Vehicle track intelligent GIS electronic fence visualization system based on GPS

Abstract

The invention belongs to the technical field of intelligent traffic monitoring, and relates to a vehicle track intelligent GIS electronic fence visualization system based on a GPS, and the system comprises a track data collection module which is used for obtaining the historical driving track point data of a plurality of vehicles in a preset time period; the aggregation area identification module is used for identifying a high-frequency aggregation area of the historical driving track point data; the elastic fence generation module is used for expanding a dynamic buffer area outwards by taking the boundary of the high-frequency aggregation area as a reference; the border crossing state judgment module is used for calculating the spatial relationship between the real-time position of the vehicle and the elastic electronic fence; the behavior analysis module is used for identifying repeated border-crossing road sections and high-incidence time periods; and the multi-source visualization module is used for overlapping and displaying the elastic electronic fence layer, the real-time vehicle position marking layer and the border-crossing thermodynamic layer on the GIS map. The problem that a closed-loop optimization mechanism based on data feedback cannot be formed is solved.

Description

Technical Field

[0001] The invention belongs to the technical field of intelligent traffic monitoring and relates to a vehicle track intelligent GIS electronic fence visualization system based on GPS. Background Art

[0002] In the field of vehicle monitoring, accurately defining the boundaries of operating areas and effectively warning of violations have long been a technical challenge. In real-world scenarios, areas where vehicles frequently gather often change dynamically with business needs. For example, loading and unloading areas in logistics parks and construction sites exhibit significant temporal and spatial migration. Traditional fixed-boundary electronic fences struggle to adapt to these dynamic changes, resulting in frequent false alarms and missed alerts in monitoring systems. The spatial and temporal patterns inherent in historical violations have not been fully explored, resulting in a lack of predictability in risk warnings and a lag in regulatory responses after incidents occur.

[0003] Current mainstream solutions employ two main technical approaches: One is to pre-set fixed geofence boundaries based on manual experience, and to determine transgression status based on the geometric relationship between GPS positioning points and the pre-set boundaries; the other is to generate fixed-width buffer zones through equidistant expansion, triggering a primary warning when a vehicle enters the buffer zone. Historical behavior analysis typically utilizes a separate reporting system, which calculates the spatial distribution and time-period frequency of transgression events over a fixed period and outputs the analysis results in the form of static charts. These solutions rely on a pre-set rule base to implement basic monitoring functions, and the data flow between modules is linear.

[0004] To address the above issues, the three functional modules of the traditional solution – fence generation, status determination, and behavior analysis – are independent of each other and cannot form a closed-loop optimization mechanism based on data feedback, which restricts the overall performance improvement of the monitoring system. Summary of the Invention

[0005] In order to solve the above problems, the present invention provides a vehicle trajectory intelligent GIS electronic fence visualization system based on GPS.

[0006] A GPS-based vehicle trajectory intelligent GIS electronic fence visualization system, including:

[0007] The trajectory data acquisition module is used to obtain the historical driving trajectory point data of multiple vehicles within a preset time period. Each driving trajectory point data contains longitude, latitude and timestamp;

[0008] Aggregation area recognition module, used to identify high-frequency aggregation areas of historical driving trajectory point data;

[0009] The elastic fence generation module is used to implement a dynamic buffer zone that expands outward based on the boundary of the high-frequency clustering area. The width of the buffer zone increases as the density of trajectory points at the edge of the area decreases.

[0010] An out-of-bounds status determination module is used to calculate the spatial relationship between the vehicle's real-time position and the elastic electronic fence, marking the vehicle as in a warning state when it enters the buffer zone and as in an out-of-bounds state when it crosses the core boundary of the fence;

[0011] The behavior analysis module is used to establish a spatiotemporal distribution model of historical boundary crossing behaviors. By correlating the location of boundary crossings with the time of occurrence, it can identify sections of road where repeated boundary crossings occur and the time periods when they are most frequent.

[0012] The multi-source visualization module is used to overlay and display the elastic electronic fence layer, real-time vehicle location mark layer, and out-of-bounds thermal layer on the GIS map. The out-of-bounds thermal layer uses color concentration to represent the frequency of out-of-bounds in high-incidence areas identified by the behavior analysis module.

[0013] A further solution of the present invention is to obtain historical driving trajectory point data of multiple vehicles within a preset time period, including the following steps:

[0014] Obtain the original trajectory point sequence from the preset vehicle terminal interface;

[0015] Delete abnormal track points whose longitude and latitude are beyond the preset effective range;

[0016] Remove drifting track points whose distance between adjacent track points exceeds the preset maximum reachable distance;

[0017] The processed denoised trajectory dataset is stored in groups according to vehicle ID.

