Illegal mining site identification method and device, storage medium and electronic equipment

Abstract

The invention discloses an illegal mining site identification method and device, a storage medium and electronic equipment, and the method comprises the steps: obtaining historical vehicle track data associated with electronic fences of all production and manufacturing enterprises in a preset period, and the production and manufacturing enterprises comprise mineral processing factories for mineral processing and mine enterprises for legal mining of mineral; through the historical vehicle trajectory data, screening out a to-be-identified truck conforming to a preset vehicle type and historical vehicle trajectory data of the to-be-identified truck; determining a plurality of candidate parking points of the to-be-identified truck according to the to-be-identified truck and the historical vehicle trajectory data thereof, and constructing feature factors of the candidate parking points; inputting the feature factors of each candidate stop point into a pre-trained illegal mining site identification model, and outputting the probability that each candidate stop point is a suspicious illegal mining site; and based on the probability corresponding to each candidate stop point, identifying existing suspicious illegal mining sites. Therefore, by adopting the embodiment of the invention, the supervision efficiency and accuracy of illegal illegal mining behaviors can be improved.

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 illegal mining sites. Background Art

[0002] In the process of mining and transportation of mineral resources, illegal mining has caused serious impacts on the ecological environment and resource management. In order to effectively supervise and crack down on these illegal activities, relevant departments urgently need an efficient and accurate monitoring method to detect and deal with illegal mining in a timely manner.

[0003] Related technologies usually rely on video surveillance, observation and reporting by on-site personnel. The installation and maintenance of video surveillance equipment requires a large amount of capital investment, especially in remote areas, where the cost is more significant. The combination of manual inspections and video surveillance relies on manual judgment, which has problems such as slow response and low efficiency, and cannot achieve real-time monitoring and rapid response to illegal mining. Summary of the invention

[0004] The embodiments of the present application provide a method, device, storage medium and electronic device for identifying illegal mining sites. 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 illegal mining sites, the method comprising:

[0006] From the vehicle trajectory database, historical vehicle trajectory data associated with the electronic fences of various manufacturing enterprises within a preset period are obtained, and each manufacturing enterprise includes a mineral processing plant for mineral processing and a mining enterprise that legally mines minerals;

[0007] Through the historical vehicle trajectory data, the trucks to be identified that meet the preset vehicle model and their historical vehicle trajectory data are screened out;

[0008] According to the truck to be identified and its historical vehicle trajectory data, multiple candidate stop points of the truck to be identified are determined, and characteristic factors of each candidate stop point are constructed;

[0009] Input the characteristic factors of each candidate stop point into the pre-trained illegal mining site recognition model, and output the probability that each candidate stop point is a suspected illegal mining site;

[0010] Based on the probability that each candidate stop point is a suspected illegal mining point, the existing suspected illegal mining points are identified.

[0011] Optionally, before obtaining the historical vehicle trajectory data associated with the electronic fences of each manufacturing enterprise within a preset period, the method further includes:

[0012] Get the display name of each manufacturing enterprise;

[0013] Based on the pre-acquired industrial and commercial information, the displayed names of each manufacturing enterprise are cleaned to obtain the real names of each manufacturing enterprise;

[0014] Get the location corresponding to the real name of each manufacturing enterprise;

[0015] An electronic fence is constructed for the location corresponding to the real name of each manufacturing enterprise to obtain the electronic fence of each manufacturing enterprise.

[0016] Optionally, multiple candidate stop points of the truck to be identified are determined based on the truck to be identified and its historical vehicle trajectory data, including:

[0017] Based on the truck to be identified and its historical vehicle trajectory data, multiple initial stops within a preset time distance from the current moment are analyzed;

[0018] Get the stop duration and stop type of each initial stop;

[0019] The docking points whose docking duration is greater than a preset duration threshold and whose docking type is not a preset docking type are divided to obtain a plurality of spare docking points, and the preset docking type is a docking type not related to loading and unloading;

[0020] From the multiple backup stops, a stop point that meets the preset screening conditions is obtained to obtain multiple candidate stop points for the truck to be identified; wherein the preset screening conditions include:

[0021] From multiple alternative stop points, multiple candidate stop points are obtained whose distance from the electronic fence of each manufacturing enterprise is less than a preset number of kilometers; the distance between adjacent candidate stop points among the multiple candidate stop points is less than a preset distance value; the stop type of each candidate stop point does not match the POI tag of the mining enterprise.

[0022] Optionally, construct characteristic factors for each candidate stop, including:

[0023] Calculate the distance between each candidate stop and the mineral processing plant;

[0024] Count the distance ratio of unnamed roads between each candidate stop and the mineral processing plant in the roads of vehicle trajectory data;

[0025] Count the POI tags of mining enterprises and other POI tags within the preset range of each candidate stop point;

[0026] The time characteristics and weather characteristics from each candidate stop to the mineral processing plant are counted. The time characteristics include the trajectory driving time, the trajectory running time interval distribution, and whether the running time is the weekend. The weather characteristics include the weather distribution during the trajectory running period.

[0027] The distance between each candidate stop and the mineral processing plant, the distance ratio of unnamed roads, the POI tags of mining enterprises within a preset range and other POI tags, as well as time characteristics and weather characteristics are taken as characteristic factors of each candidate stop.

