Forest region man-made risk assessment method and system based on trajectory big data

Through POI data and mobile phone signaling data analysis, the trajectory characteristics of forest area personnel were extracted, and risk assessment was used to identify high-risk areas. The problem of large amount and diverse forms of risk assessment in urban forest area was solved, and man-made in forest fire prevention was realized Factor prevention and control and advanced perception and real-time early warning of urban safety risks.

CN119090260APending Publication Date: 2024-12-06BEIJING INST OF EMERGENCY MANAGEMENT SCI & TECH (BEIJING EMERGENCY MANAGEMENT EXAMINATION CENT BEIJING HAZARDOUS CHEM REGISTRATION CENT) +2

Patent Information

Application Number
CN202411071912.8
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-06
Publication Date
2024-12-06

AI Technical Summary

Technical Problem

The data on personnel activity in urban forest areas is large in volume and diverse in form, making it difficult to effectively conduct risk assessment, resulting in insufficient human factors prevention and control capabilities in forest fires, and it is difficult for existing technology to achieve advanced perception and real-time early warning of safety risks in key urban areas.

Method used

The forest area coverage is determined through POI data, mobile phone signaling data is collected, personnel trajectory characteristics are extracted, people flow changes and trajectory routes are analyzed, risk assessment is used using DBSCAN clustering algorithm, high-risk areas are identified, and security risk reminders are sent.

Benefits of technology

It has improved the prevention and control capabilities of human factors in forest fire prevention, achieved advanced perception and dynamic identification of safety risks in key urban areas, provided real-time early warning functions, and improved the efficiency of urban safety management.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119090260A_ABST
    Figure CN119090260A_ABST
Patent Text Reader

Abstract

The invention discloses a forest region man-made risk assessment method and system based on trajectory big data, and relates to the technical field of risk assessment, and the method comprises the steps: extracting a forest region coverage range according to map POI data in combination with map grid division; collecting mobile phone signaling data according to the time period; carrying out statistics, analysis and extraction according to the mobile phone signaling data to obtain a personnel track, and carrying out identity recognition on personnel; carrying out pedestrian flow change and forest region personnel track route analysis on the forest region coverage area in a grid division manner; and determining a risk assessment result of the corresponding grid region according to a comparison result of a preset change value of the human traffic change or a comparison result of a clustering result of the trajectory routes of the people in the forest region and a preset density threshold. Through the technical scheme of the invention, the risk assessment of different areas of the forest region is realized, the prevention and control capability of human factors in forest fire prevention is improved, and advanced perception, dynamic identification and real-time early warning of safety risks of urban key areas are realized.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of risk assessment, and in particular to a method for assessing human risk in forest areas based on trajectory big data and a system for assessing human risk in forest areas based on trajectory big data. Background Art

[0002] The safe operation of cities faces various severe challenges, and daily administrative inspections and monitoring are becoming increasingly difficult. It has now become a difficulty and key to stabilizing the security situation in various places. How to use modern technical means to achieve efficient, timely and accurate prediction of urban public safety risks is the research frontier and hot spot to ensure the safety of urban operations.

[0003] A large amount of data related to the location of people is accumulated during the operation of the city, involving information such as the people, tools, and starting and ending points of the trip. These data have the characteristics of individual relevance and comprehensive coverage. These data have good description and analysis capabilities for analyzing the movement patterns and travel portraits of urban personnel. In-depth research on trajectory big data analysis algorithms in urban operation safety management, mining the spatiotemporal aggregation characteristics of human flows, and the correlation characteristics between places and personnel behaviors, will help achieve advanced perception, dynamic identification, and real-time warning of safety risks in key urban areas.

[0004] Regarding the risk management of forest fires, since forest fire prevention is a difficult task involving many contents, large investment and difficult to achieve significant results, human factors are usually the main factor causing forest fires. The workload of forest rangers patrolling fire prevention is heavy, and the range that video cameras can monitor is small. Using mobile phone signaling data to analyze the characteristics of human behavior after entering the forest area can improve the prevention and control capabilities of human factors in forest fire prevention.

[0005] For urban forest areas, a large number of people come in and out every day, which not only leads to a huge amount of data, but also various information data related to the patterns of people's activities. Different data forms have different sampling accuracy, sampling density and coverage. This large-scale and diverse data has brought huge challenges to urban safety. Summary of the invention

[0006] In response to the above problems, the present invention provides a method and system for assessing human risk in forest areas based on trajectory big data. The forest coverage is determined by POI (Point of Interest) data, and mobile phone signaling data within the forest coverage is collected. The trajectory characteristics of personnel are extracted from the mobile phone signaling data, and the user's stay status and movement status when entering the forest are analyzed. Risk assessment is performed on different areas of the forest area through changes in human flow and analysis of the trajectory of personnel in the forest area, and then risk management is performed on high-risk areas, thereby improving the prevention and control capabilities of human factors in forest fire prevention, and realizing advanced perception, dynamic identification and real-time warning of safety risks in key urban areas.

[0007] To achieve the above objectives, the present invention provides a method for assessing human risk in forest areas based on trajectory big data, comprising:

[0008] Extract forest coverage based on map POI data combined with map grid division;

[0009] Collecting mobile phone signaling data within the coverage area of ​​the forest area according to time periods;

[0010] According to the collected mobile phone signaling data, statistics are performed, and personnel trajectories based on time series are analyzed and extracted, and the identities of personnel are identified according to the characteristics of the personnel trajectories;

[0011] Based on the identity of the personnel, the forest area coverage area is divided into grids to analyze the change of human flow and the trajectory of personnel in the forest area;

[0012] The risk assessment result of the corresponding grid area is determined based on the comparison result of the preset change value of the human flow change or the comparison result of the clustering result of the trajectory route of the personnel in the forest area with the preset density threshold.