[0018] A further solution of the present invention is to identify high-frequency clustered areas of historical driving trajectory point data, including the following steps:

[0019] Construct rectangular grid cells with preset longitude and latitude differences;

[0020] Calculate the product of the number of trajectory points and the dwell time of each grid cell, where the dwell time is the difference between the timestamps of the first and last trajectory points in the grid;

[0021] The preset density threshold is dynamically set according to the ratio of the total number of trajectory points to the total area of ​​the grid.

[0022] A further solution of the present invention, identifying high-frequency clustered areas of historical driving trajectory point data, further includes the following steps:

[0023] Mark the grids whose product values ​​exceed the preset density threshold as high-density cells;

[0024] Merge adjacent high-density units to form continuous blocks;

[0025] When the total stay time in the block exceeds the preset time threshold, the coordinate sequence of the clustering area boundary is output.

[0026] A further solution of the present invention is to expand the dynamic buffer zone outward based on the boundary of the high-frequency aggregation area, including the following steps:

[0027] Set calculation points along the boundary line at preset intervals;

[0028] A circular analysis area with a preset radius is established with each calculation point as the center;

[0029] The ratio of the number of trajectory points in the analysis area to the maximum trajectory point carrying capacity is calculated as the local density ratio;

[0030] Get the local density ratio of the current point, subtract the density ratio from 1 to get the scarcity coefficient, and dynamically determine the width of each point by adding the basic buffer value and the scarcity coefficient multiplied by the maximum expansion value.

[0031] A further solution of the present invention is to calculate the spatial relationship between the real-time position of the vehicle and the elastic electronic fence, comprising the following steps:

[0032] Convert the vehicle's current position coordinates into plane coordinate system coordinates;

[0033] The ray method is used to determine the positional relationship between the coordinate point and the inner core boundary;

[0034] When the coordinate point is within the outer dynamic buffer zone, the warning state is activated;

[0035] The out-of-bounds state is activated when the coordinate points cross the inner core boundary at the same time.

[0036] A further solution of the present invention is to establish a spatiotemporal distribution model of historical boundary crossing behaviors, comprising the following steps:

[0037] Create an analysis circle of out-of-bounds points with a preset radius;

[0038] Merge the cross-border points contained in the overlapping analysis circles to form the cross-border road section;

[0039] Divide and calculate the percentage of boundary crossing incidents by preset time periods;

[0040] When the frequency of crossing the boundary on a specific road section during a specific period exceeds the preset risk threshold, it will be marked as a high-frequency risk point.

[0041] A further solution of the present invention is to overlay and display the elastic electronic fence layer, the real-time vehicle location mark layer, and the out-of-bounds thermal layer on the GIS map, including the following steps:

[0042] Divide map units into preset grid sizes;

[0043] Calculate the sum of the cross-border frequency ratios of each unit based on the cross-border behavior distribution feature table;

[0044] Match the preset time period weight coefficient according to the current time period;

[0045] The heat value is obtained by multiplying the sum of the frequency ratios of crossing the boundary by the time period weight coefficient;

[0046] Rendering is done by mapping a gradient color spectrum according to the thermal value interval.

[0047] A further solution of the present invention activates the out-of-bounds state when the coordinate points simultaneously cross the inner core boundary, including the following steps:

[0048] Get the out-of-bounds road section identifier of the current out-of-bounds location;

[0049] Retrieve the high-incidence time period of the corresponding road section in the cross-border behavior distribution feature table;

[0050] When the current time falls into the high-incidence period, the warning level is raised and the preset emergency response protocol is triggered.

[0051] In summary, the present invention has the following beneficial technical effects:

[0052] 1. Significantly improve the accuracy and adaptability of regional monitoring. Dynamic analysis of historical trajectory data automatically identifies high-frequency clusters and adaptively generates non-uniform buffer zones based on boundary trajectory point density, breaking through the limitations of traditional fixed boundaries of electronic fences. Core operating areas maintain tight protection, while buffer zones are expanded in low-activity areas, precisely matching monitoring intensity with regional risk levels. This adaptive mechanism effectively reduces false alarms and missed alerts caused by inappropriate boundary demarcation in complex environments, improving the monitoring system's adaptability to real-world business scenarios.

[0053] 2. The solution innovatively integrates real-time location determination with historical behavior modeling to establish a dual-dimensional spatial and temporal risk analysis framework. By linking border crossing locations with high-incidence periods to generate heat maps, discrete border crossing events are transformed into a continuous, visual representation of the risk landscape. When a vehicle enters the buffer zone during real-time monitoring, the system automatically links historical risk data at that location and implements an upgraded warning for border crossings occurring during high-frequency periods or on high-risk road sections.