[0028] Optionally, follow the steps below to generate a pre-trained illegal mining site identification model, including:

[0029] Filtering multiple target trucks that meet the preset vehicle model and their first vehicle trajectory data through historical vehicle trajectory data;

[0030] Determining a plurality of candidate stop points for each target truck based on each target truck and its first vehicle trajectory data;

[0031] According to multiple candidate stop points of each target truck, a model training sample is constructed, and the model training sample includes positive samples and negative samples;

[0032] Divide the model training samples into training set and test set according to the preset division ratio;

[0033] Create a model to identify illegal mining sites;

[0034] Based on the training set and the test set, machine learning is performed on the illegal mining site recognition model to obtain a pre-trained illegal mining site recognition model.

[0035] Optionally, a model training sample is constructed based on multiple candidate stops for each target truck, including:

[0036] Match multiple candidate stop points of each target truck with the POI tags of the mining enterprise;

[0037] The POI tags of the mining enterprises are matched and the stop points are located in the electronic fence as the first candidate stop points, and the characteristic factors of each first candidate stop point are constructed as positive samples;

[0038] Filtering a first candidate stop point from a plurality of candidate stop points of each target truck to obtain a plurality of final stop points of each target truck;

[0039] Construct the characteristic factors of each final stop of each target truck and mark them as negative samples of illegal mining;

[0040] The positive samples and negative samples are used as model training samples.

[0041] Optionally, based on the training set and the test set, the illegal mining site recognition model is subjected to machine learning to obtain a pre-trained illegal mining site recognition model, including:

[0042] The training set is input into the illegal mining site recognition model for machine learning, and when the learning is completed, an initial illegal mining site recognition model is obtained;

[0043] Input the test set into the initial illegal mining site recognition model and output the model recognition results;

[0044] When the model recognition result is consistent with the labeling result of the test set, the initial illegal mining site recognition model is used as the pre-trained illegal mining site recognition model.

[0045] In a second aspect, an embodiment of the present application provides a device for identifying illegal mining sites, the device comprising:

[0046] An acquisition module is used to acquire historical vehicle trajectory data associated with the electronic fences of various manufacturing enterprises within a preset period from a vehicle trajectory database, wherein the various manufacturing enterprises include mineral processing plants for mineral processing and mining enterprises that legally mine minerals;

[0047] A screening module is used to screen out trucks to be identified that meet the preset vehicle model and their historical vehicle trajectory data through historical vehicle trajectory data;

[0048] A determination module, used to determine multiple candidate stop points of the truck to be identified based on the truck to be identified and its historical vehicle trajectory data, and to construct characteristic factors of each candidate stop point;

[0049] A model processing module is used to input the characteristic factors of each candidate stop point into a pre-trained illegal mining site recognition model, and output the probability that each candidate stop point is a suspected illegal mining site;

[0050] The identification module is used to identify the existing suspected illegal mining sites based on the probability that each candidate stop point is a suspected illegal mining site.

[0051] 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.

[0052] 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.

[0053] The technical solution provided by the embodiments of the present application may have the following beneficial effects:

[0054] In the embodiment of the present application, on the one hand, by associating vehicle trajectory data with the electronic fences of each manufacturing enterprise, it is possible to rely on the existing vehicle trajectory data for big data analysis to identify illegal mining sites. This method is low-cost and makes the monitoring system more economical and efficient. On the other hand, by constructing characteristic factors and using pre-trained models, the characteristic factors can truly reflect the parking information and vehicle operation information of the stop points, so that the illegal mining probability of each candidate stop point can be quickly analyzed and evaluated, so as to automatically identify suspicious sites and realize real-time monitoring and rapid response to illegal mining.

[0055] 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

[0056] 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.

[0057] Figure 1 It is a flow chart of a method for identifying illegal mining sites provided in an embodiment of the present application;

[0058] Figure 2 It is a schematic diagram of a vehicle entering an ore processing plant provided by the present application;

[0059] Figure 3 It is a stop point screening schematic diagram provided by this application;

[0060] Figure 4 It is a flowchart of a model training method provided in an embodiment of the present application;

[0061] Figure 5 It is a structural schematic diagram of an illegal mining point identification device provided in an embodiment of the present application;

[0062] Figure 6 It is a structural schematic diagram of a device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0063] 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.

[0064] 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.

[0065] 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.

[0066] 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.

[0067] The present application provides a method, device, storage medium and electronic device for identifying illegal mining sites to solve the problems existing in the above-mentioned related technical issues. In the embodiments of the present application, on the one hand, by associating vehicle trajectory data with the electronic fences of each manufacturing enterprise, it is possible to rely on the existing vehicle trajectory data for big data analysis to identify illegal mining sites. This method is low-cost and makes the monitoring system more economical and efficient. On the other hand, by constructing characteristic factors and using pre-trained models, the characteristic factors can truly reflect the stop information and vehicle operation information of the stop point, so that the illegal mining probability of each candidate stop point can be quickly analyzed and evaluated, so as to automatically identify suspicious points and realize real-time monitoring and rapid response to illegal mining activities. The following is a detailed description using an exemplary embodiment.

[0068] The following will be combined with the attached Figure 1 -Attached Figure 4 , the illegal mining site identification method provided by the embodiment of the present application is introduced in detail. The method can be implemented by relying on a computer program and can be run on an illegal mining site identification device based on the von Neumann system. The computer program can be integrated into an application or run as an independent tool application.