[0013] In the above technical solution, preferably, the specific process of extracting the forest coverage range according to the map POI data combined with the map grid division includes:

[0014] The acquired map POI data is divided into cells, and the map division grids occupied by the forest coverage area and the forest boundary are determined as the forest coverage range.

[0015] In the above technical solution, preferably, the collected mobile phone signaling data is statistically analyzed and extracted to obtain personnel trajectories based on time series, and personnel are identified according to the characteristics of the personnel trajectory. The specific process includes:

[0016] According to the stop point labels in the mobile phone signaling data, lines are connected in time series to obtain the personnel trajectory corresponding to the mobile phone signaling data;

[0017] Using the spatial deviation method, the offset distance between the stop point of the person's trajectory and the preset normal behavior trajectory or standard model is calculated;

[0018] Determine the stay points whose offset distance exceeds the preset distance threshold as abnormal stay points, remove the abnormal stay points, and reconnect them to form a new personnel trajectory;

[0019] According to whether the personnel trajectory conforms to the crowd trajectory within the coverage of the forest area, the distribution time characteristics of the personnel trajectory and the distribution position characteristics of the personnel trajectory relative to the forest area boundary, the personnel are divided into residents, workers or tourists.

[0020] In the above technical solution, preferably, the forest area coverage is divided into grids for analyzing the change of human flow and the trajectory route of personnel in the forest area based on the identity of the personnel, and the risk assessment result of the corresponding grid area is determined according to the comparison result of the preset change value of the human flow change or the comparison result of the clustering result of the trajectory route of personnel in the forest area with the preset density threshold. The specific process includes:

[0021] Conduct density analysis on the stop points of the personnel trajectories in each grid area within the coverage of the forest area;

[0022] Based on the resident, worker and tourist identities of the people, the number of residents, workers and tourists in each grid area is determined by time period;

[0023] Grid areas where the number of residents or workers is much larger than the number of tourists are regarded as residential areas or work areas, and grid areas where the number of tourists is much larger than the number of residents or workers are regarded as recreational areas;

[0024] According to the personnel trajectory analysis, the grid areas where the change of human flow in the preset time period exceeds the preset ratio are regarded as high-risk areas, and the grid areas where the change of human flow is lower than the preset ratio are regarded as low-risk areas;

[0025] The personnel trajectories are clustered by using the DBSCAN (Density-Based Spatial Clustering of Applications with Noise) clustering algorithm, and the density threshold is set by using the average value of the number of trajectories in multiple grid areas. The clustering results exceeding the density threshold are taken as the main routes, and the clustering results below the density threshold are taken as the deviation routes.

[0026] The grid area through which the trunk route passes is regarded as a low-risk area, and the grid area through which the deviated route passes is regarded as a high-risk area.

[0027] In the above technical solution, preferably, the method for assessing human risk in forest areas based on trajectory big data also includes:

[0028] Security risk reminder information is sent to the mobile phone numbers corresponding to the mobile phone signaling data entering the high-risk area.

[0029] The present invention also proposes a forest area human risk assessment system based on trajectory big data, which uses the forest area human risk assessment method based on trajectory big data disclosed in any one of the above technical solutions, including:

[0030] The forest area determination module is used to extract the forest area coverage based on the map POI data combined with the map grid division;

[0031] A signaling collection module is used to collect mobile phone signaling data within the coverage area of ​​the forest area according to time periods;

[0032] The identity recognition module is used to collect statistics from the mobile phone signaling data, analyze and extract the personnel trajectory based on the time series, and identify the personnel according to the characteristics of the personnel trajectory;

[0033] A route analysis module, for analyzing the change of human flow and the trajectory of people in the forest area by dividing the forest area coverage into grids based on the identity of the people;

[0034] The risk assessment module is used to determine the risk assessment result of the corresponding grid area based on the comparison result of the preset change value of the human flow change or the comparison result of the clustering result of the trajectory route of the personnel in the forest area with the preset density threshold.

[0035] In the above technical solution, preferably, the forest area determination module is specifically used to:

[0036] The acquired map POI data is divided into cells, and the map division grids occupied by the forest coverage area and the forest boundary are determined as the forest coverage range.

[0037] In the above technical solution, preferably, the identity recognition module is specifically used for:

[0038] According to the stop point labels in the mobile phone signaling data, lines are connected in time series to obtain the personnel trajectory corresponding to the mobile phone signaling data;

[0039] Using the spatial deviation method, the offset distance between the stop point of the person's trajectory and the preset normal behavior trajectory or standard model is calculated;

[0040] Determine the stay points whose offset distance exceeds the preset distance threshold as abnormal stay points, remove the abnormal stay points, and reconnect them to form a new personnel trajectory;

[0041] According to whether the personnel trajectory conforms to the crowd trajectory within the coverage of the forest area, the distribution time characteristics of the personnel trajectory and the distribution position characteristics of the personnel trajectory relative to the forest area boundary, the personnel are divided into residents, workers or tourists.