[0054] 3. A dynamic response mechanism, based on the distribution characteristics of boundary violations, enables intelligent scheduling of regulatory resources. The system automatically identifies the distribution patterns of repeated boundary violations along sections of road and within time periods, directing patrols to high-risk areas identified by heat maps. When a boundary violation occurs, a hierarchical response protocol is automatically triggered by real-time matching of historical high-incidence period data, initiating pre-defined handling procedures for patterned violations. This mechanism significantly shortens the time window from incident occurrence to handling intervention, while also reducing regulatory resource consumption in low-risk areas and improving the overall operational efficiency of the security system. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. The drawings are used to provide a further understanding of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0056] Figure 1 It is a schematic diagram of the framework in the embodiment of the present application.

[0057] Figure 2 It is a schematic diagram of the process flow in the embodiment of this application. DETAILED DESCRIPTION

[0058] To make the objectives, technical solutions, and advantages of the embodiments of the present invention more clear, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of them. All other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0059] The following is combined with Figure 1-Figure 2 The preferred embodiments of the present invention are described in detail.

[0060] Refer to the attached Figure 1-Figure 2 ,This paper proposes a GPS-based vehicle trajectory intelligent GIS electronic fence visualization system, which includes the following modules:

[0061] The trajectory data acquisition module is used to obtain the historical driving trajectory point data of multiple vehicles within a preset time period. Each driving trajectory point data contains longitude, latitude and timestamp;

[0062] Aggregation area recognition module, used to identify high-frequency aggregation areas of historical driving trajectory point data;

[0063] The elastic fence generation module is used to implement a dynamic buffer zone that expands outward based on the boundary of the high-frequency clustering area. The width of the buffer zone increases as the density of trajectory points at the edge of the area decreases.

[0064] An out-of-bounds status determination module is used to calculate the spatial relationship between the vehicle's real-time position and the elastic electronic fence, marking the vehicle as in a warning state when it enters the buffer zone and as in an out-of-bounds state when it crosses the core boundary of the fence;

[0065] The behavior analysis module is used to establish a spatiotemporal distribution model of historical boundary crossing behaviors. By correlating the location of boundary crossings with the time of occurrence, it can identify sections of road where repeated boundary crossings occur and the time periods when they are most frequent.

[0066] The multi-source visualization module is used to overlay and display the elastic electronic fence layer, real-time vehicle location mark layer, and out-of-bounds thermal layer on the GIS map. The out-of-bounds thermal layer uses color concentration to represent the frequency of out-of-bounds in high-incidence areas identified by the behavior analysis module.

[0067] In one embodiment of the present invention, obtaining historical driving trajectory point data of multiple vehicles within a preset time period includes the following steps:

[0068] Connect to the data interface of a vehicle-mounted GPS terminal or vehicle monitoring platform to batch retrieve driving records for multiple vehicles based on a preset time period (e.g., the past 30 days). When a vehicle's engine is started, its GPS sensor collects longitude, latitude, and corresponding timestamps accurate to the second at fixed intervals (typically 15-60 seconds), forming a sequence of raw track points. This raw data is denoised by removing outliers with latitude values ​​outside the range of -90° to 90° and longitude values ​​outside the range of -180° to 180°. If the distance between adjacent track points exceeds the maximum achievable distance of the vehicle (e.g., for a truck traveling at 120 km / h, the maximum displacement within a 10-second interval is 333 meters), the point is identified as a signal drift and removed. The maximum achievable distance for denoising depends on the vehicle type. For trucks, the value is the product of the time interval (in seconds) between adjacent track points and a speed of 120 km / h.

[0069] Historical driving trajectory data refers to the periodic recording of vehicle location information during driving. Each vehicle's data is stored independently as a time-ordered sequence. The preset time period is set by the system administrator based on business needs, with the minimum unit being one day.

[0070] Longitude and latitude are recorded in the WGS-84 coordinate system, directly from the vehicle's GPS sensor's built-in positioning chip. Timestamps are recorded in UTC, synchronized with GPS satellite signals, in the format of "year-month-day hour:minute:second."

[0071] For example, a logistics company retrieved data from the on-board terminals of 50 cold chain transport vehicles from June 1 to 30, 2023; it was found that the recording point of a certain vehicle at 09:30:02 on June 15 was (longitude 181.5° latitude 35.2°), which was deleted because the longitude exceeded the valid range; it was detected that the time interval between adjacent recording points of a certain vehicle was 10 seconds, and the coordinates jumped from point A (116.404°E, 39.915°N) to point B (116.900°E, 40.002°N), and the straight-line distance between the two points reached 45 kilometers (exceeding the maximum displacement of 333 meters of the truck in 10 seconds), and it was determined to be signal drift and point B was removed; finally, a data set with an average of 1,200 valid trajectory points per vehicle per day was generated.

