A roadside parking space management method and system based on Beidou positioning system

Abstract

The present invention discloses a roadside parking space management method and system based on the Beidou positioning system, which relates to the technical field of roadside parking space management and includes the following steps: collecting the current position coordinates of the vehicle through the Beidou positioning system, and combining it with the parking space position information database in the urban road parking management platform to identify the standard parking space number where the vehicle is currently located, and generating an analysis area with the parking space center as the center according to the center coordinates of the parking space and a preset analysis radius; calling the connected urban three-dimensional geographic information platform to retrieve all building boundary objects in the analysis area. The present invention effectively offsets the positioning drift error of Beidou in densely built areas by introducing intelligent prediction of building density and adaptive fuzzy buffering mechanism, combined with secondary verification of vehicle body orientation and inertial navigation trajectory, improves the fault tolerance of parking space determination, significantly reduces the risk of misjudgment and miscontrol, and enhances the accuracy and stability of roadside parking management.

Description

Technical Field

[0001] The present invention relates to the technical field of roadside parking space management, and in particular to a roadside parking space management method and system based on the Beidou positioning system. Background Art

[0002] Roadside parking management based on the Beidou positioning system refers to a management method that uses the high-precision positioning capabilities provided by the Beidou satellite navigation system to conduct real-time monitoring and intelligent scheduling of parking spaces on both sides of urban roads. This method installs Beidou positioning terminals on vehicles or parking spaces to collect real-time information on the vehicle's location, parking status, and duration, and transmits the data to the urban traffic management platform. The platform can combine electronic maps with parking rules to dynamically determine whether the vehicle is parked in a standardized manner, whether it has exceeded the parking time limit, and other situations, thereby realizing intelligent supervision of parking behavior. Compared with traditional methods that rely on manual inspections or geomagnetic induction, the Beidou positioning system has the advantages of all-weather, high precision, and wide coverage. It improves the utilization efficiency of urban parking resources and provides important support for smart transportation and refined urban management.

[0003] The existing technology has the following deficiencies:

[0004] Existing Beidou-based roadside parking management technologies often use a static grid-based location information judgment mechanism to determine whether vehicles are parked properly. However, in densely built-up areas, Beidou positioning systems are prone to sub-meter positioning errors due to factors such as multipath effects and signal obstruction. When multiple vehicles are close together or within the critical boundary of the same parking space, the Beidou positioning system may mistakenly identify vehicles in adjacent parking spaces as "overlapping" within the same parking unit. This type of misjudgment not only triggers erroneous penalties for illegal double parking, but can also, when combined with automated equipment (such as locking devices and lifting systems), result in the forced locking of non-violating vehicles, accidental damage to the vehicle body, or obstruction of normal departure, leading to serious consequences such as property damage and even traffic jams.

[0005] The above information disclosed in this Background section is only for enhancement of understanding of the background of the present disclosure and therefore it may contain information that does not form the prior art that is already known to a person of ordinary skill in the art. Summary of the Invention

[0006] The purpose of the present invention is to provide a roadside parking space management method and system based on the Beidou positioning system. By introducing intelligent prediction of building density based on three-dimensional urban semantic perception before the Beidou static grid judgment logic, and adaptively expanding the fuzzy buffer zone of the parking space grid in high-density scenarios, and superimposing a secondary verification mechanism of vehicle body orientation and inertial navigation trajectory, the Beidou positioning system can dynamically offset the sub-meter positioning drift error caused by the building complex, and achieve high fault tolerance judgment of critical parking space boundary scenes, thereby significantly reducing the "overlapping parking" misjudgment rate and the risk of false triggering of automatic vehicle locking and other equipment, improving the accuracy, safety and user experience of roadside parking management in complex urban environments, and solving the problems in the above-mentioned background technology.

[0007] To achieve the above objectives, the present invention provides the following technical solution: a roadside parking space management method based on the Beidou positioning system, comprising the following steps:

[0008] The vehicle's current location coordinates are collected through the Beidou positioning system and combined with the parking space location information database in the urban road parking management platform to identify the standard parking space number the vehicle is currently in. Based on the center coordinates of the parking space, an analysis area with the center of the parking space as the center is generated according to a preset analysis radius (such as 10 meters or 15 meters);

[0009] Call the connected urban 3D geographic information platform (such as urban BIM data, urban LIDAR mapping data, or GIS building vector map) to retrieve all building boundary objects in the analysis area. For each building covering the analysis area, map it to the coordinate system of the parking space center according to the spatial relationship;

[0010] Graphically fuse the extracted building boundary information to generate a building information map;

[0011] Using feature engineering techniques, we extract key features from architectural infographics that reflect the presence of high-density buildings in the analysis area. After in-depth analysis of these features, we input them into a convolutional neural network, previously trained on a large amount of data, to predict the density of buildings within the analysis area.