[0042] In the above technical solution, preferably, the route analysis module and the risk assessment module are specifically used for:

[0043] Conduct density analysis on the stop points of the personnel trajectories in each grid area within the coverage of the forest area;

[0044] Based on the resident, worker and tourist identities of the people, the number of residents, workers and tourists in each grid area is determined by time period;

[0045] Grid areas where the number of residents or workers is much larger than the number of tourists are regarded as residential areas or work areas, and grid areas where the number of tourists is much larger than the number of residents or workers are regarded as recreational areas;

[0046] According to the personnel trajectory analysis, the grid areas where the change of human flow in the preset time period exceeds the preset ratio are regarded as high-risk areas, and the grid areas where the change of human flow is lower than the preset ratio are regarded as low-risk areas;

[0047] Perform cluster analysis on the personnel trajectory by using the DBSCAN clustering algorithm, set a density threshold by using the average value of the number of trajectories in multiple grid areas, take the clustering results exceeding the density threshold as the main route, and take the clustering results below the density threshold as the deviation route;

[0048] The grid area through which the trunk route passes is regarded as a low-risk area, and the grid area through which the deviated route passes is regarded as a high-risk area.

[0049] In the above technical solution, preferably, the forest area human risk assessment system based on trajectory big data also includes a risk reminder module, which is used to send safety risk reminder information to the mobile phone number corresponding to the mobile phone signaling data entering the high-risk area.

[0050] Compared with the prior art, the beneficial effects of the present invention are: determining the coverage of the forest area through POI data, and collecting mobile phone signaling data within the coverage of the forest area, extracting the trajectory characteristics of the personnel from the mobile phone signaling data, analyzing the user's stay status and movement status when entering the forest, and conducting risk assessment on different areas of the forest area through changes in human flow and analysis of the trajectory of personnel in the forest area, and then conducting risk management in high-risk areas, thereby improving the prevention and control capabilities of human factors in forest fire prevention, and realizing advanced perception, dynamic identification and real-time warning of safety risks in key urban areas. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Figure 1 A schematic diagram of a process of a method for assessing human risk in forest areas based on trajectory big data disclosed in an embodiment of the present invention;

[0052] Figure 2 A visualized schematic diagram of the flow of people in a region disclosed in an embodiment of the present invention;

[0053] Figure 3 A schematic diagram of a process for dividing personnel trajectories based on cluster analysis disclosed in an embodiment of the present invention. DETAILED DESCRIPTION

[0054] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0055] The present invention is further described in detail below in conjunction with the accompanying drawings:

[0056] like Figure 1 As shown, a method for assessing human risk in forest areas based on trajectory big data provided by the present invention includes:

[0057] Extract forest coverage based on map POI data combined with map grid division;

[0058] Collect mobile phone signaling data within the forest area coverage area according to time periods;

[0059] According to the collected mobile phone signaling data, statistics are collected, analyzed and extracted to obtain the personnel trajectory based on the time series, and the identity of the personnel is identified according to the characteristics of the personnel trajectory;

[0060] Based on the identity of the people, the forest coverage area is divided into grids to analyze the changes in human flow and the trajectory routes of people in the forest area;

[0061] The risk assessment result of the corresponding grid area is determined based on the comparison result of the preset change value of the human flow change or the comparison result of the clustering result of the trajectory route of the personnel in the forest area with the preset density threshold.

[0062] In this implementation, the coverage of the forest area is determined through POI data, and mobile phone signaling data within the coverage of the forest area is collected. The trajectory characteristics of the personnel are extracted from the mobile phone signaling data, and the user's stay and movement status when entering the forest are analyzed. Risk assessment is performed on different areas of the forest area through changes in human flow and analysis of the trajectory of personnel in the forest area, and then risk management is carried out in high-risk areas, which improves the prevention and control capabilities of human factors in forest fire prevention and realizes advanced perception, dynamic identification and real-time warning of safety risks in key urban areas.

[0063] Specifically, the signaling data of mobile phones is collected by determining the forest coverage and base station information, and the data is statistically analyzed to provide data support for risk assessment. The statistical data is used to extract labels, analyze the user's stay state and movement state when entering the forest area, and establish a recognition model for the user's stay state. The trajectory characteristics of people are extracted from the mobile phone signaling data, and the security risk level of people entering the area is divided.

[0064] In the above implementation, preferably, the specific process of extracting the forest coverage range according to the map POI data combined with the map grid division includes:

[0065] The acquired map POI data is divided into cells, and the map division grids occupied by the forest coverage area and the forest boundary are determined as the forest coverage range.

[0066] During the implementation process, we must first determine the location information of the forest to be studied and search through the obtained map POI data. However, POI information cannot accurately identify the coverage of the forest, so after determining the location of the forest, we need to extract the forest coverage area by combining the network map and the grid division. Since the coverage of the forest is generally irregular in shape, and the cells divided by the project are square, in order to ensure that all areas of the forest can be analyzed, the grids occupied by the forest boundary should also be summarized as forest area grids.

[0067] Due to the irregular shape of the forest area, when dividing the grid, the forest area boundary is included in the coverage grid of the forest area even though it only occupies a small part of the grid. This may cause the area to include residents near the forest area or people who come to work nearby. Therefore, it is necessary to identify the identity of each person in the area before conducting a risk assessment of the personnel. In this implementation, the population in a single area can be divided into three research categories, as shown in Table 1 below.

[0068] Table 1 Definition of population study categories within the region

[0069]

[0070] The classification of personnel characteristics can be divided according to the length and frequency of their stay in the forest area. The specific parameters are shown in Table 2 below.

[0071] Table 2 Related symbols and meanings

[0072]

[0073] The calculation formula of the arrival frequency threshold τ of the location within the forest area is shown in Formula 1, t d is the number of arrival days within the period, and k is the research period.

