the vehicle is determined to be a sea-going vehicle;

[0031] Based on the sea driving trajectory of the sea-going vehicle, the sea driving time, landing terminal, landing time, sea driving route angle, sea stop time, sea stop duration, and sea stop point latitude and longitude are calculated.

[0032] In one embodiment, before inputting the vehicle behavior characteristic factor and the corresponding preset weight into the pre-trained smuggled vehicle identification model, the method further includes:

[0033] Construct training data sets and test data sets based on the behavioral characteristic factors of multiple vehicles and preset weight data;

[0034] Training the XGBoost algorithm model based on the training data set and using a grid search method to perform parameter tuning;

[0035] Model evaluation is performed based on the test data set to obtain a trained smuggled vehicle recognition model.

[0036] According to another aspect of the embodiment of the present application, a shore smuggling vehicle inspection device is provided, comprising:

[0037] A characteristic factor extraction module is used to extract vehicle behavior characteristic factors based on vehicle trajectory data within a preset time period;

[0038] A model scoring module, used to input the vehicle behavior characteristic factors and corresponding preset weights into a pre-trained smuggled vehicle identification model, and output a suspicion score for each vehicle;

[0039] The smuggling identification module is used to determine that the vehicle is a shore smuggling vehicle when the suspicion score of the vehicle is greater than or equal to a preset score threshold.

[0040] According to another aspect of an embodiment of the present application, there is also provided an electronic device, including a memory and a processor, wherein a computer program is stored in the memory, and the processor is configured to execute the shore smuggling vehicle inspection method through the computer program.

[0041] According to another aspect of the embodiment of the present application, a computer-readable storage medium is provided, in which a computer program is stored, wherein the computer program is configured to execute the above-mentioned shore smuggling vehicle inspection method when running.

[0042] The technical solution provided by the embodiments of the present application may have the following beneficial effects:

[0043] The smuggled vehicle inspection method of the present application analyzes the characteristics of smuggling behavior of vehicle-ship docking, extracts multiple smuggling behavior characteristic factors of the vehicle, and then analyzes and identifies based on the trained machine learning model to obtain the identified smuggled vehicles. By using a machine learning algorithm and combining feature extraction technology, the key features of smuggling behavior can be identified from a large amount of data, thereby improving the accuracy and efficiency of the inspection. This intelligent inspection method not only improves the efficiency of inspection and capture, but also reduces the workload of the customs anti-smuggling department. By timely discovering smuggling behavior, it provides accurate clues for the customs anti-smuggling department and improves the success rate of inspection and capture. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0045] Figure 1 It is a flow chart of a shore smuggling vehicle inspection method according to an embodiment of the present application;

[0046] Figure 2 is a flow chart of another shore smuggling vehicle inspection method according to an embodiment of the present application;

[0047] Figure 3 It is a structural schematic diagram of a shore smuggling vehicle inspection device according to an embodiment of the present application;

[0048] Figure 4 It is a schematic diagram of the structure of an optional electronic device according to an embodiment of the present application. DETAILED DESCRIPTION

[0049] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present application.

[0050] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0051] The following is combined with Figure 1-2 The shore smuggling vehicle inspection method of the present application embodiment is described in detail. Figure 1 As shown, the method mainly includes the following steps:

[0052] S101 extracts vehicle behavior characteristic factors based on vehicle trajectory data within a preset time period.

[0053] In one embodiment, the truck trajectory data within a preset time period is obtained, and the national POI fence data, coastline tidal flat area fence data, national road data, inspection sea area fence data, etc. are obtained. Based on multiple data sources and truck trajectory data analysis, the vehicle behavior characteristic factors are obtained.

[0054] Among them, the vehicle behavior characteristic factors include the stop time in the shore mudflat area, the stop duration in the shore mudflat area, the stop POI type in the shore mudflat area, the number of gathered parked vehicles, the number of vehicles with similar trajectories in the gathered vehicles, the number of the same stop points for gathered vehicles, whether there is a waiting pick-up point, the terminal POI type, the number of days docked at the shore, the length of time driving at sea, the landing terminal, the landing time, the angle of the sea driving route, the stop time at sea, the length of time stopping at sea, and the longitude and latitude of the sea stop point.

[0055] In one implementation, vehicle parking data is first obtained based on vehicle trajectory data within a preset time period.