[0012] When the analysis area is identified as a densely built-up area, a fuzzy edge buffer zone is extended around each grid, and the width of the fuzzy edge buffer zone is adaptively adjusted according to the density of the building complex. When the vehicle positioning point is within the buffer zone, it is not immediately judged as a straddle. Instead, a secondary verification mechanism is triggered, using vehicle orientation recognition and inertial navigation displacement tracking to determine whether the vehicle is parked reasonably.

[0013] Preferably, the process of identifying the standard parking space number of the vehicle through the Beidou positioning system can be divided into the following key steps: first, the Beidou high-precision positioning terminal installed on the vehicle is used to collect the latitude and longitude coordinates of the vehicle's current position in real time, and upload the position information to the urban road parking management platform; second, the platform calls a pre-built parking space location information database, which records the spatial attributes such as the number, center coordinates, boundary polygons, etc. of all standard parking spaces; then, the Beidou positioning system uses a spatial matching algorithm (such as point-in-polygon judgment, nearest neighbor matching or vector space search) to geometrically compare the vehicle's current position with all parking space boundaries to identify the standard parking space number where the vehicle is located; if the vehicle positioning point falls in the overlapping area of ​​multiple parking space boundaries, the optimization judgment is performed based on the principle of closest distance or historical trajectory directionality, thereby achieving accurate identification of the vehicle's actual parking space.

[0014] Preferably, each building in the analysis area is mapped to the coordinate system of the parking space center according to the spatial relationship, which can be done by the following steps:

[0015] First, determine the geographic coordinates of the center point of the target parking space (such as WGS-84 longitude and latitude or XY coordinates in the city projection coordinate system), and set this point as the origin of the local coordinate system; second, perform coordinate transformation on the coordinate data of the retrieved building boundary objects one by one, and calculate the relative displacement of each building vertex relative to the center point of the parking space, that is, convert the original absolute coordinates into local coordinate representation based on the parking space center through translation transformation; then, if the city data uses different projections or coordinate systems (such as LIDAR data as local coordinates and BIM as building internal coordinates), it is necessary to unify the coordinate reference system first and then perform mapping; finally, a set of building boundary polygon data aligned with the parking space center is formed, which is convenient for subsequent spatial relative position relationship analysis and graphic fusion operations.

[0016] Preferably, the process of graphically fusing the extracted building boundary information to generate a building information map can be carried out in the following steps:

[0017] First, based on the building boundary data mapped to the parking space center coordinate system, the boundary polygons of each building are drawn on a unified two-dimensional plane, and their spatial outlines are expressed using vector graphics formats (such as GeoJSON, Shapefile, or SVG path). Second, all building boundary layers are overlaid and rendered using graphics processing libraries (such as OpenCV, matplotlib, QGIS, or custom rendering engines) to merge them into a complete regional building information map. Different attribute layer information (such as height, material, and use) can be assigned to different buildings to enhance the layer semantics. Subsequently, to improve the analysis accuracy, auxiliary image information such as occlusion areas, boundary continuity, number of enclosed spaces, building density and thermal distribution in the graphics can be further calculated, and this information can be embedded in the building information map in the form of pixel matrices, partitioned vector grids, or raster codes. The final output building information map will be visual and computable, and can be used as an input image for neural network recognition and for subsequent spatial structure feature extraction and analysis.

[0018] Preferably, key features reflecting the presence of high-density buildings in the analysis area are extracted from the building information map using feature engineering techniques, wherein the extracted features include the overlap ratio of adjacent building boundaries on the projection plane and the concentration of direction vectors from polygonal building edges to parking space centers. Within the analysis area, after in-depth analysis of the extracted features, boundary overlap ratio indices and polygon concentration ratio indices are generated, respectively. The boundary overlap ratio indices and polygon concentration ratio indices are used to quantify the building density in the analysis area.

[0019] The analyzed boundary overlap rate index and polygon concentration rate index are used as feature vectors expressing the high-dimensional spatial structure and input into a convolutional neural network that has been trained with a large amount of data in advance. The convolutional neural network conducts an in-depth analysis of the input feature vectors and finally outputs a density index score that comprehensively reflects the density of building distribution in the analysis area. Based on the density index score, an intelligent prediction is made on the density of the building complex in the analysis area.