[0069] See also Figure 1 , which is a flow chart of a method for identifying illegal mining sites in an embodiment of the present application. Figure 1 As shown, the method of the embodiment of the present application may include the following steps:

[0070] S101, acquiring historical vehicle trajectory data associated with the electronic fences of various manufacturing enterprises within a preset period from a vehicle trajectory database, wherein the various manufacturing enterprises include mineral processing plants for mineral processing and mining enterprises that legally mine minerals;

[0071] Among them, the vehicle trajectory database refers to a database system that stores vehicle driving trajectory data, which records the driving path, stop location, time and other information of the vehicle within a specific time period. The preset period refers to a time range pre-defined in data analysis, which is used to filter and extract data within a specific time period. For example, a preset period can be set to the most recent month or the most recent week to obtain historical data within this time period from the vehicle trajectory database. The electronic fence is a virtual geographical boundary defined by GPS or other positioning technologies. The electronic fence is used to identify the geographical scope of mineral processing plants and legal mining enterprises in order to track and monitor vehicles entering and leaving these areas. Manufacturing enterprises refer to enterprises involved in the processing and mining of mineral resources, including mineral processing plants and legal mining enterprises. Mineral processing plants refer to industrial enterprises specializing in the processing of mineral resources. They process the mined raw ore into higher value-added products such as concentrates and metal ingots through crushing, screening, flotation, smelting and other processes. Mining enterprises that legally mine minerals refer to those enterprises that have legal mining rights and mine mineral resources in accordance with national laws and regulations.

[0072] In some embodiments of the present application, before obtaining the historical vehicle trajectory data associated with the electronic fence of each manufacturing enterprise within a preset period, it also specifically includes: obtaining the display name of each manufacturing enterprise; based on the pre-acquired industrial and commercial information, the display name of each manufacturing enterprise is cleansed to obtain the real name of each manufacturing enterprise; obtaining the location corresponding to the real name of each manufacturing enterprise; constructing an electronic fence for the location corresponding to the real name of each manufacturing enterprise to obtain the electronic fence of each manufacturing enterprise.

[0073] Among them, the display name refers to the name used by the company in daily operations and external communications. This may be the company's common name, brand name or abbreviation. Data cleaning refers to the process of processing the collected data to correct errors, remove duplications, and exclude irrelevant information to improve data quality. The purpose of data cleaning in this application is to correct the erroneous or non-standard names in the company's display name to the official registered real name to ensure the accuracy of subsequent analysis. The real name refers to the name of the company officially registered with a government agency, which is used in legal documents, official registrations, and business information. Location refers to the actual geographical location of the company, including latitude and longitude coordinates or specific address information.

[0074] In some possible implementations, suppose that several mineral processing companies need to be supervised. First, the display names of these companies are collected from public channels such as the company's official website and promotional materials, such as "Beijing XX Mining Group" and "XX Mineral Processing Co., Ltd.". The official registration information of these companies is queried using industrial and commercial information platforms (such as Qichacha and Tianyancha). Data cleaning is performed by comparing the display names with the industrial and commercial information. During the data cleaning process, all inconsistent names are corrected to ensure that the name of each company is its real name registered with the industrial and commercial department. The geographical location information corresponding to the real names of these companies is further extracted from the industrial and commercial information. For example, the registered address of "Beijing XX Mining Group" is an industrial park in Chaoyang District, Beijing, while "XX Mineral Processing Co., Ltd." is located in another industrial park in Fengtai District, Beijing. Using the geographic information system (GIS), electronic fences are drawn for each company on the map based on the real names and geographical location information of these companies.

[0075] In some possible implementations, authorized database access rights are used to connect to a freight vehicle trajectory database. The trajectory data of all trucks in the past three months are extracted from the database. Specifically, the vehicle trajectory data associated with the electronic fences of mineral processing enterprises and legal mining enterprises can be screened out through a spatial retrieval algorithm, including all truck trajectory records entering or leaving these electronic fences.

[0076] S102, filtering out trucks to be identified that meet the preset vehicle model and their historical vehicle trajectory data through historical vehicle trajectory data;

[0077] Among them, the preset vehicle type refers to the vehicle type that is pre-set according to a specific purpose or need before the vehicle trajectory analysis is performed. For example, in the context of combating illegal mining, special attention may be paid to large trucks used to transport minerals, such as dump trailers. Dump trailers are a special type of trailer whose design allows the carriage to tilt at a certain angle to unload and rely on the carriage to self-reposition. Dump trailers are suitable for a variety of cargo types, such as sand, stone, soil, garbage, building materials, coal, ore, etc. Its carriage structure and lifting method can be adjusted and optimized according to the type and weight of the cargo.

[0078] In some embodiments of the present application, the historical vehicle trajectory data are only the vehicle trajectories associated with the electronic fences of various manufacturing companies, including the trajectories of private cars and the trajectories of food transported in canteens. Therefore, it is necessary to set a preset vehicle model (such as a dump trailer). Based on the preset vehicle model in the vehicle static data, multiple dump trailers and their historical trajectory data are determined from the historical vehicle trajectory data.

[0079] For example Figure 2 As shown, the vehicles entering the ore processing plant are V1, V2, V3, and V4, and the only dump trailer that passes the preset vehicle model screening is V4.