[0072] In one embodiment of the present invention, identifying high-frequency clustered areas in historical driving trajectory point data includes the following steps:

[0073] The historical driving trajectory point data output by the trajectory data acquisition module is grouped by vehicle ID, and spatial grid division is performed for the data of a single vehicle: a rectangular grid is constructed with 0.001° longitude difference and 0.001° latitude difference as units to cover the trajectory point distribution range.

[0074] For each grid cell, the product of the number of track points and the dwell time in that grid cell is calculated. The dwell time is determined by the maximum time difference between adjacent track points within the grid cell. First, the total number of track points is counted, and 5% of this is taken as a baseline value. This baseline value is then divided by the total grid area of ​​the entire analysis region to determine the required density per unit area. For example, if 100,000 track points correspond to a baseline value of 5,000 points and the total grid area is 50 square kilometers, the density threshold is 100 points per square kilometer. A grid cell is marked as having a high density if the result (number of track points x dwell time) exceeds the density threshold. Adjacent high-density cells are merged to form a continuous block, which is output as a sequence of coordinates of the boundaries of the high-frequency cluster area.

[0075] Spatial gridding refers to the process of dividing a geographic area into equal-sized rectangular cells, with longitude and latitude differences fixed at 0.001° (approximately 100 meters per side). Dwell duration refers to the difference in timestamps between the first and last track points within a single grid. Adjacent track points are those with consecutive timestamps and a time interval less than twice the sensor sampling period.

[0076] The density threshold is dynamically set by the ratio of the total number of trajectory points to the total area of ​​the grid, and its value is proportional to the scale of the trajectory data. A continuous geographic area is defined by the circumscribed polygon boundary of adjacent high-density grids, where the grids share edges or corners.

[0077] For example, when a construction company analyzed dump truck data, it first performed spatial grid division processing on the 21,600 historical trajectory points collected by a single vehicle over 30 days. The vehicle activity area was divided into rectangular grid cells with a side length of 0.001° longitude difference and 0.001° latitude difference, generating a total of 1,200 grid cells. For a specific grid cell, the number of trajectory points it contained was counted as 85. The first trajectory point in this cell was recorded at 08:00:00, and the last trajectory point was recorded at 08:15:30. By calculating the time difference between the first and last points, it was determined that the dwell time of this grid cell was 930 seconds.

[0078] When the system dynamically calculates the density threshold, it takes 5% of the total number of track points (21,600) as a baseline value of 1,080 points. This value is then divided by the total grid area of ​​12 square kilometers, resulting in a density threshold of 90 track points per square kilometer. The density assessment for this grid cell is 79,050 (units uniformly converted to points-seconds per square kilometer), which is the product of 85 track points and a dwell time of 930 seconds. Because this value exceeds the density threshold of 90, the grid cell is marked as a high-density cell. Finally, the system detects a continuous block of six spatially adjacent high-density cells and calculates the total dwell time of all track points within this block to be 62 minutes. This output is a sequence of coordinates of the boundaries of the high-frequency cluster area.

[0079] In one embodiment of the present invention, the dynamic buffer zone is expanded outward based on the boundary of the high-frequency concentration area, including the following steps:

[0080] First, the coordinate sequence of the high-frequency clustering area boundary output by the clustering area identification module is converted into a polygonal geographic range. A calculation point is set every 10 meters along the boundary line, and a circular analysis area with a radius of 20 meters is established with this point as the center. The number of historical driving trajectory points captured by the trajectory data acquisition module within each analysis area is counted, and the ratio of this number to the maximum trajectory point capacity (analysis area × standard density per unit area) is calculated, defining this as the local density ratio. The local density ratio of the current point is obtained, and the scarcity coefficient is calculated by subtracting this density ratio from 1 (the lower the density, the larger the coefficient). Finally, the base buffer value is added to the scarcity coefficient multiplied by the maximum expansion value. For example, if the local density ratio at a point is 0.3, the buffer width is 5 + (1 - 0.3) × 50 = 40 meters. The base buffer value is fixed at 5 meters, and the maximum expansion value is 50 meters.

[0081] Finally, the buffer widths of each calculation point are connected into a smooth curve to generate a non-uniform buffer boundary that wraps the original aggregation area. This boundary and the original aggregation area boundary together constitute a double-layer structure of the elastic electronic fence.