[0020] Preferably, the density index score generated when predicting the density of the building complex in the analysis area through the convolutional neural network is compared with a pre-set density threshold to identify the building-dense area. The specific identification steps are as follows:

[0021] If the density index score is greater than the preset density threshold, the analysis area is identified as a building-dense area; if the density index score is less than or equal to the preset density threshold, the analysis area is identified as a normal building area.

[0022] Preferably, when the analysis area is identified as a densely built area, a fuzzy edge buffer zone is extended around each grid, and the width of the fuzzy edge buffer zone is adaptively adjusted according to the density of the building group. The specific steps are as follows:

[0023] According to the degree of deviation between the density index score output by the convolutional neural network and the pre-set density threshold, the building density excess ratio of the current analysis area is calculated. Then, the basic expansion radius of the fuzzy edge buffer zone is calculated based on the building density excess ratio. The calculation expression is: ,in: It is the density index score output by the convolutional neural network. The larger the value, the denser the building distribution. is the building density threshold, used to determine whether to enter a high-density area; The building density exceeds the standard ratio, which reflects the degree to which the building density exceeds the building density threshold; The basic buffer zone extension radius represents the default extension width (e.g. 0.5 m) under critical density conditions. is the sensitivity amplification factor (e.g., 1.5 to 3.0), which is used to control the intensity of the extended response; is the nonlinear amplification power exponent (1.2~2.0 is recommended) to enhance the growth trend of the buffer width response in high-density areas;

[0024] Traverse all parking grid cells in the analysis area and perform fuzzy buffer zone expansion on their boundaries to construct a new adaptive fuzzy extended grid area. The geometric expansion rule of the adaptive fuzzy extended grid area is expressed as: ,in: Represents the morphological dilation operation (similar to the concept of structural dilation convolution); For is a two-dimensional dilated structuring element of radius (usually a circle or an eight-connected neighborhood); is the w-th parking grid area; is the fuzzy grid area after the w-th parking grid is expanded;

[0025] In the actual construction process, in order to prevent unnecessary intersections between adjacent raster buffers, a buffer zone boundary compression function is introduced to compress and adjust the overlapping areas. The adjustment expression is: ,in: is the center distance between adjacent grids; and is the buffer compression control coefficient (e.g. , ), used to avoid excessive expansion of high-density areas and resulting in buffer overlap; The actual edge buffer radius of the w-th parking grid after dynamic compression based on the interference relationship of adjacent grids.

[0026] Preferably, within the analysis area, the specific steps of generating a boundary overlap rate index after performing an in-depth analysis on the overlapping ratio of adjacent building boundary lines on the projection plane in the building information diagram are as follows:

[0027] In the analysis area, all adjacent building pairs (Meet the minimum distance between two buildings , is the preset adjacency threshold) to perform boundary overlap detection and identify all overlapping boundary line segments. , for each pair of overlapping boundary segments, obtain their overlapping length, and combine their spatial spacing to calculate the overlapping weight factor of the boundary segment pair. The calculation expression is: ,in: is the length of the overlapping part of the boundary segments; For buildings and The Euclidean distance of the centroid; is a small constant used to suppress the numerical explosion caused by extremely small spatial spacing; Boundary segment pair The angle between the direction vectors of is the directional consistency factor, which encourages parallel overlapping boundary segments to have higher weights. ; is the overlap weight factor of the boundary segment pair;

[0028] The overlap weight factors of all adjacent building pairs that meet the conditions Perform nonlinear fusion processing and calculate the boundary overlap index of the analysis area. The calculation expression is: ,in: is the boundary overlap rate indicator; Used to analyze all buildings in the area that meet the adjacency conditions. Perform traversal; Indicates that in the building Count all the boundary line segments that actually overlap , these boundary segment pairs are recorded as ; is the structural trend amplification coefficient (value range is 0.01 to 0.05) used to control the nonlinear growth rate; is a high-order nonlinear compression function, which makes the output of the boundary overlap rate index stable at , used for nonlinear normalization.