[0056] Specifically, the trajectory data is cleaned to obtain the basic trajectory data of the truck, such as car_number (license plate), lng (longitude), lat (latitude), time (current timestamp), speed (vehicle speed), duration (direction), link, etc. The trajectory points are matched to the road network and the link of each vehicle is calculated. Based on the points with a speed of 0, the parking details of the vehicle are calculated.

[0057] Furthermore, based on the matching of the shore tidal flat area fence with the vehicle stop data, the vehicle stop time at the shore tidal flat area, the stop duration at the shore tidal flat area, the stop POI type at the shore tidal flat area, and the shore stop days are calculated.

[0058] Specifically, the shore tidal flat area fence is matched according to the vehicle stop details, the vehicle stop time in the shore tidal flat area is calculated, the stop duration is obtained based on the stop start time and end time of the shore tidal flat area, and the longitude and latitude of the stop point are obtained.

[0059] Furthermore, the number of days of docking at the shore within a preset time period is calculated based on the docking details. For example, the number of days of docking at the shore within the past month is counted.

[0060] Furthermore, the docking points with a preset duration are obtained, and the national POI fence data is matched to calculate the POI type of the docking point, and the docking points of large ports operating 24 hours a day are filtered out to obtain the remaining docking points in the shore tidal flat area.

[0061] Furthermore, the number of clustered parked vehicles is calculated, including: clustering the parked vehicles in the shore tidal flat area based on an improved DBSCAN algorithm; the improved DBSCAN algorithm includes a set stop distance threshold, a stop time interval threshold and a density threshold.

[0062] The clustering algorithm selected in the embodiment of the present application is the DBSCAN algorithm. The DBSCAN algorithm is a density-based clustering algorithm. The basic idea is to obtain a cluster connected with maximum density through density-reachable relationship. It mainly has two parameters: Eps: specifies the neighborhood radius (distance of the point), MinPts: density threshold (minimum number of points).

[0063] This application has further optimized this algorithm and added a stop time interval threshold parameter. This is because the stop times of the gathering vehicles must be close, otherwise even if two vehicles are in the same place, if one vehicle stops at 9 am and the other stops at 16 pm, they cannot be the same group of people.

[0064] Use the improved DBSCAN algorithm to cluster the vehicles and obtain multiple clusters. Calculate the number of clustered parking vehicles in each cluster. The number of clustered parking vehicles in each cluster is used as the number of clustered parking vehicles for each vehicle in the cluster. For example, if there are 8 trucks in a cluster, then the number of clustered parking vehicles for each vehicle is 8.

[0065] Furthermore, the number of similar vehicles in the clustered vehicle trajectories is calculated. This includes: for each cluster, obtaining the driving trajectory of each vehicle in the cluster within a preset time period before docking at the shore. For example, for vehicle A, obtaining the driving trajectory of vehicle A 6 hours before docking at the shore. Obtaining the road link before docking.

[0066] Furthermore, the driving trajectories of other vehicles in the same cluster are obtained and compared with the driving trajectories of other vehicles in the same cluster to see whether the road links passed are the same and obtain the mileage of the same trajectory.

[0067] The trajectory similarity is obtained based on the ratio of the same trajectory mileage of the two vehicles to the total mileage. For example, based on the ratio of the same trajectory mileage of vehicle A and vehicle B to the total mileage of vehicle A in the first 6 hours before the stop, the similarity between B and A is obtained. When the trajectory similarity is greater than the preset similarity threshold, it is determined that the trajectories of the two vehicles are similar.

[0068] Furthermore, the number of vehicles in the cluster that are similar to each vehicle's trajectory is calculated separately, and the maximum value is used as the number of vehicles with similar aggregated vehicle trajectories for each vehicle in the cluster. For example, the number of vehicles in the cluster that are similar to vehicle A's trajectory is calculated separately, and the number of vehicles in the cluster that are similar to vehicle B's trajectory is counted separately, until the number of vehicles with similar trajectories to all vehicles is counted. The maximum value among the obtained values ​​is used as the number of vehicles with similar aggregated vehicle trajectories for each vehicle in the cluster.

[0069] Furthermore, the number of common stopping points of the clustered vehicles is calculated, and the stopping points of each vehicle in each cluster within a preset time period before and after stopping at the shore are obtained; based on the stopping distance and time, the stopping points of all vehicles in each cluster are clustered respectively to obtain the clustered clustered stopping area; when the ratio of the number of vehicles in the clustered stopping area to the total number of vehicles in the cluster is greater than a preset threshold, the clustered stopping area is determined to be a common stopping point; the number of clustered stopping areas that meet the threshold condition is summarized to obtain the number of common stopping points of the clustered vehicles in the cluster.