[0080] S103, determining multiple candidate stop points of the truck to be identified based on the truck to be identified and its historical vehicle trajectory data, and constructing characteristic factors of each candidate stop point;

[0081] In some embodiments of the present application, the specific process of determining multiple candidate stop points for the truck to be identified based on the truck to be identified and its historical vehicle trajectory data includes: analyzing multiple initial stop points within a preset time distance from the current moment based on the truck to be identified and its historical vehicle trajectory data; obtaining the stop time and stop type of each initial stop point; dividing the stop points whose stop time is greater than the preset time threshold and whose stop type is not a preset stop type, and obtaining multiple backup stop points, the preset stop type is a stop type that is not related to loading and unloading; from the multiple backup stop points, obtaining the stop points that meet the preset screening conditions, and obtaining multiple candidate stop points for the truck to be identified.

[0082] Among them, the preset screening conditions include: obtaining multiple candidate stop points from multiple alternative stop points, whose distance from the electronic fence of each manufacturing enterprise is less than a preset number of kilometers; the distance between adjacent candidate stop points among the multiple candidate stop points is less than a preset distance value; the stop type of each candidate stop point does not match the POI tag of the mining enterprise.

[0083] Among them, the initial stop points refer to all the recorded stop points of the truck to be identified in its historical trajectory data. These stop points are the basis for analysis, including but not limited to the starting point, the end point and all the places where the vehicle stops during the journey. The stop duration refers to the length of time the truck stays at each initial stop point, which can be determined by analyzing the time and location information in the vehicle trajectory data. The stop type refers to the classification of the stop points according to the nature of the vehicle's activities at the stop points. For example, it can be divided into loading, unloading, rest, maintenance and other types. The preset time threshold is a pre-set time length standard used to filter out the stops whose stay time exceeds this threshold in the initial stop points. The preset screening conditions refer to a series of standards or rules based on which candidate stops are selected from the backup stops. Candidate stops refer to backup stops that meet the preset screening conditions. These points are considered to be high-risk locations where the truck to be identified may be involved in illegal mining activities.

[0084] In some possible implementations, set N=24 hours, M=30 minutes, K=50 kilometers, and r=10 kilometers. Obtain all stop records of dump trailers in the past 24 hours. From these records, filter out all stops with a stop time of more than M minutes as initial stops. Further filter out those stops with a stop time greater than a preset time threshold (M minutes) and a stop type of loading and unloading, and obtain multiple backup stops. For multiple backup stops, analyze the distances between them, find out the backup stops among multiple candidate stops that are less than K kilometers away from the electronic fences of each manufacturing enterprise, ensure that the distance between consecutive stops between multiple backup stops is less than r kilometers, and filter out the stops for normal mining transportation to obtain multiple candidate stops. For example Figure 3 As shown, the initial stop points may include road stops, refueling stops, and remaining stops, and the final candidate stop points can be obtained after screening according to the preset screening conditions.

[0085] In some embodiments of the present application, the specific process of constructing the characteristic factors of each candidate stop point includes: calculating the distance between each candidate stop point and the mineral processing plant; counting the distance ratio of unnamed roads in the roads in the vehicle trajectory data between each candidate stop point and the mineral processing plant; counting the POI tags of mining enterprises and other POI tags within the preset range of each candidate stop point; counting the time characteristics and weather characteristics from each candidate stop point to the mineral processing plant, the time characteristics include the trajectory driving time, the trajectory operation time interval distribution, and whether the operation time is the weekend, and the weather characteristics include the weather distribution during the trajectory operation; the distance between each candidate stop point and the mineral processing plant, the distance ratio of unnamed roads, the POI tags of mining enterprises and other POI tags within the preset range, as well as the time characteristics and weather characteristics are used as the characteristic factors of each candidate stop point.

[0086] Among them, the distance refers to the actual road distance between the candidate stop and the mineral processing plant. The distance ratio of unnamed roads refers to the proportion of unnamed roads (i.e., roads without formal names or clearly marked on the map) in the vehicle's trajectory data that include the roads between the candidate stop and the mineral processing plant. Unnamed roads may be related to illegal mining activities because they may be used to avoid formal monitoring. The proportion of unnamed roads is calculated by counting the distance of the roads passed by the vehicle according to the road grade and calculating their respective proportions. The time characteristics include the trajectory driving time, the distribution of the trajectory running time interval (day and night), and whether the running time is weekends. The weather characteristics include the weather distribution during the trajectory running (rainy days or bad weather). These features are used to analyze the vehicle's operating mode to identify possible illegal mining activities. For example, if the vehicle is running at night or in bad weather conditions, or on weekends, this may indicate that the vehicle is trying to avoid supervision. The characteristic factor refers to a series of quantitative indicators used to describe and evaluate the candidate stop. The characteristic factors of each candidate stop are shown in Table 1.

[0087] Table 1

[0088]

[0089] S104, inputting the characteristic factors of each candidate stop point into a pre-trained illegal mining site identification model, and outputting the probability that each candidate stop point is a suspected illegal mining site;

[0090] Among them, the illegal mining site identification model is a pre-trained statistical or machine learning model that is used to predict whether a candidate stop point is a suspected illegal mining site based on the input feature factors. The output probability means that the model gives a value between 0 and 1 for each input candidate stop point, indicating the degree of suspicion of the point being an illegal mining site. The higher the probability value, the greater the possibility that the model believes that the candidate stop point is an illegal mining site.