[0029] Preferably, within the analysis area, the specific steps of generating a polygon concentration rate index after performing an in-depth analysis on the concentration of direction vectors pointing from the edges of polygonal buildings to the centers of parking spaces in the building information diagram are as follows:

[0030] In a local coordinate system with the center of the parking space as the origin, edge vectors are extracted for all polygonal buildings within the analysis area. For each boundary segment, its unit direction vector is constructed, and a unit vector is constructed from the midpoint of the boundary segment to the center of the parking space. The degree of directional alignment between the two (i.e., directional consistency) is calculated using the following formula: ,in: is the unit direction vector of the nth boundary segment; The unit direction vector from the midpoint of the nth boundary segment to the center of the parking space; Represents a vector and vector The dot product operation; Indicates absolute value, used to uniformly handle positive and negative directions; The consistency weight value of the direction between the nth boundary line segment and the parking space center is used to quantify whether the boundary line segment points to the parking space center;

[0031] After obtaining the consistency weight values ​​of all boundary segments and the parking space center direction, a normalized entropy evaluation is performed based on the sparsity of the direction alignment weights to construct a polygon concentration index. The constructed expression is: , where: N is the total number of all boundary segments in the analysis area; Represents the energy normalized distribution of each directional consistency weight value, ; Used to normalize the entropy value to interval; The polygon concentration index of the final output is in the range of ,The higher the value, the more concentrated the building direction is towards the center of the parking space.

[0032] A roadside parking space management system based on the Beidou positioning system includes a vehicle location recognition module, an analysis area generation module, a building boundary retrieval and mapping module, a building information map construction module, a building density prediction module, a buffer zone adaptive expansion module, and a secondary verification and determination module;

[0033] The vehicle location recognition module collects the vehicle's current location coordinates through the Beidou positioning system and combines it with the parking space location information database in the urban road parking management platform to identify the standard parking space number the vehicle is currently in.

[0034] An analysis area generation module generates an analysis area with the center of the parking space as the center according to the center coordinates of the parking space and a preset analysis radius;

[0035] The building boundary retrieval and mapping module calls the connected urban 3D geographic information platform to retrieve all building boundary objects in the analysis area. For each building covering the analysis area, it maps it to the coordinate system of the parking space center according to the spatial relationship;

[0036] The building information map construction module graphically integrates the extracted building boundary information to generate a building information map;

[0037] The building density prediction module uses feature engineering technology to extract key features from the building information map that indicate the presence of high-density buildings in the analysis area. After in-depth analysis of the extracted features, the analyzed features are input into a convolutional neural network that has been trained on a large amount of data in advance. The convolutional neural network then predicts the building density within the analysis area.

[0038] The buffer zone adaptive expansion module, when the analysis area is identified as a densely built area, expands the fuzzy edge buffer zone around each grid and adaptively adjusts the width of the fuzzy edge buffer zone according to the density of the building group;

[0039] The secondary verification and judgment module does not immediately judge a vehicle as being parked in a buffer zone when its positioning point is within the buffer zone. Instead, it triggers the secondary verification mechanism, using vehicle orientation recognition and inertial navigation displacement tracking to determine whether the vehicle is parked reasonably.

[0040] In the above technical solution, the technical effects and advantages provided by the present invention are:

[0041] The present invention introduces intelligent prediction of building density based on three-dimensional urban semantic perception before the Beidou static grid judgment logic, and adaptively expands the fuzzy buffer zone of the parking grid in high-density scenarios, and superimposes a secondary verification mechanism of vehicle body orientation and inertial navigation trajectory. This enables the Beidou positioning system to dynamically offset the sub-meter positioning drift error caused by building clusters, and achieve high fault tolerance judgment of critical parking boundary scenarios, thereby significantly reducing the "overlapping parking" misjudgment rate and the risk of false triggering of automatic vehicle locking and other equipment, and improving the accuracy, safety and user experience of roadside parking management in complex urban environments. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, a brief introduction to the drawings required for use in the embodiments will be given below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0043] Figure 1 This is a flow chart of a method for roadside parking space management based on the Beidou positioning system of the present invention.

[0044] Figure 2 This is a module schematic diagram of a roadside parking space management system based on the Beidou positioning system of the present invention. DETAILED DESCRIPTION

[0045] Example embodiments will now be described more fully with reference to the accompanying drawings. However, example embodiments can be implemented in many forms and should not be construed as limited to the examples set forth herein; rather, these example embodiments are provided so that the description of this disclosure will be thorough and complete and will fully convey the concepts of the example embodiments to those skilled in the art.