[0082] The high-frequency cluster boundary coordinate sequence refers to the set of longitude and latitude coordinates of the polygon vertices output by the cluster identification module. The local density ratio indicates the density of trajectory points within the local boundary area. This is the ratio of the actual number of trajectory points to the theoretical maximum capacity (ranging from 0 to 1). The basic buffer value is the minimum buffer width, used to ensure safe spacing within the core area. The maximum expansion value is the maximum width to which the buffer can be expanded.

[0083] A non-uniform buffer zone is an area where the buffer width varies dynamically with location. Its boundary is formed by linear interpolation of the buffer widths of adjacent calculation points. The two-layer structure is defined by the boundary of the original aggregation area (inner layer) and the boundary of the non-uniform buffer zone (outer layer).

[0084] For example, in the port container transport area, the polygonal boundary of the loading and unloading area output by the cluster area identification module is obtained. The boundary consists of 12 vertex coordinates. A calculation point is set every 10 meters along the boundary line, generating a total of 86 calculation points. For point P35 (121.5 degrees east longitude, 31.2 degrees north latitude), a circular analysis area with a radius of 20 meters is established with this point as the center. The number of historical trajectory points contained in this area is counted as 42.

[0085] The analysis area is approximately 1,256 square meters. Based on a pre-defined standard density of 0.1 points per square meter, the theoretical maximum carrying capacity is calculated to be 125.6 track points. Dividing the actual number of track points (42) by the maximum carrying capacity of 125.6 yields a local density ratio of approximately 0.33. Using the buffer width calculation formula, the base buffer value of 5 meters is added to the expansion value (1 minus the local density ratio of 0.33, 0.67, multiplied by the maximum expansion value of 50 meters) to obtain a buffer width of 38.5 meters at this point.

[0086] For corner location P68, whose analysis area contains only eight trajectory points and a local density ratio of 0.06, the calculated buffer width is 52 meters (exceeding the maximum expansion limit of 50 meters, so 50 meters is used). The resulting elastic fence maintains a minimum width of 5 meters on the main freight corridor and expands to a maximum width of 50 meters in remote areas, forming a buffer boundary that changes dynamically with location.

[0087] In one embodiment of the present invention, calculating the spatial relationship between the real-time position of the vehicle and the elastic electronic fence includes the following steps:

[0088] The vehicle's current location coordinates (longitude and latitude) uploaded by the on-board terminal are received in real time, and geographic relationship judgment is performed in combination with the elastic electronic fence spatial data generated by the elastic fence generation module. For a single vehicle location point, its distance to the boundary of the outer buffer zone of the elastic electronic fence is calculated: first, the longitude and latitude of the vehicle and the boundary point are converted into plane coordinates x1, y1 and x2, y2; then the sum of the square of the difference between the x coordinates and the square of the y coordinates is calculated; finally, the square root of the square sum is taken to obtain the actual distance. If the distance is less than or equal to zero (that is, the location point is within the buffer zone), it is marked as a warning state; if the distance is less than zero and crosses the inner core boundary at the same time (that is, the location point enters the original gathering area), it is marked as an out-of-bounds state.

[0089] The positional relationship between the position point and the boundary is determined by the ray method:

[0090] Draw a horizontal ray from the location point to the right and count the number of times the vehicle crosses the fence boundary. An odd number of crossings indicates entry into the area. The status is marked and continuously updated until the vehicle leaves the relevant area. The timestamp and coordinates of each status change are recorded.

[0091] Elastic electronic fence spatial data includes the coordinate sequence of the inner core boundary vertices and the outer buffer boundary vertex coordinate sequence defined by the elastic fence generation module. The vehicle's real-time location refers to the longitude and latitude coordinates updated in seconds by the onboard GPS sensor. The data format is consistent with the historical trajectory points of the trajectory data acquisition module.

[0092] Distance calculation uses the spherical distance formula, converting longitude and latitude into Euclidean distance in a planar coordinate system (such as the Web Mercator projection). The ray method is a geometric method that determines point-to-surface relationships based on the odd or even number of times a horizontal line crosses a boundary segment, formed by connecting adjacent fence vertices. A warning state indicates that the vehicle has entered the buffer zone but not the core area, triggering a low-level monitoring alert. An out-of-bounds state indicates that the vehicle has entered the core boundary, triggering a high-level alert.

[0093] For example, in a construction site monitoring scenario, the system acquires the vehicle's coordinates (116.403 degrees east longitude, 39.917 degrees north latitude) in real time. The calculated minimum distance from this point to the outer buffer zone of the elastic geo-fence is -3.2 meters (a negative value indicates it is within the buffer zone), but detection shows that the point has not crossed the inner core boundary.