[0091] In some embodiments of the present application, characteristic factors of candidate stop points are collected and sorted, including the distance to the mineral processing plant, the distance ratio of unnamed roads, the POI tags of mining enterprises existing within the preset range and other POI tags, as well as time characteristics (such as driving time, trajectory operation time interval distribution, whether it is a weekend) and weather characteristics (operating time weather distribution). These characteristic factors are input into the illegal mining point identification model. The model evaluates each candidate stop point and outputs the probability that each point is a suspected illegal mining point.

[0092] S105: Based on the probability that each candidate stop point is a suspected illegal mining point, identifying the existing suspected illegal mining points.

[0093] Among them, the suspected illegal mining sites are finally identified as candidate stops that exceed the probability threshold, that is, they are identified as suspected illegal mining sites. These sites are the most likely locations for illegal mining activities that the model believes, requiring further on-site investigation and confirmation by law enforcement agencies.

[0094] In some embodiments of the present application, the model calculates a probability value between 0 and 1 for each candidate stop point, indicating the possibility that the point is a suspected illegal mining point. The probability value output by the model is normalized to a standard interval of [300, 1000]. For example, if the probability output by the model is 0.9, the normalized score may be 900 points. By analyzing the distribution of existing illegal mining points in the model evaluation curve, a suitable score threshold is determined for identifying suspicious points. For example, through analysis, it is found that the scores of actual illegal mining points are mostly distributed above 800 points. According to the determined probability interval (for example, the score is greater than 800 points), the points among the candidate stop points whose scores exceed the threshold are identified, and these points are marked as suspected illegal mining points.

[0095] In the embodiment of the present application, on the one hand, by associating vehicle trajectory data with the electronic fences of each manufacturing enterprise, it is possible to rely on the existing vehicle trajectory data for big data analysis to identify illegal mining sites. This method is low-cost and makes the monitoring system more economical and efficient. On the other hand, by constructing characteristic factors and using pre-trained models, the characteristic factors can truly reflect the parking information and vehicle operation information of the stop points, so that the illegal mining probability of each candidate stop point can be quickly analyzed and evaluated, so as to automatically identify suspicious sites and realize real-time monitoring and rapid response to illegal mining.

[0096] See also Figure 4 , which is a flow chart of a model training method for a green pass vehicle recognition model provided in an embodiment of the present application. Figure 4 As shown, the method of the embodiment of the present application may include the following steps:

[0097] S201, screening out multiple target trucks that meet the preset vehicle model and their first vehicle trajectory data through historical vehicle trajectory data;

[0098] It should be noted that the target truck is the vehicle determined in the model training scenario. The specific implementation process of screening out multiple target trucks that meet the preset model and their first vehicle trajectory data is the same as the specific implementation logic of "screening out multiple target trucks that meet the preset model and their first vehicle trajectory data" in the actual application scenario, which will not be repeated here.

[0099] S202, determining a plurality of candidate stop points for each target truck according to each target truck and its first vehicle trajectory data;

[0100] It should be noted that the specific implementation process of determining multiple candidate stop points for each target truck is the same as the specific implementation logic of "determining multiple candidate stop points for trucks to be identified" in the actual application scenario, and will not be repeated here.

[0101] S203, constructing a model training sample according to the multiple candidate stop points of each target truck, where the model training sample includes a positive sample and a negative sample;

[0102] In some embodiments of the present application, based on multiple candidate stop points for each target truck, the specific process of constructing a model training sample includes: matching the multiple candidate stop points of each target truck with the POI tag of the mining enterprise respectively; taking the stop point that matches the POI tag of the mining enterprise and is located in the electronic fence as the first candidate stop point, and constructing the characteristic factor of each first candidate stop point as a positive sample; filtering the first candidate stop points from the multiple candidate stop points of each target truck to obtain multiple final stop points for each target truck; constructing the characteristic factor of each final stop point of each target truck, and marking it as a negative sample of illegal mining; using the positive samples and negative samples as model training samples.

[0103] It should be noted that the specific implementation process of constructing the characteristic factors of each final stop point of each target truck is the same as the specific implementation logic of "constructing the characteristic factors of each candidate stop point" in the actual application scenario, and will not be repeated here.

[0104] S204, dividing the model training samples into a training set and a test set according to a preset division ratio;

[0105] In some embodiments of the present application, the preset division ratio is 70% for the training set and 30% for the test set.

[0106] S205, creating an illegal mining site identification model;

[0107] In some embodiments of the present application, an illegal mining site identification model can be created using an XGBoost-related neural network. XGBoost (eXtreme Gradient Boosting) is a popular machine learning algorithm that is an optimized implementation of a gradient boosting machine (GBM). XGBoost is an ensemble learning algorithm based on a gradient boosting decision tree (GBDT).

[0108] S206, based on the training set and the test set, perform machine learning on the illegal mining site recognition model to obtain a pre-trained illegal mining site recognition model.

[0109] In some embodiments of the present application, based on the training set and the test set, the illegal mining site recognition model is subjected to machine learning, and the specific process of obtaining the pre-trained illegal mining site recognition model includes: inputting the training set into the illegal mining site recognition model for machine learning, and when the learning is completed, obtaining the initial illegal mining site recognition model; inputting the test set into the initial illegal mining site recognition model, and outputting the model recognition result; when the model recognition result is consistent with the annotation result of the test set, the initial illegal mining site recognition model is used as the pre-trained illegal mining site recognition model.