[0074]

[0075] The research period is set to seven days, that is, k = 7, and the frequency threshold is 0.71. By comparing the arrival frequency of people within the range with the frequency threshold, people can be divided. The classification of different groups of people can be divided by formula calculation, such as formula 2 to formula 4.

[0076]

[0077]

[0078]

[0079] In the above implementation, preferably, statistics are performed based on the collected mobile phone signaling data, and personnel trajectories based on time series are analyzed and extracted, and personnel identities are identified based on the characteristics of the personnel trajectories. The specific process includes:

[0080] According to the stop point labels in the mobile phone signaling data, the lines are connected in the time series order to obtain the personnel trajectory corresponding to the mobile phone signaling data;

[0081] The spatial deviation method is used to calculate the offset distance between the stop point of the personnel trajectory and the preset normal behavior trajectory or standard model;

[0082] Determine the stay points whose offset distance exceeds the preset distance threshold as abnormal stay points, remove the abnormal stay points, and reconnect them to form a new personnel trajectory;

[0083] People are classified as residents, workers or tourists based on whether their trajectories conform to the trajectories of the crowd within the coverage of the forest area, the distribution time characteristics of their trajectories, and the distribution position characteristics of their trajectories relative to the forest area boundary.

[0084] In this embodiment, the trajectory feature extraction of the person is to connect the user's stay points by using the location information updated by the signaling data in different time periods, and the connection lines between the stay points can be formed by connecting the adjacent location points in chronological order. When the user moves and updates the location within a period of time, each location update can be regarded as a stay point, and these stay points can be formed by drawing lines on the map to form the movement trajectory of the person. The connection lines of each stay point should be based on the time series, that is, the adjacent location stay points are connected in chronological order to reflect the movement trajectory of the person, but it should be noted that the connection lines between the trajectory points are only a means of fuzzy visualization of the movement trajectory, and do not mean that the person actually walks along this route.

[0085] When extracting personnel trajectories, it is necessary to detect deviations in the trajectory data and analyze abnormalities in the stop points and path deviations in the personnel trajectories. The present invention uses a spatial deviation method to detect deviations in personnel trajectories, and detects deviations in trajectories by comparing the deviations between personnel trajectories and normal behaviors or standard models. Since personnel trajectories are extracted based on stop points, deviations in trajectories can be extracted by detecting abnormal stop points.

[0086] The spatial deviation method is used to detect whether the location of the trajectory stop point is similar to the location distribution of the normal trajectory stop point. This can be judged by calculating the distance between different stop points at the same time point of each trajectory. The Euclidean distance is the most commonly used distance metric, which is used to calculate the straight-line distance between two points. In abnormal trajectory detection, the straight-line distance of different trajectory points at the same time point of each trajectory is calculated in time sequence, and the distance threshold r is set. If the distance between the two stop points is greater than the threshold r, it means that the stop point is abnormal and is not easy to be a stop point on the main trajectory route. The trajectory with this stop point is also summarized as a deviation route. The calculation of the threshold r is shown in Formula 5, where Mean is the distance mean, StandardDeviation is the distance standard deviation, and N is a natural number.

[0087] r=Mean±(N*StandardDeviation) (Formula 5)

[0088] The calculation of the straight-line distance dis between the two trajectory stop points is shown in Formula 6. The data are the coordinates of the two points and

[0089]

[0090] The standard deviation of the distance is calculated as shown in Formula 7.

[0091]

[0092] Where n is the number of stop points.

[0093] By comparing the stop points of the two trajectories with the threshold, the stop points that deviate from the trajectory are eliminated, providing technical support for extracting the main personnel trajectories and providing a data basis for risk assessment.

[0094] In the above implementation, preferably, based on the identity of the personnel, the change of human flow and the trajectory route of the personnel in the forest area are analyzed in the grids of the forest area coverage area, and the risk assessment result of the corresponding grid area is determined according to the comparison result of the preset change value of the human flow change or the comparison result of the clustering result of the trajectory route of the personnel in the forest area with the preset density threshold. The specific process includes:

[0095] The density of the stop points of the personnel trajectories in each grid area within the coverage of the forest area is analyzed;

[0096] Based on the resident, worker and tourist identities of the people, the number of residents, workers and tourists in each grid area is determined by time period;

[0097] Grid areas where the number of residents or workers is much larger than the number of tourists are regarded as residential areas or work areas, and grid areas where the number of tourists is much larger than the number of residents or workers are regarded as recreational areas;

[0098] According to the personnel trajectory analysis, the grid areas where the change of human flow in the preset time period exceeds the preset ratio are regarded as high-risk areas, and the grid areas where the change of human flow is lower than the preset ratio are regarded as low-risk areas;

[0099] The DBSCAN clustering algorithm is used to cluster the personnel trajectories. The density threshold is set by taking the average value of the number of trajectories in multiple grid areas. The clustering results exceeding the density threshold are taken as the main routes, and the clustering results below the density threshold are taken as the deviation routes.

[0100] The grid area passed by the main route is regarded as the low-risk area, and the grid area passed by the deviation route is regarded as the high-risk area.