[0046] The present invention provides Figure 1 A roadside parking space management method based on the Beidou positioning system is shown, comprising the following steps:

[0047] The vehicle's current location coordinates are collected through the Beidou positioning system and combined with the parking space location information database in the urban road parking management platform to identify the standard parking space number the vehicle is currently in. Based on the center coordinates of the parking space, an analysis area with the center of the parking space as the center is generated according to a preset analysis radius (such as 10 meters or 15 meters);

[0048] The process of identifying the standard parking space number of a vehicle through the Beidou positioning system can be divided into the following key steps: First, the Beidou high-precision positioning terminal installed on the vehicle is used to collect the latitude and longitude coordinates of the vehicle's current position in real time, and upload the location information to the urban road parking management platform; second, the platform calls a pre-built parking space location information database, which records the spatial attributes such as the number, center coordinates, boundary polygons, etc. of all standard parking spaces; then, the Beidou positioning system uses spatial matching algorithms (such as point-in-polygon judgment, nearest neighbor matching, or vector space search) to geometrically compare the vehicle's current position with all parking space boundaries to identify the standard parking space number where the vehicle is located; if the vehicle positioning point falls in the overlapping area of ​​multiple parking space boundaries, the optimization judgment is made based on the principle of closest distance or historical trajectory directionality, thereby achieving accurate identification of the vehicle's actual parking space.

[0049] This process provides basic positioning support for subsequent parking behavior judgment and intelligent management.

[0050] This analysis area serves as a spatial window for subsequent spatial environment semantic analysis, ensuring that potential influencing factors, such as obstruction by high-rise buildings and dense street structures, can be fully captured when the vehicle's surrounding environment is complex.

[0051] Call the connected urban 3D geographic information platform (such as urban BIM data, urban LIDAR mapping data, or GIS building vector map) to retrieve all building boundary objects in the analysis area. For each building covering the analysis area, map it to the coordinate system of the parking space center according to the spatial relationship;

[0052] To retrieve building boundary objects within the analysis area, follow these steps:

[0053] First, the Beidou positioning system constructs a spatial query area based on the identified center coordinates of the target parking space and the set analysis radius, and converts the coordinates of the area into a unified geographic spatial reference system (such as WGS-84 or the city's unified GIS projection coordinate system); second, the Beidou positioning system calls the connected urban three-dimensional geographic information platform, such as the urban BIM data platform, LIDAR mapping database or GIS building vector map interface, to perform spatial range query operations (such as GeoBoundingBox query or GeoWithin spatial filtering) to retrieve all building boundary objects within the analysis area.

[0054] Map each building in the analysis area to the coordinate system of the parking space center according to its spatial relationship. This can be done by following the steps below:

[0055] First, determine the geographic coordinates of the center point of the target parking space (such as WGS-84 longitude and latitude or XY coordinates in the city projection coordinate system), and set this point as the origin of the local coordinate system; second, perform coordinate transformation on the coordinate data of the retrieved building boundary objects one by one, and calculate the relative displacement of each building vertex relative to the center point of the parking space, that is, convert the original absolute coordinates into local coordinate representation based on the parking space center through translation transformation; then, if the city data uses different projections or coordinate systems (such as LIDAR data as local coordinates and BIM as building internal coordinates), it is necessary to unify the coordinate reference system first and then perform mapping; finally, a set of building boundary polygon data aligned with the parking space center is formed, which is convenient for subsequent spatial relative position relationship analysis and graphic fusion operations.

[0056] This process ensures the consistency of spatial semantic expression and is the basis for accurately analyzing the relationship between building density and occlusion.

[0057] Graphically fuse the extracted building boundary information to generate a building information map;

[0058] The process of graphically integrating the extracted building boundary information to generate a building information map can be carried out in the following steps:

[0059] First, based on the building boundary data mapped to the parking space center coordinate system, the boundary polygons of each building are drawn on a unified two-dimensional plane, and their spatial outlines are expressed using vector graphics formats (such as GeoJSON, Shapefile, or SVG path). Second, all building boundary layers are overlaid and rendered using graphics processing libraries (such as OpenCV, matplotlib, QGIS, or custom rendering engines) to merge them into a complete regional building information map. Different attribute layer information (such as height, material, and use) can be assigned to different buildings to enhance the layer semantics. Subsequently, to improve the analysis accuracy, auxiliary image information such as occlusion areas, boundary continuity, number of enclosed spaces, building density and thermal distribution in the graphics can be further calculated, and this information can be embedded in the building information map in the form of pixel matrices, partitioned vector grids, or raster codes. The final output building information map will be visual and computable, and can be used as an input image for neural network recognition and for subsequent spatial structure feature extraction and analysis.

[0060] Using feature engineering techniques, we extract key features from architectural infographics that reflect the presence of high-density buildings in the analysis area. After in-depth analysis of these features, we input them into a convolutional neural network, previously trained on a large amount of data, to predict the density of buildings within the analysis area.