[0110] In the embodiment of the present application, on the one hand, by associating vehicle trajectory data with the electronic fences of each manufacturing enterprise, it is possible to rely on the existing vehicle trajectory data for big data analysis to identify illegal mining sites. This method is low-cost and makes the monitoring system more economical and efficient. On the other hand, by constructing characteristic factors and using pre-trained models, the characteristic factors can truly reflect the parking information and vehicle operation information of the stop points, so that the illegal mining probability of each candidate stop point can be quickly analyzed and evaluated, so as to automatically identify suspicious sites and realize real-time monitoring and rapid response to illegal mining.

[0111] 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.

[0112] See also Figure 5 , which shows a schematic diagram of the structure of an illegal mining site identification device provided by an exemplary embodiment of the present application. The illegal mining site 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 screening module 20, a determination module 30, a model processing module 40, and an identification module 50.

[0113] The acquisition module 10 is used to acquire historical vehicle trajectory data associated with the electronic fences of various manufacturing enterprises within a preset period from the vehicle trajectory database, and each manufacturing enterprise includes a mineral processing plant for mineral processing and a mining enterprise that legally mines minerals;

[0114] A screening module 20 is used to screen out trucks to be identified that meet a preset vehicle model and their historical vehicle trajectory data through historical vehicle trajectory data;

[0115] A determination module 30, for determining a plurality of candidate stop points of the truck to be identified based on the truck to be identified and its historical vehicle trajectory data, and constructing a characteristic factor of each candidate stop point;

[0116] The model processing module 40 is used to input the characteristic factors of each candidate stop point into a pre-trained illegal mining site identification model, and output the probability that each candidate stop point is a suspected illegal mining site;

[0117] The identification module 50 is used to identify the existing suspected illegal mining sites based on the probability that each candidate stop point is a suspected illegal mining site.

[0118] It should be noted that the illegal mining site identification device provided in the above embodiment only uses the division of the above functional modules as an example when executing the illegal mining site 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 is divided into different functional modules to complete all or part of the functions described above. In addition, the illegal mining site identification device provided in the above embodiment and the illegal mining site identification method embodiment belong to the same concept. The implementation process is detailed in the method embodiment and will not be repeated here.

[0119] 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.

[0120] In the embodiment of the present application, on the one hand, by associating vehicle trajectory data with the electronic fences of each manufacturing enterprise, it is possible to rely on the existing vehicle trajectory data for big data analysis to identify illegal mining sites. This method is low-cost and makes the monitoring system more economical and efficient. On the other hand, by constructing characteristic factors and using pre-trained models, the characteristic factors can truly reflect the parking information and vehicle operation information of the stop points, so that the illegal mining probability of each candidate stop point can be quickly analyzed and evaluated, so as to automatically identify suspicious sites and realize real-time monitoring and rapid response to illegal mining.

[0121] The present application also provides a computer-readable medium having program instructions stored thereon, which, when executed by a processor, implement the illegal mining site identification method provided by the above-mentioned various method embodiments.

[0122] The present application also provides a computer program product comprising instructions, which, when executed on a computer, enables the computer to execute the illegal mining site identification method of each of the above method embodiments.

[0123] See also Figure 6 , which is a schematic diagram of the structure of a device provided in the embodiment of the present application. Figure 6 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 .

[0124] The communication bus 1002 is used to realize the connection and communication between these components.

[0125] 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.

[0126] The network interface 1004 may optionally include a standard wired interface or a wireless interface (such as a WI-FI interface).

[0127] 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.

[0128] 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 6 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 mining site identification application.

[0129] exist Figure 6 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 mining site identification application stored in the memory 1005, and specifically perform the following operations:

[0130] From the vehicle trajectory database, historical vehicle trajectory data associated with the electronic fences of various manufacturing enterprises within a preset period are obtained, and each manufacturing enterprise includes a mineral processing plant for mineral processing and a mining enterprise that legally mines minerals;

[0131] Through the historical vehicle trajectory data, the trucks to be identified that meet the preset vehicle model and their historical vehicle trajectory data are screened out;

[0132] According to the truck to be identified and its historical vehicle trajectory data, multiple candidate stop points of the truck to be identified are determined, and characteristic factors of each candidate stop point are constructed;

[0133] Input the characteristic factors of each candidate stop point into the pre-trained illegal mining site recognition model, and output the probability that each candidate stop point is a suspected illegal mining site;

[0134] Based on the probability that each candidate stop point is a suspected illegal mining point, the existing suspected illegal mining points are identified.

[0135] In one embodiment, before executing the acquisition of historical vehicle trajectory data associated with the electronic fences of each manufacturing enterprise within a preset period, the processor 1001 further performs the following operations:

[0136] Get the display name of each manufacturing enterprise;

[0137] Based on the pre-acquired industrial and commercial information, the displayed names of each manufacturing enterprise are cleaned to obtain the real names of each manufacturing enterprise;

[0138] Get the location corresponding to the real name of each manufacturing enterprise;

[0139] An electronic fence is constructed for the location corresponding to the real name of each manufacturing enterprise to obtain the electronic fence of each manufacturing enterprise.