[0101] In this implementation, the activity trajectories of different people entering the forest area, but when dividing the risk level of people, it is necessary to combine the characteristics of the people and the trajectory of the people. The characteristics of people can be simply divided into residents, workers and tourists. For residents and workers living and working at the border of the forest area, combined with their trajectory routes, if the activity range is only in the border area, the risk level of these people is low. However, the risk level of different people entering the forest area can be further subdivided. This part needs to be combined with the judgment of the extraction of personnel trajectories above. If the trajectory route of tourists entering the forest area is similar to the trajectory route of most tourists, the risk level of these people is low. If the trajectory of some tourists deviates from the mass trajectory, the risk level of these people is high. The specific risk level of personnel is divided as shown in Table 3.

[0102] Table 3 Classification of personnel risk levels

[0103]

[0104] Based on the signaling data and personnel trajectory characteristics of the extracted forest-covered area, the density of the personnel stay points in each grid of the forest-covered area is judged. The judgment needs to be combined with the following characteristics of the area, as shown in Table 4.

[0105] Table 4 Regional characteristics description

[0106]

[0107]

[0108] According to the above implementation, there are generally two ways to divide the risk area in combination with the above feature analysis. One is to analyze the changes in the flow of people in the area at different time periods, and the other is to judge based on the trajectory of people after entering the forest area. When dividing the risk level of the forest area, the residential area and the work area at the forest boundary can be analyzed. By visualizing the flow of people in the area, Figure 2 As shown, if the number of residents or workers in the area is much larger than the number of tourists, the area can be regarded as a residential area or a work area, and risk assessment can be carried out by analyzing the changes in the flow of different personnel. When the number of tourists is much larger than the number of residents or staff, the area can be visualized as a recreational area, and risk assessment can be carried out based on the personnel trajectory routes or changes in the flow of personnel.

[0109] Determine the risk area by the change of human flow. When the human flow in a region increases suddenly during a certain period of time, it can be considered that a large number of people enter the region during that period of time. The increase in the number of people increases the safety risk of the region, and the region can be classified as a high-risk area. If the change trend of human flow in the region is relatively stable during the research period, without sudden increases or decreases, the flow of people in the region is stable and the risk is low.

[0110] The trajectory of the personnel is used to determine the risk. The trajectory data set is clustered using the DBSCAN clustering algorithm. The density threshold is set using the clustering results (it can be set to the average value of the number of trajectories in multiple regions) to distinguish between high-density clusters and low-density clusters, and then the risk area is assessed based on the clusters. The division flow chart is as follows: Figure 3 shown.

[0111] By clustering the trajectories of people, we can determine the main routes for entering the forest area. When the area contains the main route, the area can be divided into a low-risk area, while the offset routes and the areas without personnel trajectories are divided into high-risk areas.

[0112] In the above implementation, preferably, the forest area human risk assessment method based on trajectory big data also includes:

[0113] Security risk reminder information is sent to the mobile phone numbers corresponding to the mobile phone signaling data entering the high-risk area.

[0114] According to the forest area human risk assessment method based on trajectory big data disclosed in the above implementation mode, the assessment process of the above method is specifically described below through specific examples.

[0115] Example 1: Jindingshan Analysis

[0116] (1) Regional division

[0117] There are some villages and factories around the area where Jinding Mountain is located, and the changes in population flow in the area will be more obvious, so this area was chosen as the object of research and analysis.

[0118] After determining the location of the forest, it is necessary to extract the grid of the forest area. Since the forest coverage in this area is irregular in shape, in order to ensure that all forest coverage areas are divided into grids, the residential and working areas on the edge of the forest should also be divided into grids.

[0119] Statistics are made on the daily changes in the flow of people in the area, and the number of people is counted at 30-minute intervals. The flow of people is counted for seven days in total, and the visualization of the flow data for the seven days is given. Statistics are made on the changes in the flow of people of three types in the area: residents, workers, and tourists. If the number of tourists in the area is significantly less than the number of residents, and the number of residents is large, attention should be paid to the changes in the flow of people in marginal residential areas when conducting risk assessment.

[0120] Due to the irregular shape of the forest, in order to ensure that the entire forest area is included in the grid, the residential areas and work sites near the forest are also partially divided during the regional division. By finding the Jindingshan area on the network map and checking the area included in the grid, the location of the village at the edge of the area is determined.

[0121] Visualize the flow of people in the left grid of Jinding Mountain for seven days. Determine the changes and number of staff in this area from the data in the figure. If the flow of tourists and residents is not only large in number but also has a significant change trend, the flow of these two parts of people will have a greater impact on the risk level of the area when conducting risk assessment.

[0122] According to the results of the visualization of regional human flow in the above text, if the human flow data in the grid within seven days, whether it is tourists or residents, has no sudden increase in data changes, it means that the number of people coming to play in the area every day is relatively stable, and this part of the area can be divided into a low-risk area. If the seven-day human flow data changes in some grid areas clearly show a sudden increase in the human flow data of residents and tourists on Saturdays and Sundays, it means that a large number of people come to play in the area on weekends. The sudden increase in people increases the risk level of the area, so the area can be divided into a high-risk area.

[0123] Example 2: Analysis of Sanhuang Mountain

[0124] The Sanhuang Mountain area is a tourist attraction, and the changes in the flow of people are relatively obvious, so this area is selected for analysis. After determining the location of the forest, the forest-covered area is divided into grids.

[0125] When analyzing personnel characteristics, we should pay attention to all POI data included in the grid division range. For example, in the above analysis of the grid division results, the upper half of the grid of Sanhuang Mountain contains the POI data of Yiduqingqing Town. When analyzing the flow of people, we should include this part of the flow of people data to avoid risk assessment errors.