[0094] Position verification is performed using the ray method: a horizontal ray is drawn from this point to the right, and the number of times it crosses the outer boundary is counted as a single (odd number of times), thus confirming that the vehicle is within the buffer zone. The system then marks it as a warning state, and the monitoring interface displays a yellow triangle. Two minutes later, the vehicle moves to the new coordinates (116.402 degrees east longitude, 39.914 degrees north latitude). The ray method detection shows that the horizontal ray crosses the inner core boundary, and the state is immediately updated to the out-of-bounds state, triggering the sound and light alarm device, and recording the current timestamp and coordinate position.

[0095] In one embodiment of the present invention, establishing a spatiotemporal distribution model of historical boundary crossing behaviors includes the following steps:

[0096] The boundary crossing status data recorded by the boundary crossing status determination module is used to extract all boundary crossing events. Each event includes the boundary crossing location coordinates (longitude and latitude) and the occurrence timestamp. Spatial clustering is performed on this location data: a 50-meter radius analysis circle is established with a single boundary crossing point as the center. If two analysis circles overlap, the boundary crossing points contained in them are merged to form a single boundary crossing section.

[0097] Hourly segmented statistics are generated for time data: the entire day is divided into 24 time periods, and the proportion of boundary crossing incidents in each period to the total number of incidents for the entire day is calculated. If the proportion of a period exceeds a preset threshold (such as 10%), it is marked as a high-incidence period. Finally, the spatial and temporal analysis results are linked to generate a two-dimensional distribution table with road sections as rows and time periods as columns. The cell value in the table represents the proportion of boundary crossing frequency of the road section in that time period. Cells with a proportion exceeding 5% are marked as high-frequency risk points.

[0098] Out-of-bounds status data refers to the set of records marked as "out-of-bounds" by the out-of-bounds status determination module. Each record contains the vehicle ID, longitude and latitude coordinates, and UTC time accurate to the second. Out-of-bounds road sections refer to spatially continuous out-of-bounds areas, defined by the minimum circumscribed polygon formed by merging overlapping analysis circles.

[0099] The high-incidence period is the time period when cross-border incidents occur in a single day, and the threshold is dynamically adjusted according to the fluctuation range of historical data; the two-dimensional distribution table is a data matrix where rows represent road sections and columns represent 24-hour time periods; high-frequency risk points refer to time-space units where the cross-border frequency of a specific road section in a specific time period accounts for more than 5%. The threshold is set by the system administrator.

[0100] For example, the logistics park monitoring system obtained 2,185 records of cross-border events within 30 days through the cross-border status determination module. The coordinate point (121.501 degrees east longitude, 31.202 degrees north latitude) was triggered 42 times in 15 days. A 50-meter radius analysis circle was established with the coordinate as the center, and the adjacent point (121.503 degrees east longitude, 31.201 degrees north latitude) with a center distance of 80 meters was detected. Because the distance between the two circle centers was less than the 100-meter threshold, the analysis circles were merged to generate a rectangular cross-border section covering the two points. Statistics show that 386 cross-border events occurred during the 08:00-09:00 period every day, accounting for 17.7% of the total 2,185 events (exceeding the set threshold of 10%), and were marked as a high-incidence period.

[0101] A spatiotemporal distribution analysis revealed that 112 border crossings occurred on this combined road section between 8:00 AM and 9:00 AM, accounting for 26.5% of the total number of border crossings on this section (exceeding the 5% threshold). The final high-frequency risk point conclusion was: "The East Gate Passage area has a significant risk of border crossings during the morning rush hour between 8:00 AM and 9:00 AM."

[0102] In one embodiment of the present invention, the overlay display of the elastic electronic fence layer, the real-time vehicle location mark layer, and the out-of-bounds thermal layer on the GIS map includes the following steps:

[0103] First, the elastic electronic fence spatial data generated by the elastic fence generation module is used to draw two polygon layers on the GIS map basemap: the inner core boundary is represented by a red dashed line, and the outer dynamic buffer boundary is represented by a yellow solid line, filled with semi-transparent blocks (light red for the core area and light yellow for the buffer area). The real-time vehicle location data stream output by the out-of-bounds status determination module is simultaneously connected to the vehicle location data stream, and dynamic icons are rendered at the corresponding coordinate points: a blue circular icon is displayed for areas not entering the fence, a yellow triangle icon is displayed for warning status, and a red flashing icon is displayed for out-of-bounds status.

[0104] At the same time, the high-frequency risk point data from the behavior analysis module is loaded, and the map is divided into 1 km x 1 km grids. The crossing frequency percentages for all road sections within the current grid are first accumulated. A time period weight coefficient is then matched based on the current time (for example, 1.5 for the morning rush hour). Finally, the accumulated value is multiplied by the weight coefficient. For example, if a grid contains three road sections (with percentages of 12%, 8%, and 5%, respectively), the morning rush hour calculation process is: 12 + 8 + 5 = 25, and 25 × 1.5 = 37.5.