[0061] Feature engineering technology is used to extract key features from the building information map that reflect the presence of high-density buildings in the analysis area. The extracted features include the overlap ratio of adjacent building boundaries on the projection plane and the concentration of direction vectors from polygonal building edges to parking space centers. Within the analysis area, after in-depth analysis of the extracted features, boundary overlap rate indicators and polygon concentration rate indicators are generated, respectively. The boundary overlap rate indicators and polygon concentration rate indicators are used to quantify the building density in the analysis area.

[0062] Within the analysis area, a high overlap ratio of adjacent building boundaries on the projected plane in the building infographic generally indicates a high building density. This is because in high-density building environments, buildings are often closely spaced along streets, with walls close together or even sharing structural boundaries, resulting in their outlines on the two-dimensional projected plane partially overlapping or fitting together with minimal spacing. When multiple building boundaries exhibit a high spatial overlap ratio, it reflects physical structural characteristics such as small gaps between buildings, a crowded layout, and a lack of open space. This geometric overlap relationship can serve as a direct quantitative indicator of spatial compression. It is a relatively stable, highly sensitive feature that is easy to extract at the graphical level. Therefore, a higher boundary overlap ratio indicates a high building density in the area, which is valuable for identifying "high-occlusion areas" or "misidentified high-risk areas" in cities.

[0063] The specific steps for generating the boundary overlap ratio index after in-depth analysis of the overlap ratio of adjacent building boundary lines on the projection plane in the building information diagram within the analysis area are as follows:

[0064] In the analysis area, all adjacent building pairs (Meet the minimum distance between two buildings , is the preset adjacency threshold) to perform boundary overlap detection and identify all overlapping boundary line segments. , for each pair of overlapping boundary segments, obtain their overlapping length, and combine their spatial spacing to calculate the overlapping weight factor of the boundary segment pair. The calculation expression is: ,in: is the length of the overlapping part of the boundary segments; For buildings and The Euclidean distance of the centroid; is a small constant used to suppress the numerical explosion caused by extremely small spatial spacing; Boundary segment pair The angle between the direction vectors of is the directional consistency factor, which encourages parallel overlapping boundary segments to have higher weights. ; is the overlap weight factor of the boundary segment pair;

[0065] For each pair of overlapping boundary line segments, the overlapping length can be obtained by combining projection coincidence detection with line segment clipping calculation. Specifically, first determine whether the two boundary line segments are collinear or approximately collinear in the two-dimensional plane. The judgment condition can be that the angle between the direction vectors is less than a set threshold (such as 5°); if it is satisfied, the two line segments are projected onto a one-dimensional axis (such as the main axis direction) in the same direction, and the length of the overlapping interval is calculated by comparing the start and end projection intervals of the two line segments on the axis; this length is the spatial projection distance of the actual overlapping part. This method is not only applicable to strictly collinear situations, but can also tolerate "approximately overlapping" structures with slight offsets. It has high robustness and is easy to embed into a batch calculation framework in graphics processing or spatial analysis. It is an efficient means for identifying building fit relationships.

[0066] This step comprehensively models the length, directional consistency and spatial proximity of overlapping line segments to generate an overlapping weight factor that represents the "local spatial fit" , providing a physical representation of the boundary structure for subsequent dense measurement.

[0067] The overlap weight factors of all adjacent building pairs that meet the conditions Perform nonlinear fusion processing and calculate the boundary overlap index of the analysis area. The calculation expression is: ,in: is the boundary overlap rate indicator; Used to analyze all buildings in the area that meet the adjacency conditions. Perform traversal; Indicates that in the building Count all the boundary line segments that actually overlap , these boundary segment pairs are recorded as ; is the structural trend amplification coefficient (value range is 0.01 to 0.05) used to control the nonlinear growth rate; is a high-order nonlinear compression function, which makes the output of the boundary overlap rate index stable at , used for nonlinear normalization;

[0068] This step performs global weighted superposition of local overlapping relationships and uses a nonlinear suppression function for normalization, so that the boundary overlap rate index It not only reflects the degree of building fit, but also produces a response amplification effect on extremely dense structures.