[0126] The Sanhuang Mountain area is a tourist attraction, but it is remote. Although there are a few tourists, most of them are residents of small towns. Due to the existence of small towns in the area, the flow of residents and workers in this area is much greater than the number of tourists visiting here, so when conducting regional risk assessment, some areas can be analyzed by combining the trend of changes in the flow of people in small towns.

[0127] Through the grid division of the area, some grids in the Sanhuang Mountain area grid contain the POI data of Yidu Qingqing Town. When performing regional analysis, it is necessary to first search for the area, determine that the area is located in the grid area of ​​the grid division, and visualize the seven-day passenger flow of the grid.

[0128] The flow of people in the grid under the Sanhuang Mountain area is visualized, and the area is determined to be a recreational area or a residential area based on the fluctuations in the flow of different people in the area. When conducting risk assessment and division, it is necessary to make a judgment based on the changing trends of the flow of different people.

[0129] According to the results of the visualization of the flow of people in the area, if the flow of people with different characteristics within the grid area changes relatively stably within seven days without obvious differences, the area can be classified as a low-risk area. If the flow of residents in the regional grid increases sharply during Saturday and Sunday, but there is no obvious trend of change from Monday to Friday, it is relatively stable, and the number of tourists and staff in the area also fluctuates relatively greatly, the large fluctuation of the flow of people increases the risk level of the area, and the area can be classified as a high-risk area.

[0130] Example 3: Miaofengshan Analysis

[0131] The Miaofeng Mountain area is located in the Miaofeng Mountain Scenic Area. There are many people visiting this area, but relatively few residents living in this area. The changes in the flow of people are relatively obvious, which meets the project requirements.

[0132] After determining the location of Miaofeng Mountain, an analysis of the forest coverage of the area was conducted to ensure that the subsequent forest area grid division would include the entire forest area.

[0133] When identifying and analyzing regional personnel characteristics, it is necessary to statistically analyze the traffic data of all POIs in the grid, which can be obtained by querying the POI data. Miaofeng Mountain is located in the scenic area and is a tourist attraction. However, due to its remote location, the traffic data is small. There are many tourists in the area, but relatively few residents and staff. The results of personnel characteristics division are more obvious. In the risk assessment stage, the changes in the traffic of tourists in the area can be analyzed.

[0134] The population of the regional grid is visualized. There are relatively few residents, workers, and tourists in this area. There are even situations where no one enters the forest area during certain periods of time. Therefore, when conducting risk assessment, the changes in the total number of people entering the area can be used for analysis.

[0135] According to the data analysis of the chart, the number of tourists in some areas increased sharply on Saturday and Sunday, while the trend of the flow of people was relatively stable on the other five days. This shows that a large number of tourists entered the area for fun on the two days of the weekend. The sudden increase in the flow of people increases the risk level of the area, so the area can be classified as a high-risk area. In other parts of the grid area, the flow of people with the three characteristics is relatively small, with only a few people entering the area every day, and there is no obvious sudden increase or decrease in the total number of people within seven days, so the area can be classified as a low-risk area.

[0136] According to the analysis of the above embodiments, it can be seen that POI data is used to find suitable forest areas, and statistical analysis is performed on the flow of people and the places where people stay within the forest area to extract the movement trajectory of people and analyze the travel activity patterns of group people, so as to classify people with different behavioral characteristics. When identifying the forest area, all POI data in the grid where the forest area is located are used to visualize the changes in the flow of people for all people in the area. The changes in the flow of people in different time periods within the cycle are analyzed through the visualization results. If there is a sudden increase or decrease in the flow of people, the area is divided into a high-risk area. On the contrary, if the flow of people changes steadily, the risk of the area is low.

[0137] The present invention further proposes a forest area human risk assessment system based on trajectory big data, which uses the forest area human risk assessment method based on trajectory big data disclosed in any one of the above embodiments, including:

[0138] The forest area determination module is used to extract the forest area coverage based on the map POI data combined with the map grid division;

[0139] The signaling collection module is used to collect mobile phone signaling data within the forest coverage area according to time periods;

[0140] The identity recognition module is used to collect statistics from the mobile phone signaling data, analyze and extract the personnel trajectory based on the time series, and identify the personnel according to the characteristics of the personnel trajectory;

[0141] Route analysis module, used to analyze the flow of people and the trajectory of people in the forest area by dividing the forest area into grids based on the identity of the people;

[0142] The risk assessment module is used to determine the risk assessment result of the corresponding grid area based on the comparison result of the preset change value of the human flow change or the comparison result of the clustering result of the trajectory route of the personnel in the forest area with the preset density threshold.

[0143] In the above implementation, preferably, the forest area determination module is specifically used to:

[0144] The acquired map POI data is divided into cells, and the map division grids occupied by the forest coverage area and the forest boundary are determined as the forest coverage range.

[0145] In the above implementation, preferably, the identity recognition module is specifically used for:

[0146] According to the stop point labels in the mobile phone signaling data, the lines are connected in the time series order to obtain the personnel trajectory corresponding to the mobile phone signaling data;

[0147] The spatial deviation method is used to calculate the offset distance between the stop point of the personnel trajectory and the preset normal behavior trajectory or standard model;

[0148] Determine the stay points whose offset distance exceeds the preset distance threshold as abnormal stay points, remove the abnormal stay points, and reconnect them to form a new personnel trajectory;

[0149] People are classified as residents, workers or tourists based on whether their trajectories conform to the trajectories of the crowd within the coverage of the forest area, the distribution time characteristics of their trajectories, and the distribution position characteristics of their trajectories relative to the forest area boundary.