[0070] In one embodiment, the stop points of each vehicle in each cluster in a preset time period are first obtained, and each vehicle has multiple stop points. The stop points of all vehicles in each cluster are clustered again, and the clustering condition is that the stop distance is less than the preset distance and the stop time is greater than the preset time. The number of vehicles in each clustered stop area after clustering is calculated. If the ratio of the number of vehicles in the clustered stop area to the total number of vehicles in the cluster is greater than 60%, the clustered stop area is a common stop point. The number of clustered stop areas that meet the threshold condition is summarized to obtain the number of common stop points.

[0071] For example, suppose there are 10 vehicles in a cluster and a total of 100 stops. The 100 stops are clustered again. If there are 7 vehicles in one stop cluster, the ratio is 70%, then the clustered stop is considered to be a same stop.

[0072] Further, determine whether there is a waiting point.

[0073] Obtain road stop points with a stop time greater than a preset threshold within a preset time period before the vehicle stops at the shore. For example, obtain the stop points within 12 hours before the vehicle stops at the shore tidal flat area, and filter the stop points with a stop time greater than the threshold, such as greater than 30 minutes, as road stop points. Calculate the distance between the road stop point and the shore stop point; if the distance is less than the preset threshold, determine that there is a waiting point for connection. This road stop point is the calculated waiting point for connection.

[0074] Among them, the specific values ​​of the preset distance threshold and the preset time threshold can be set according to actual conditions, and this application does not make any specific restrictions.

[0075] Furthermore, the destination POI type is calculated, and the national POI type is matched according to the stop point after the vehicle stops at the shore to obtain the destination POI type.

[0076] Furthermore, we analyze the situation of vehicles going out to sea. There are two forms of vehicle-ship connection. The first is shore connection, where a small boat transports the goods to the shore and the vehicle loads the goods at the shore. The second is sea connection, where a truck goes out to sea by boat and meets another ship full of goods at sea. After the truck is fully loaded with goods, it takes the boat to land.

[0077] In one embodiment, when the vehicle trajectory passes through the ferry terminal fence and the inspection sea area fence, the vehicle is determined to be a sea-going vehicle.

[0078] Specifically, the ferry terminal fence and the inspection sea area fence are matched according to the vehicle trajectory. Vehicles with trajectories at both the ferry terminal and the inspection sea area are outbound vehicles. The outbound terminal, outbound time, landing terminal, and landing time are obtained, and the outbound time is subtracted from the landing time to calculate the sea travel time.

[0079] Furthermore, based on the sea-going vehicle's sea-going track, a preset number of consecutive adjacent track points are taken to calculate the angle formed by the track points to obtain the sea-going route angle.

[0080] Furthermore, the docking situation at sea is analyzed. Based on the vehicle's sea driving trajectory, the docking point at sea is obtained. Check whether the docking time at the docking point at sea exceeds the preset minute threshold. For the docking points at sea that meet the conditions, extract their docking time at sea, docking time at sea, and longitude and latitude information of the docking point at sea.

[0081] This step extracts multiple vehicle behavior characteristic factors by analyzing the characteristics of smuggling behavior of vehicle-ship connections.

[0082] S102 inputs the vehicle behavior characteristic factors and the corresponding preset weights into the pre-trained smuggled vehicle identification model, and outputs the suspicion score of each vehicle.

[0083] In one embodiment, before inputting the vehicle behavior characteristic factors and the corresponding preset weights into the pre-trained smuggled vehicle identification model, the method further includes: firstly constructing a training data set and a test data set based on the behavior characteristic factors and preset weight data of multiple vehicles; training an XGBoost algorithm model based on the training data set, and using a grid search method to perform parameter tuning; and performing model evaluation based on the test data set to obtain a trained smuggled vehicle identification model.

[0084] In one implementation scenario, 800 vehicles are obtained as a training data set and 200 vehicles are obtained as a test data set. The above factors are input into the XGBoost algorithm model according to their pre-set weights to perform model parameter training. The weight setting can be adjusted according to actual conditions, and this application does not impose any restrictions.

[0085] Furthermore, grid search (GridSearchCV) is used to traverse the parameter grid to find the optimal parameter combination. After each parameter combination, cross validation is used to evaluate the model performance and the best parameters are recorded. The trained smuggled vehicle recognition model is obtained.