[0105] The thermal layer is rendered using a red-yellow-green gradient color spectrum, with red representing the top 20% of areas with the highest thermal values, and green representing the bottom 40% with the lowest. When superimposing each layer, the layer priority is set (the thermal layer is at the bottom, the fence layer is in the middle, and the vehicle icon is at the top).

[0106] GIS maps refer to loaded online or offline electronic map services. Real-time vehicle location data is formatted in the same format as the trajectory points in the trajectory data acquisition module, with an update frequency of no less than 10 seconds. High-frequency risk points refer to the collection of space-time units marked in the two-dimensional distribution table output by the behavior analysis module.

[0107] The thermal value reflects the overall intensity of the risk of crossing the boundary within a grid cell. It is calculated by multiplying the sum of the crossing frequency percentages across all road sections within the cell by the preset weight for that time period (1.5 for morning peak hours and 0.8 for nighttime hours). The time period weight coefficient is dynamically set based on the high-incidence periods identified by the behavior analysis module, with a baseline value of 1.0 for non-high-incidence periods.

[0108] For example, the sanitation vehicle monitoring center loads a GIS basemap of the city's administrative divisions and draws a flexible electronic fence around the garbage transfer station: a 200-square-meter core area is marked with a red dashed line, while a buffer zone is defined by a solid yellow line 5 to 30 meters wide based on regional characteristics. The system receives real-time location data from 58 operating vehicles, 53 of which display blue circular icons (normal status), four vehicles in the buffer zone display yellow triangular warning icons, and one vehicle entering the core zone displays a flashing red out-of-bounds icon. The behavioral analysis module uses data on high-frequency risk points during the morning rush hour to divide the eastern part of the city into ten 1-square-kilometer grid cells. For grid A05 (containing three out-of-bounds road sections), the percentage of out-of-bounds violations is calculated: Section 1 accounts for 15%, Section 2 accounts for 9%, and Section 3 accounts for 6%. Using a weighting factor of 1.5 for the morning rush hour, the thermal value is calculated by summing the percentages of each section and multiplying by the weight to obtain a thermal value of (15 + 9 + 6) × 1.5 = 45.

[0109] Based on the grid-wide thermal value distribution (highest value 82 to lowest value 7), grid A05 was rendered orange-red by numerical quantile. The administrator observed through the overlay view that the red flashing vehicle crossing the boundary was located within both the grid cell with a thermal value of 45 and the core area of ​​the fence, and immediately initiated an intervention command through the dispatch system.

[0110] It should be noted that the formulas described above can translate physical quantities of different attributes into unitless standard values ​​or superimposable parameters of the same dimension through the principle of dimensional consistency and mathematical standardization (e.g., normalization, dimensionless parameter conversion, or unit system unification). This eliminates the interference of different dimensions on the operational logic, allowing the formulas to retain the distribution characteristics of the original data while maintaining mathematical rationality and adaptability to objective laws. The above are merely exemplary embodiments of the present invention and are not intended to limit the scope of the present invention.

[0111] The modules can be implemented in whole or in part through software, hardware, or a combination thereof, supporting hardware embedded in or independent of a processor in a computer device, and also supporting software stored in a memory in a computer device, so that the processor can call and execute operations corresponding to the modules.

[0112] It should be noted that the human body information (including but not limited to human device information and personal information, etc.) and data (including but not limited to data used for analysis, stored data and displayed data, etc.) involved in the present invention are all information and data authorized by the human body or fully authorized by all parties. The collection, use and processing of relevant data require relevant legal standards.

[0113] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. These modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present invention, and should all be included in the scope of protection of the present invention.

Claims

1. A GPS-based vehicle trajectory intelligent GIS electronic fence visualization system, characterized by: include: The trajectory data acquisition module is used to obtain the historical driving trajectory point data of multiple vehicles within a preset time period. Each driving trajectory point data contains longitude, latitude and timestamp; Aggregation area recognition module, used to identify high-frequency aggregation areas of historical driving trajectory point data; The elastic fence generation module is used to implement a dynamic buffer zone that expands outward based on the boundary of the high-frequency clustering area. The width of the buffer zone increases as the density of trajectory points at the edge of the area decreases. An out-of-bounds status determination module is used to calculate the spatial relationship between the vehicle's real-time position and the elastic electronic fence, marking the vehicle as in a warning state when it enters the buffer zone and as in an out-of-bounds state when it crosses the core boundary of the fence; The behavior analysis module is used to establish a spatiotemporal distribution model of historical boundary crossing behaviors. By correlating the location of boundary crossings with the time of occurrence, it can identify sections of road where repeated boundary crossings occur and the time periods when they are most frequent. The multi-source visualization module is used to overlay and display the elastic electronic fence layer, real-time vehicle location mark layer, and out-of-bounds thermal layer on the GIS map. The out-of-bounds thermal layer uses color concentration to represent the frequency of out-of-bounds in high-incidence areas identified by the behavior analysis module.