[0150] In the above implementation, preferably, the route analysis module and the risk assessment module are specifically used to:

[0151] The density of the stop points of the personnel trajectories in each grid area within the coverage of the forest area is analyzed;

[0152] Based on the resident, worker and tourist identities of the people, the number of residents, workers and tourists in each grid area is determined by time period;

[0153] Grid areas where the number of residents or workers is much larger than the number of tourists are regarded as residential areas or work areas, and grid areas where the number of tourists is much larger than the number of residents or workers are regarded as recreational areas;

[0154] According to the personnel trajectory analysis, the grid areas where the change of human flow in the preset time period exceeds the preset ratio are regarded as high-risk areas, and the grid areas where the change of human flow is lower than the preset ratio are regarded as low-risk areas;

[0155] The DBSCAN clustering algorithm is used to cluster the personnel trajectories. The density threshold is set by taking the average value of the number of trajectories in multiple grid areas. The clustering results exceeding the density threshold are taken as the main routes, and the clustering results below the density threshold are taken as the deviation routes.

[0156] The grid area passed by the main route is regarded as the low-risk area, and the grid area passed by the deviation route is regarded as the high-risk area.

[0157] In the above implementation, preferably, the forest area human risk assessment system based on trajectory big data also includes a risk reminder module, which is used to send safety risk reminder information to the mobile phone number corresponding to the mobile phone signaling data entering the high-risk area.

[0158] According to the forest area human risk assessment system based on trajectory big data disclosed in the above-mentioned implementation mode, the functions to be implemented by each module correspond to the steps of the forest area human risk assessment method based on trajectory big data disclosed in the above-mentioned implementation mode. During the implementation process, the operations are performed with reference to the above-mentioned implementation mode, which will not be repeated here.

[0159] The method and system for assessing human risk in forest areas based on trajectory big data disclosed in the above-mentioned implementation method use mobile phone signaling data to study the activity patterns of users, which requires data cleaning, acquisition of stop points, and classification of user behaviors in combination with POI data. After data cleaning using median filtering and sliding window methods, the density-based DBSCN algorithm is used to identify the user's stop points. By integrating map POI data, the user's work-residence characteristics are further analyzed, and the user's work location and residence location are obtained, providing a reliable basis for risk assessment and analysis. In the human risk assessment of urban gas stations, POI data and user stop point information are used to identify various situations of dangerous places and personnel situations near the research object. A risk assessment model is established using hierarchical analysis to evaluate the risk level. Trajectory data is used to comprehensively assess the potential risks in the forest area. By analyzing the user's trajectory chain and personnel gathering situation, and combining POI data, users with abnormal behavior can be quickly identified, providing a reference for managers to conduct safety management.

[0160] Actual cases show that there are man-made risks around many gas stations, and the risk level is high. This shows that the situation where gas stations have man-made risks due to urban construction is common in cities. The cause of the risk is mainly due to the close distance between key places and gas stations. Usually, key places are also places where people gather, and the population density is relatively high, so when conducting risk assessment, the weight of the distance between key places and gas stations will be larger. When conducting risk assessment in key forest areas, there are differences in risk levels in different grid areas, and these risk levels are divided according to the fluctuation of human flow in the area. Usually, high-risk areas in forests are often areas with abnormally large fluctuations in human flow. Therefore, when conducting risk assessment, it is necessary to carefully analyze the changes in human flow of different people in the area.

[0161] The above are only preferred embodiments of the present invention and are not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for assessing human risk in forest areas based on trajectory big data, characterized in that: include: Extract forest coverage based on map POI data combined with map grid division; Collecting mobile phone signaling data within the coverage area of ​​the forest area according to time periods; According to the collected mobile phone signaling data, statistics are performed, and personnel trajectories based on time series are analyzed and extracted, and the identities of personnel are identified according to the characteristics of the personnel trajectories; Based on the identity of the personnel, the forest area coverage area is divided into grids to analyze the change of human flow and the trajectory of personnel in the forest area; The risk assessment result of the corresponding grid area is determined based on the comparison result of the preset change value of the human flow change or the comparison result of the clustering result of the trajectory route of the personnel in the forest area with the preset density threshold.

2. The method for assessing human risk in forest areas based on trajectory big data according to claim 1 is characterized in that: The specific process of extracting the forest area coverage according to the map POI data combined with the map grid division includes: The acquired map POI data is divided into cells, and the map division grids occupied by the forest coverage area and the forest boundary are determined as the forest coverage range.

3. The method for assessing human risk in forest areas based on trajectory big data according to claim 2 is characterized in that: The process of collecting mobile phone signaling data, analyzing and extracting personnel trajectories based on time series, and identifying personnel based on the characteristics of the personnel trajectories includes: According to the stop point labels in the mobile phone signaling data, lines are connected in time series to obtain the personnel trajectory corresponding to the mobile phone signaling data; Using the spatial deviation method, the offset distance between the stop point of the person's trajectory and the preset normal behavior trajectory or standard model is calculated; Determine the stay points whose offset distance exceeds the preset distance threshold as abnormal stay points, remove the abnormal stay points, and reconnect them to form a new personnel trajectory; According to whether the personnel trajectory conforms to the crowd trajectory within the coverage of the forest area, the distribution time characteristics of the personnel trajectory and the distribution position characteristics of the personnel trajectory relative to the forest area boundary, the personnel are divided into residents, workers or tourists.