[0069] The boundary overlap rate indicator shows that within the analysis area, the larger the boundary overlap rate indicator, which is generated after an in-depth analysis of the overlap ratio of adjacent building boundary lines on the projection plane in the building information map, the higher the building density in the analysis area, and vice versa. The reason is that in high-density building areas, buildings are often arranged compactly, with walls close to each other and even showing spatial relationships such as shared boundaries and parallel edges, resulting in a large number of overlapping or near-overlapping boundary segments in the two-dimensional projection map. The boundary overlap rate indicator accurately captures the structural characteristics of this "spatial fit" by integrating multiple factors such as overlap length, building spacing, and boundary direction consistency. Therefore, the larger its value, the more crowded the building distribution, the smaller the spacing, and the more consistent the directional arrangement, which is a direct structural response to the high density of the building cluster. Conversely, in areas with sparse building distribution and distant boundary lines, the overlap rate is significantly reduced, and the boundary overlap rate indicator naturally decreases, thus reflecting the low-density spatial characteristics of the area.

[0070] If the edges or normal vectors of polygonal buildings in the building infographics within the analysis area are concentrated toward the center of parking spaces, this indicates a degree of spatial convergence, convergence, or arraying of these buildings. This phenomenon often occurs in densely built urban blocks or gated communities. Limited spatial resources in high-density building complexes drive planning to enclose, edge, or inward-facing certain core areas, resulting in a convergence of building outline edge directions toward the center. This clustering characteristic can be quantified by calculating the concentration of these direction vectors (e.g., using mean cosine similarity or vector concentration coefficient). Higher values ​​generally indicate a more compact spatial distribution of buildings, more continuous boundaries, and narrower gaps, reflecting a higher building density in the area. Therefore, the concentration of edge direction vectors is a semantically meaningful indicator of spatial structure that can effectively identify local high-density building complexes.

[0071] The specific steps for generating the polygon concentration ratio index by performing an in-depth analysis of the concentration of direction vectors pointing from the edges of polygonal buildings to the centers of parking spaces in the building information map within the analysis area are as follows:

[0072] In a local coordinate system with the center of the parking space as the origin, edge vectors are extracted for all polygonal buildings within the analysis area. For each boundary segment, its unit direction vector is constructed, and a unit vector is constructed from the midpoint of the boundary segment to the center of the parking space. The degree of directional alignment between the two (i.e., directional consistency) is calculated using the following formula: ,in: is the unit direction vector of the nth boundary segment; The unit direction vector from the midpoint of the nth boundary segment to the center of the parking space; Represents a vector and vector The dot product operation; Indicates absolute value, used to uniformly handle positive and negative directions; The consistency weight value of the direction between the nth boundary line segment and the parking space center is used to quantify whether the boundary line segment points to the parking space center;

[0073] In this step, the consistency weight value of all boundary segments and the parking space center direction The value range is , indicating the change from “completely vertical (0)” to “completely pointing to the center (1)”. This consistency weight value essentially reflects whether the building edge is facing the center of the parking space as a whole, and is the weight basis for constructing the core feature of aggregation.

[0074] After obtaining the consistency weight values ​​of all boundary segments and the parking space center direction, a normalized entropy evaluation is performed based on the sparsity of the direction alignment weights to construct a polygon concentration index. The constructed expression is: , where: N is the total number of all boundary segments in the analysis area; Represents the energy normalized distribution of each directional consistency weight value, ; Used to normalize the entropy value to interval; The polygon concentration index of the final output is in the range of ,The higher the value, the more the building direction is concentrated on the center of the parking space;

[0075] The spatial direction aggregation effect is modeled by entropy, and the response to high alignment boundaries is strengthened by square normalization to eliminate the disturbance of a small number of extreme values. A value close to 1 indicates high concentration, which is suitable for identifying "ring-shaped" building layouts or high-density enclosed blocks.

[0076] The polygon concentration ratio index shows that within the analysis area, a higher value, generated by in-depth analysis of the concentration of directional vectors pointing from polygonal building edges to parking space centers in the building infographic, indicates a higher building density within the analysis area; conversely, a lower value indicates a lower building density within the analysis area. This is because the polygon concentration ratio measures spatial aggregation by calculating the degree of consistency between the directional vectors of building edge segments and the direction of parking space centers. A higher value indicates that the building edges within the area tend to surround, converge, or point toward the parking space center, reflecting a distinct "spatial crowding" or "enclosed compression" layout characteristic. This characteristic is often found in densely packed blocks, enclosed communities, or continuous street-facing building clusters, representing high-density development spatial forms. Therefore, a higher polygon concentration ratio index indicates a stronger structural concentration and directional aggregation of buildings within a limited space, indirectly reflecting a higher building density.