2. The GPS-based vehicle trajectory intelligent GIS electronic fence visualization system according to claim 1 is characterized in that: Obtaining historical driving trajectory data for multiple vehicles within a preset time period includes the following steps: Obtain the original trajectory point sequence from the preset vehicle terminal interface; Delete abnormal track points whose longitude and latitude are beyond the preset effective range; Remove drifting track points whose distance between adjacent track points exceeds the preset maximum reachable distance; The processed denoised trajectory dataset is stored in groups according to vehicle ID.

3. The GPS-based vehicle trajectory intelligent GIS electronic fence visualization system according to claim 1 is characterized in that: Identifying high-frequency clustered areas of historical driving trajectory point data includes the following steps: Construct rectangular grid cells with preset longitude and latitude differences; Calculate the product of the number of trajectory points and the dwell time of each grid cell, where the dwell time is the difference between the timestamps of the first and last trajectory points in the grid; The preset density threshold is dynamically set according to the ratio of the total number of trajectory points to the total area of ​​the grid.

4. The GPS-based vehicle trajectory intelligent GIS electronic fence visualization system according to claim 3 is characterized in that: Identifying high-frequency clustered areas of historical driving trajectory point data also includes the following steps: Mark the grids whose product values ​​exceed the preset density threshold as high-density cells; Merge adjacent high-density units to form continuous blocks; When the total stay time in the block exceeds the preset time threshold, the coordinate sequence of the clustering area boundary is output.

5. The GPS-based vehicle trajectory intelligent GIS electronic fence visualization system according to claim 1 is characterized in that: Expanding the dynamic buffer zone outward based on the boundary of the high-frequency cluster area includes the following steps: Set calculation points along the boundary line at preset intervals; A circular analysis area with a preset radius is established with each calculation point as the center; The ratio of the number of trajectory points in the analysis area to the maximum trajectory point carrying capacity is calculated as the local density ratio; Get the local density ratio of the current point, subtract the density ratio from 1 to get the scarcity coefficient, and dynamically determine the width of each point by adding the basic buffer value and the scarcity coefficient multiplied by the maximum expansion value.

6. The GPS-based vehicle trajectory intelligent GIS electronic fence visualization system according to claim 1 is characterized in that: Calculating the spatial relationship between the real-time position of the vehicle and the elastic electronic fence includes the following steps: Convert the vehicle's current position coordinates into plane coordinate system coordinates; The ray method is used to determine the positional relationship between the coordinate point and the inner core boundary; When the coordinate point is within the outer dynamic buffer zone, the warning state is activated; The out-of-bounds state is activated when the coordinate points cross the inner core boundary at the same time.

7. The GPS-based vehicle trajectory intelligent GIS electronic fence visualization system according to claim 1 is characterized in that: Establishing a spatiotemporal distribution model of historical boundary crossing behaviors includes the following steps: Create an analysis circle of out-of-bounds points with a preset radius; Merge the cross-border points contained in the overlapping analysis circles to form the cross-border road section; Divide and calculate the percentage of boundary crossing incidents by preset time periods; When the frequency of crossing the boundary on a specific road section during a specific period exceeds the preset risk threshold, it will be marked as a high-frequency risk point.

8. The GPS-based vehicle track intelligent GIS electronic fence visualization system according to claim 1 is characterized in that: Overlaying the elastic geo-fence layer, real-time vehicle location marker layer, and out-of-bounds thermal layer on the GIS map involves the following steps: Divide map units into preset grid sizes; Calculate the sum of the cross-border frequency ratios of each unit based on the cross-border behavior distribution feature table; Match the preset time period weight coefficient according to the current time period; The heat value is obtained by multiplying the sum of the frequency ratios of crossing the boundary by the time period weight coefficient; Rendering is done by mapping a gradient color spectrum according to the thermal value interval.

9. The GPS-based vehicle trajectory intelligent GIS electronic fence visualization system according to claim 1, characterized in that: When the coordinate points cross the inner core boundary at the same time, the out-of-bounds state is activated, including the following steps: Get the out-of-bounds road section identifier of the current out-of-bounds location; Retrieve the high-incidence time period of the corresponding road section in the cross-border behavior distribution feature table; When the current time falls into the high-incidence period, the warning level is raised and the preset emergency response protocol is triggered.

Images