[0086] The vehicle data to be identified is input into the trained smuggled vehicle identification model, and the suspicion score of each vehicle is output.

[0087] The embodiment of the present application is based on the XGBoost algorithm model, supports multi-threading and distributed computing, and can perform parallel computing on vehicle behavior characteristic factors of multiple vehicles to improve processing efficiency.

[0088] S103: When the suspicion score of the vehicle is greater than or equal to a preset score threshold, the vehicle is determined to be a shore smuggling vehicle.

[0089] In one embodiment, a preset scoring threshold is determined, and this threshold can be set based on historical data, business requirements, etc. This threshold will be used to distinguish suspicious vehicles from non-suspicious vehicles. The specific value of the threshold is not limited in this embodiment of the application.

[0090] The trained XGBoost model is used to score the vehicles, and for each vehicle, its model score is compared with the preset score threshold. If the model score of the vehicle is greater than or equal to the preset score threshold, the vehicle is determined to be a shore smuggling vehicle.

[0091] Finally, a list of vehicles identified as smuggled is output or stored, along with their scores.

[0092] In order to facilitate understanding of the shore smuggling vehicle identification method provided in the embodiment of the present application, the following is combined with the attached Figure 2 Further description.

[0093] like Figure 2As shown in the figure, vehicle behavior characteristic factors are extracted from vehicle trajectory data. These factors include: the time of stopover in the shore tidal flat area, the duration of stopover in the shore tidal flat area, the type of POI of stopover in the shore tidal flat area, the number of aggregated stopover vehicles, the number of vehicles with similar trajectories, the number of the same stopover points of aggregated vehicles, whether there is a waiting point, the type of terminal POI, the number of days of shore stopover, the duration of sea travel, the landing terminal, the landing time, the angle of the sea travel route, the time of sea stopover, the duration of sea stopover, and the longitude and latitude of the sea stopover point.

[0094] Furthermore, 800 vehicles are selected from the data as the training data set and 200 vehicles as the test data set.

[0095] The extracted feature factors are input into the XGBoost algorithm model according to the preset weights to train the model parameters. The hyperparameters of the model are tuned using methods such as grid search to obtain a trained model.

[0096] The test data set is input into the trained model to generate a model score for each vehicle. If the model score reaches or exceeds the preset threshold, the vehicle is judged to be a suspected vehicle for vehicle-to-ship smuggling and the final result is generated.

[0097] The smuggled vehicle inspection method of the present application analyzes the characteristics of smuggling behavior of vehicle-ship connection, extracts multiple smuggling behavior characteristic factors of the vehicle, and then analyzes and identifies based on the trained machine learning model to obtain the identified smuggled vehicles. By comprehensively analyzing the various behavioral characteristics of the vehicle, such as stop time, trajectory similarity, aggregation, etc., suspicious vehicles involved in vehicle-ship connection smuggling can be more accurately identified to improve accuracy. The XGBoost algorithm is used to automatically process and analyze a large amount of vehicle trajectory data, reducing manual intervention and improving processing efficiency. This intelligent inspection method not only improves the efficiency of inspection and capture, but also reduces the workload of the customs anti-smuggling department. By timely discovering smuggling behavior, it provides accurate clues for the customs anti-smuggling department and improves the success rate of inspection and capture.

[0098] According to another aspect of the embodiment of the present application, a shore smuggling vehicle inspection device for implementing the above shore smuggling vehicle inspection method is also provided. Figure 3 As shown, the device comprises:

[0099] A characteristic factor extraction module 301 is used to extract vehicle behavior characteristic factors based on vehicle trajectory data within a preset time period;

[0100] The model scoring module 302 is used to input the vehicle behavior characteristic factors and the corresponding preset weights into the pre-trained smuggled vehicle identification model and output the suspicion score of each vehicle;

[0101] The smuggling identification module 303 is used to determine that the vehicle is a shore smuggling vehicle when the suspicion score of the vehicle is greater than or equal to a preset score threshold.

[0102] It should be noted that the shore smuggling vehicle inspection device provided in the above embodiment only uses the division of the above functional modules as an example when executing the shore smuggling vehicle inspection 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 shore smuggling vehicle inspection device provided in the above embodiment and the shore smuggling vehicle inspection method embodiment belong to the same concept, and the implementation process thereof is detailed in the method embodiment, which will not be repeated here.