[0140] In one embodiment, when the processor 1001 determines multiple candidate stop points of the truck to be identified based on the truck to be identified and its historical vehicle trajectory data, the processor 1001 specifically performs the following operations:

[0141] Based on the truck to be identified and its historical vehicle trajectory data, multiple initial stops within a preset time distance from the current moment are analyzed;

[0142] Get the stop duration and stop type of each initial stop;

[0143] The docking points whose docking duration is greater than a preset duration threshold and whose docking type is not a preset docking type are divided to obtain a plurality of spare docking points, and the preset docking type is a docking type not related to loading and unloading;

[0144] From the multiple backup stops, a stop point that meets the preset screening conditions is obtained to obtain multiple candidate stop points for the truck to be identified; wherein the preset screening conditions include:

[0145] From multiple backup stop points, multiple candidate stop points are obtained whose distance from the electronic fence of each manufacturing enterprise is less than a preset number of kilometers; the distance between adjacent candidate stop points among the multiple candidate stop points is less than a preset distance value; the stop type of each candidate stop point does not match the POI tag of the mining enterprise.

[0146] In one embodiment, when constructing the characteristic factors of each candidate stop point, the processor 1001 specifically performs the following operations:

[0147] Calculate the distance between each candidate stop and the mineral processing plant;

[0148] Count the distance ratio of unnamed roads between each candidate stop and the mineral processing plant in the roads of vehicle trajectory data;

[0149] Count the POI tags of mining enterprises and other POI tags within the preset range of each candidate stop point;

[0150] The time characteristics and weather characteristics from each candidate stop to the mineral processing plant are counted. The time characteristics include the trajectory driving time, the trajectory running time interval distribution, and whether the running time is the weekend. The weather characteristics include the weather distribution during the trajectory running period.

[0152] The distance between each candidate stop and the mineral processing plant, the distance ratio of unnamed roads, the POI tags of mining enterprises within a preset range and other POI tags, as well as time characteristics and weather characteristics are taken as characteristic factors of each candidate stop.

[0153] In one embodiment, when the processor 1001 generates a pre-trained illegal mining site identification model, the processor 1001 specifically performs the following operations:

[0154] Filtering multiple target trucks that meet the preset vehicle model and their first vehicle trajectory data through historical vehicle trajectory data;

[0155] Determining a plurality of candidate stop points for each target truck based on each target truck and its first vehicle trajectory data;

[0156] According to multiple candidate stop points of each target truck, a model training sample is constructed, and the model training sample includes positive samples and negative samples;

[0157] Divide the model training samples into training set and test set according to the preset division ratio;

[0158] Create a model to identify illegal mining sites;

[0159] Based on the training set and the test set, machine learning is performed on the illegal mining site recognition model to obtain a pre-trained illegal mining site recognition model.

[0160] In one embodiment, when the processor 1001 constructs a model training sample according to multiple candidate stop points for each target truck, the processor 1001 specifically performs the following operations:

[0161] Match multiple candidate stop points of each target truck with the POI tags of the mining enterprise;

[0162] The POI tags of the mining enterprises are matched and the stop points are located in the electronic fence as the first candidate stop points, and the characteristic factors of each first candidate stop point are constructed as positive samples;

[0163] Filtering a first candidate stop point from a plurality of candidate stop points of each target truck to obtain a plurality of final stop points of each target truck;

[0164] Construct the characteristic factors of each final stop of each target truck and mark them as negative samples of illegal mining;

[0165] The positive samples and negative samples are used as model training samples.

[0166] In one embodiment, the processor 1001 performs machine learning on the illegal mining site identification model based on the training set and the test set to obtain the pre-trained illegal mining site identification model, and specifically performs the following operations:

[0167] The training set is input into the illegal mining site recognition model for machine learning, and when the learning is completed, an initial illegal mining site recognition model is obtained;

[0168] Input the test set into the initial illegal mining site recognition model and output the model recognition results;

[0169] When the model recognition result is consistent with the labeling result of the test set, the initial illegal mining site recognition model is used as the pre-trained illegal mining site recognition model.

[0170] In the embodiment of the present application, on the one hand, by associating vehicle trajectory data with the electronic fences of each manufacturing enterprise, it is possible to rely on the existing vehicle trajectory data for big data analysis to identify illegal mining sites. This method is low-cost and makes the monitoring system more economical and efficient. On the other hand, by constructing characteristic factors and using pre-trained models, the characteristic factors can truly reflect the stop information of the stop points, so that the illegal mining probability of each candidate stop point can be quickly analyzed and evaluated, so as to automatically identify suspicious sites and realize real-time monitoring and rapid response to illegal mining.

[0171] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program, and the program for identifying illegal mining sites 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. Among them, the storage medium of the program for identifying illegal mining sites can be a disk, an optical disk, a read-only storage memory, or a random access memory, etc.

[0172] 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 illegal mining sites, characterized in that: The method comprises: Acquire historical vehicle trajectory data associated with the electronic fences of various manufacturing enterprises within a preset period from a vehicle trajectory database, wherein the various manufacturing enterprises include mineral processing plants for mineral processing and mining enterprises that legally mine minerals; By using the historical vehicle trajectory data, a truck to be identified that meets a preset vehicle model and its historical vehicle trajectory data are screened out; Determine multiple candidate stop points of the truck to be identified based on the truck to be identified and its historical vehicle trajectory data, and construct characteristic factors of each candidate stop point; Input the characteristic factors of each candidate stop point into a pre-trained illegal mining site identification model, and output the probability that each candidate stop point is a suspected illegal mining site; Based on the probability that each candidate stop point is a suspected illegal mining point, the existing suspected illegal mining points are identified.