4. The method for assessing human risk in forest areas based on trajectory big data according to claim 3 is characterized in that: The process of analyzing the change of human flow and the trajectory of people in the forest area by dividing the forest area coverage into grids based on the identity of the people, and determining the risk assessment result of the corresponding grid area according to the comparison result of the preset change value of the human flow change or the comparison result of the clustering result of the trajectory of people in the forest area with the preset density threshold, specifically includes: Conduct density analysis on the stop points of the personnel trajectories in each grid area within the coverage of the forest area; Based on the resident, worker and tourist identities of the people, the number of residents, workers and tourists in each grid area is determined by time period; Grid areas where the number of residents or workers is much larger than the number of tourists are regarded as residential areas or work areas, and grid areas where the number of tourists is much larger than the number of residents or workers are regarded as recreational areas; According to the personnel trajectory analysis, the grid areas where the change of human flow in the preset time period exceeds the preset ratio are regarded as high-risk areas, and the grid areas where the change of human flow is lower than the preset ratio are regarded as low-risk areas; Perform cluster analysis on the personnel trajectory by using the DBSCAN clustering algorithm, set a density threshold by using the average value of the number of trajectories in multiple grid areas, take the clustering results exceeding the density threshold as the main route, and take the clustering results below the density threshold as the deviation route; The grid area through which the trunk route passes is regarded as a low-risk area, and the grid area through which the deviated route passes is regarded as a high-risk area.

5. The method for assessing human risk in forest areas based on trajectory big data according to claim 4 is characterized in that: Also includes: Security risk reminder information is sent to the mobile phone numbers corresponding to the mobile phone signaling data entering the high-risk area.

6. A forest area human risk assessment system based on trajectory big data, characterized in that: The method for assessing human risk in forest areas based on trajectory big data according to any one of claims 1 to 5 is applied, comprising: The forest area determination module is used to extract the forest area coverage based on the map POI data combined with the map grid division; A signaling collection module is used to collect mobile phone signaling data within the coverage area of ​​the forest area according to time periods; The identity recognition module is used to collect statistics from the mobile phone signaling data, analyze and extract the personnel trajectory based on the time series, and identify the personnel according to the characteristics of the personnel trajectory; A route analysis module, for analyzing the change of human flow and the trajectory of people in the forest area by dividing the forest area coverage into grids based on the identity of the people; The risk assessment module is used to determine the risk assessment result of the corresponding grid area based on the comparison result of the preset change value of the human flow change or the comparison result of the clustering result of the trajectory route of the personnel in the forest area with the preset density threshold.

7. The forest area human risk assessment system based on trajectory big data according to claim 6 is characterized in that: The forest area determination module is specifically used for: The acquired map POI data is divided into cells, and the map division grids occupied by the forest coverage area and the forest boundary are determined as the forest coverage range.

8. The forest area human risk assessment system based on trajectory big data according to claim 7 is characterized in that: The identity recognition module is specifically used for: According to the stop point labels in the mobile phone signaling data, lines are connected in time series to obtain the personnel trajectory corresponding to the mobile phone signaling data; Using the spatial deviation method, the offset distance between the stop point of the person's trajectory and the preset normal behavior trajectory or standard model is calculated; Determine the stay points whose offset distance exceeds the preset distance threshold as abnormal stay points, remove the abnormal stay points, and reconnect them to form a new personnel trajectory; According to whether the personnel trajectory conforms to the crowd trajectory within the coverage of the forest area, the distribution time characteristics of the personnel trajectory and the distribution position characteristics of the personnel trajectory relative to the forest area boundary, the personnel are divided into residents, workers or tourists.

9. The forest area human risk assessment system based on trajectory big data according to claim 8 is characterized in that: The route analysis module and the risk assessment module are specifically used for: Conduct density analysis on the stop points of the personnel trajectories in each grid area within the coverage of the forest area; Based on the resident, worker and tourist identities of the people, the number of residents, workers and tourists in each grid area is determined by time period; Grid areas where the number of residents or workers is much larger than the number of tourists are regarded as residential areas or work areas, and grid areas where the number of tourists is much larger than the number of residents or workers are regarded as recreational areas; According to the personnel trajectory analysis, the grid areas where the change of human flow in the preset time period exceeds the preset ratio are regarded as high-risk areas, and the grid areas where the change of human flow is lower than the preset ratio are regarded as low-risk areas; Perform cluster analysis on the personnel trajectory by using the DBSCAN clustering algorithm, set a density threshold by using the average value of the number of trajectories in multiple grid areas, take the clustering results exceeding the density threshold as the main route, and take the clustering results below the density threshold as the deviation route; The grid area through which the trunk route passes is regarded as a low-risk area, and the grid area through which the deviated route passes is regarded as a high-risk area.

10. The forest area human risk assessment system based on trajectory big data according to claim 8, characterized in that: It also includes a risk reminder module, which is used to send security risk reminder information to the mobile phone numbers corresponding to the mobile phone signaling data entering the high-risk area.

Citation Information

Patent Citations

  • Street planning demand identification method based on multi-source big data

    CN112288472A

  • Urban epidemic situation risk assessment method based on mobile phone signaling data

    CN113784292A

  • High-precision spatio-temporal trajectory restoration method based on mobile phone signaling data

    CN116132923A

  • Gas station risk assessment method, device and system and storage medium

    CN118313651A

Cited By

  • LBS (Location Based Service)-based river-lake four-disorder problem studying and judging method and system

    CN121981411A

  • A method and system for researching and judging river and lake four disorder problems based on LBS

    CN121981411B