[0103] According to another aspect of an embodiment of the present application, an electronic device corresponding to the shore smuggling vehicle inspection method provided in the aforementioned embodiment is also provided to execute the aforementioned shore smuggling vehicle inspection method.

[0104] Please refer to Figure 4 , which shows a schematic diagram of an electronic device provided by some embodiments of the present application. Figure 4 As shown, the electronic device includes: a processor 400, a memory 401, a bus 402 and a communication interface 403, and the processor 400, the communication interface 403 and the memory 401 are connected via the bus 402; the memory 401 stores a computer program that can be run on the processor 400, and when the processor 400 runs the computer program, it executes the shore smuggling vehicle inspection method provided in any of the aforementioned embodiments of the present application.

[0105] The memory 401 may include a high-speed random access memory (RAM), and may also include a non-volatile memory, such as at least one disk memory. The communication connection between the system network element and at least one other network element is realized through at least one communication interface 403 (which may be wired or wireless), and the Internet, wide area network, local area network, metropolitan area network, etc. may be used.

[0106] The bus 402 may be an ISA bus, a PCI bus, or an EISA bus, etc. The bus may be divided into an address bus, a data bus, a control bus, etc. The memory 401 is used to store programs, and the processor 400 executes the programs after receiving the execution instructions. The shore smuggling vehicle inspection method disclosed in any implementation of the aforementioned embodiment of the present application may be applied to the processor 400, or implemented by the processor 400.

[0107] The processor 400 may be an integrated circuit chip with signal processing capabilities. In the implementation process, each step of the above method can be completed by the hardware integrated logic circuit or software instructions in the processor 400. The above processor 400 can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a readily available programmable gate array (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components. The methods, steps and logic block diagrams disclosed in the embodiments of the present application can be implemented or executed. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor, etc. The steps of the method disclosed in the embodiments of the present application can be directly embodied as a hardware decoding processor to be executed, or the hardware and software modules in the decoding processor can be executed. The software module can be located in a mature storage medium in the field such as a random access memory, a flash memory, a read-only memory, a programmable read-only memory or an electrically erasable programmable memory, a register, etc. The storage medium is located in the memory 401, and the processor 400 reads the information in the memory 401 and completes the steps of the above method in combination with its hardware.

[0108] The electronic device provided in the embodiment of the present application and the shore smuggling vehicle inspection method provided in the embodiment of the present application are based on the same inventive concept and have the same beneficial effects as the methods adopted, operated or implemented therein.

[0109] According to another aspect of the embodiments of the present application, there is also provided a computer-readable storage medium corresponding to the shore smuggling vehicle inspection method provided in the aforementioned embodiments, on which a computer program (i.e., program product) is stored. When the computer program is run by the processor, it will execute the shore smuggling vehicle inspection method provided in any of the aforementioned embodiments.

[0110] It should be noted that examples of computer-readable storage media may also include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other optical or magnetic storage media, which are not listed here one by one.

[0111] The computer-readable storage medium provided in the above-mentioned embodiments of the present application and the shore smuggling vehicle inspection method provided in the embodiments of the present application are based on the same inventive concept and have the same beneficial effects as the methods adopted, run or implemented by the application programs stored therein.

[0112] The technical features of the above embodiments may be arbitrarily combined. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0113] The above embodiments only express several implementation methods of the present invention, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present invention. It should be pointed out that, for a person of ordinary skill in the art, several modifications and improvements can be made without departing from the concept of the present invention, and these all belong to the protection scope of the present invention. Therefore, the protection scope of the present invention patent shall be subject to the attached claims.

Claims

1. A shore smuggling vehicle inspection method, characterized in that: include: Extracting vehicle behavior characteristic factors based on vehicle trajectory data within a preset time period; Input the vehicle behavior characteristic factors and corresponding preset weights into a pre-trained smuggled vehicle identification model, and output a suspicion score for each vehicle; When the suspicion score of the vehicle is greater than or equal to a preset score threshold, the vehicle is determined to be a shore smuggling vehicle.

2. The method according to claim 1, characterized in that The vehicle behavior characteristic factors include the stop time in the shore tidal flat area, the stop duration in the shore tidal flat area, the stop POI type in the shore tidal flat area, the number of gathered parked vehicles, the number of vehicles with similar trajectories in the gathered vehicles, the number of the same stop points for gathered vehicles, whether there is a waiting pick-up point, the terminal POI type, the number of days docked at the shore, the length of time driving at sea, the landing terminal, the landing time, the angle of the sea driving route, the docking time at sea, the docking duration at sea, and the longitude and latitude of the sea docking point.