2. The method according to claim 1, characterized in that Before obtaining the historical vehicle trajectory data associated with the electronic fences of each manufacturing enterprise within a preset period, the method further includes: Obtain the display name of each manufacturing enterprise; According to the pre-acquired industrial and commercial information, the displayed names of the manufacturing enterprises are cleaned to obtain the real names of the manufacturing enterprises; Obtain the location corresponding to the real name of each manufacturing enterprise; An electronic fence is constructed for the position corresponding to the real name of each manufacturing enterprise to obtain the electronic fence of each manufacturing enterprise.

3. The method according to claim 1, characterized in that The step of determining a plurality of candidate stop points of the truck to be identified based on the truck to be identified and its historical vehicle trajectory data includes: Based on the truck to be identified and its historical vehicle trajectory data, analyzing multiple initial stop points within a preset time from the current moment; Get the stop duration and stop type of each initial stop; The docking points whose docking duration is greater than a preset duration threshold and whose docking type is not a preset docking type are divided to obtain a plurality of spare docking points, wherein the preset docking type is a docking type not related to loading and unloading; From the multiple backup stop points, a stop point that meets a preset screening condition is obtained to obtain multiple candidate stop points for the truck to be identified; wherein the preset screening condition includes: From the multiple backup stop points, multiple candidate stop points are obtained whose distance from the electronic fence of each manufacturing enterprise is less than a preset number of kilometers; the distance between adjacent candidate stop points among the multiple candidate stop points is less than a preset distance value; the stop type of each candidate stop point does not match the POI tag of the mining enterprise.

4. The method according to claim 1, characterized in that: The characteristic factors of each candidate stop point are constructed, including: Calculating the distance between each candidate stop point and the mineral processing plant; Counting the distance ratio of the unnamed roads between each candidate stop point and the mineral processing plant in the roads in the vehicle trajectory data; Counting the POI tags of the mining enterprise and other POI tags existing within the preset range of each candidate stop point; Counting the time characteristics and weather characteristics from each candidate stop point to the mineral processing plant, the time characteristics include the trajectory driving time, the trajectory running time interval distribution and whether the running time is a weekend, and the weather characteristics include the weather distribution during the trajectory running period; The distance between each candidate stop point and the mineral processing plant, the distance ratio of the unnamed road, the POI tag of the mining enterprise and other POI tags within the preset range, and the time characteristics and weather characteristics are used as characteristic factors of each candidate stop point.

5. The method according to any one of claims 1 to 4, characterized in that: The following steps are used to generate a pre-trained illegal mining site identification model, including: Filtering out a plurality of target trucks that meet a preset vehicle model and their first vehicle trajectory data through the historical vehicle trajectory data; Determining a plurality of candidate stop points for each target truck based on each target truck and its first vehicle trajectory data; Constructing a model training sample according to a plurality of candidate stop points of each target truck, wherein the model training sample includes a positive sample and a negative sample; Dividing the model training samples into a training set and a test set according to a preset division ratio; Create a model to identify illegal mining sites; Based on the training set and the test set, the illegal mining site identification model is subjected to machine learning to obtain a pre-trained illegal mining site identification model.

6. The method according to claim 5, characterized in that The step of constructing a model training sample according to multiple candidate stop points for each target truck includes: Matching the multiple candidate stop points of each target truck with the POI tags of the mining enterprise respectively; The POI tag of the mining enterprise is matched and the stop point is located within the electronic fence as the first candidate stop point, and the characteristic factor of each first candidate stop point is constructed as a positive sample; Filtering the first candidate stop point from the multiple candidate stop points of each target truck to obtain multiple final stop points of each target truck; Constructing a characteristic factor of each final stop of each target truck and marking it as a negative sample of illegal mining; The positive samples and the negative samples are used as model training samples.

7. The method according to claim 5, characterized in that The method of performing machine learning on the illegal mining site identification model based on the training set and the test set to obtain a pre-trained illegal mining site identification model includes: Inputting the training set into the illegal mining site identification model for machine learning, and obtaining an initial illegal mining site identification model when the learning is completed; Inputting the test set into the initial illegal mining site identification model and outputting the model identification result; When the model recognition result is consistent with the labeling result of the test set, the initial illegal mining site recognition model is used as a pre-trained illegal mining site recognition model.

8. An illegal mining point identification device, characterized in that: The device comprises: An acquisition module, used to acquire historical vehicle trajectory data associated with the electronic fences of various manufacturing enterprises within a preset period from a vehicle trajectory database, wherein the various manufacturing enterprises include mineral processing plants for mineral processing and mining enterprises that legally mine minerals; A screening module, used to screen out trucks to be identified that meet a preset vehicle model and their historical vehicle trajectory data through the historical vehicle trajectory data; A determination module, used to determine a plurality of candidate stop points of the truck to be identified based on the truck to be identified and its historical vehicle trajectory data, and to construct a characteristic factor of each candidate stop point; A model processing module, used for inputting the characteristic factors of each candidate stop point into a pre-trained illegal mining site identification model, and outputting the probability that each candidate stop point is a suspected illegal mining site; The identification module is used to identify the existing suspected illegal mining sites based on the probability that each candidate stop point is a suspected illegal mining site.

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