3. The method according to claim 2, characterized in that Based on the vehicle trajectory data within a preset time period, the vehicle parking data is obtained; Based on the vehicle stop data matching the shore tidal flat area fence, the vehicle's shore tidal flat area stop time, shore tidal flat area stop duration, shore tidal flat area stop POI type, and shore stop days are calculated.

4. The method according to claim 2, characterized in that: Calculating the number of aggregated parked vehicles includes: Based on the improved DBSCAN algorithm, clustering is performed on the vehicles parked in the shore tidal flat area; the improved DBSCAN algorithm includes a set parking distance threshold, a parking time interval threshold and a density threshold; Calculate the number of clustered parked vehicles in each cluster of the cluster; The aggregated number of parked vehicles for each cluster is taken as the aggregated number of parked vehicles for each vehicle in the cluster.

5. The method according to claim 4, characterized in that Calculate the number of similar vehicles in the clustered vehicle trajectories, including: For each cluster, obtain the driving trajectory of each vehicle in the cluster within a preset time period before docking at the shore, and compare it with the driving trajectory of other vehicles in the same cluster to obtain the mileage of the same trajectory; Based on the ratio of the same trajectory mileage of the two vehicles and the total mileage, the trajectory similarity is obtained; When the trajectory similarity is greater than a preset similarity threshold, determining that the trajectories of the two vehicles are similar; The number of vehicles in the cluster that are similar to each vehicle's trajectory is calculated separately, and the maximum value is used as the number of vehicles with similar aggregated vehicle trajectories for each vehicle in the cluster.

6. The method according to claim 4, characterized in that Calculate the number of common stops for the assembled vehicles, including: Obtain the docking points of each vehicle in each cluster within a preset time period before and after docking at the shore; Based on the stop distance and time, the stop points of all vehicles in each cluster are clustered to obtain the clustered stop area; When the ratio of the number of vehicles in the aggregated parking area to the total number of vehicles in the cluster is greater than a preset threshold, determining that the aggregated parking area is a same parking point; The number of clustered parking areas that meet the threshold condition is summarized to obtain the number of common parking points of the clustered vehicles.

7. The method according to claim 2, characterized in that Determine whether there is a waiting point, including: Obtain road stop points with a stop time greater than a preset threshold within a preset time period before the vehicle stops at the shore; Calculating the distance between the road stop and the shore stop; When the distance is less than a preset threshold, it is determined that there is a waiting docking point.

8. The method according to claim 2, characterized in that: When the vehicle track passes through the ferry terminal fence and the inspection sea area fence, the vehicle is determined to be a sea-going vehicle; Based on the sea driving trajectory of the sea-going vehicle, the sea driving time, landing terminal, landing time, sea driving route angle, sea stop time, sea stop duration, and sea stop point latitude and longitude are calculated.

9. The method according to claim 1, characterized in that: Before inputting the vehicle behavior characteristic factor and the corresponding preset weight into the pre-trained smuggled vehicle identification model, the method further includes: Construct training data sets and test data sets based on the behavioral characteristic factors of multiple vehicles and preset weight data; Training the XGBoost algorithm model based on the training data set and using a grid search method to perform parameter tuning; Model evaluation is performed based on the test data set to obtain a trained smuggled vehicle recognition model.

10. A shore smuggling vehicle inspection device, characterized in that: include: A characteristic factor extraction module is used to extract vehicle behavior characteristic factors based on vehicle trajectory data within a preset time period; A model scoring module, used to input the vehicle behavior characteristic factors and corresponding preset weights into a pre-trained smuggled vehicle identification model, and output a suspicion score for each vehicle; The smuggling identification module is used to determine that the vehicle is a shore smuggling vehicle when the suspicion score of the vehicle is greater than or equal to a preset score threshold.

11. An electronic device, characterized in that: It comprises a processor and a memory storing program instructions, wherein the processor is configured to execute the shore smuggling vehicle inspection method as described in any one of claims 1 to 9 when executing the program instructions.

12. A computer-readable medium, characterized in that Computer-readable instructions are stored thereon, and the computer-readable instructions are executed by a processor to implement a shore smuggling vehicle inspection method as described in any one of claims 1 to 9.

Images