[0077] The analyzed boundary overlap rate index and polygon concentration rate index are used as feature vectors expressing the high-dimensional spatial structure and input into a convolutional neural network that has been trained with a large amount of data in advance. The convolutional neural network conducts an in-depth analysis of the input feature vectors and finally outputs a density index score that comprehensively reflects the density of building distribution in the analysis area. Based on the density index score, an intelligent prediction is made on the density of the building complex in the analysis area.

[0078] A convolutional neural network trained with extensive data refers to a method used before formal deployment for intelligent building density recognition. Developers pre-train a specially designed convolutional neural network (CNN) model using supervised learning methods based on existing large-scale urban 3D spatial data, building distribution samples, and known density labels. This allows the model to automatically identify boundary features and building distribution patterns. The training input consists of boundary overlap and polygon concentration ratios extracted from a large number of real urban areas. These metrics are fed into the network as structural feature vectors in the form of a two-dimensional matrix or image. The output is a building density score or classification label (e.g., sparse, medium, dense) corresponding to the area. During training, the neural network uses convolutional layers to extract local spatial patterns and trends in the features, such as overlapping clusters of building boundaries and the direction of changes in outline concentration. Multiple hidden layers then continuously adjust weights to minimize the discrepancy between the output score and the manually labeled density, thereby forming a generalizable building density pattern recognition model.

[0079] Specifically, during the training phase, a convolutional neural network, fed with a large amount of labeled data, learns how to extract spatial structural features highly correlated with building density from metrics such as boundary overlap and polygon concentration. The boundary overlap reflects the degree of edge overlap between buildings, revealing whether buildings exhibit a spatial trend of "seamless" or "closely spaced" layouts. The polygon concentration measures the geometric concentration of building boundaries toward parking spaces, reflecting whether buildings are clustered around a functional core. The neural network performs sliding convolutions on the input feature matrix using multiple sets of convolutional kernels, extracting edge distribution patterns and spatial density patterns at different scales. It then applies nonlinear transformations using activation functions (such as ReLU) to enhance the model's ability to model the relationship between local variations and global structure. Subsequently, a pooling layer further reduces the dimensionality of the patterns, retaining the most representative structural information and reducing model complexity and overfitting risk. Through repeated forward and backward error propagation, the network gradually adjusts its internal parameters, enabling it to effectively identify building density. After training, the model is "frozen," fixing its structure and parameters, allowing it to be used directly for rapid prediction in real-world applications.

[0080] During the application phase, the trained CNN model uses boundary overlap and polygon concentration ratios extracted from the target analysis area as input feature vectors for in-depth analysis. These feature vectors can be constructed into a two-dimensional spatial grid, heat map, or vector stacking diagram to preserve the spatial correlation and geometric structure of the area. Based on pre-trained parameters, the model automatically identifies the input data and outputs a density index score. This score is a comprehensive assessment of the density of buildings in the area, typically ranging from 0 to 1. Higher values ​​indicate a denser building layout, which is more likely to cause Beidou positioning errors and cross-location misjudgments.

[0081] The density index score generated by the convolutional neural network when predicting the building density in the analysis area is compared with the pre-set density threshold to identify the building-dense areas. The specific identification steps are as follows:

[0082] If the density index score is greater than the preset density threshold, the analysis area is identified as a building-dense area; if the density index score is less than or equal to the preset density threshold, the analysis area is identified as a normal building area.

[0083] When the analysis area is identified as a densely built-up area, a fuzzy edge buffer zone is extended around each grid cell, and its width is adaptively adjusted based on the density of the building cluster. When the vehicle positioning point is within the buffer zone, it is not immediately judged as a straddle. Instead, a secondary verification mechanism is triggered, using vehicle orientation recognition and inertial navigation displacement tracking to determine whether the vehicle is parked properly.

[0084] When the analysis area is identified as a densely built area, a fuzzy edge buffer zone is extended around each grid, and the width of the fuzzy edge buffer zone is adaptively adjusted according to the density of the building group. The specific steps are as follows:

[0085] According to the degree of deviation between the density index score output by the convolutional neural network and the pre-set density threshold, the building density excess ratio of the current analysis area is calculated. Then, the basic expansion radius of the fuzzy edge buffer zone is calculated based on the building density excess ratio. The calculation expression is: ,in: It is the density index score output by the convolutional neural network. The larger the value, the denser the building distribution. is the building density threshold, used to determine whether to enter a high-density area; The building density exceeds the standard ratio, which reflects the degree to which the building density exceeds the building density threshold; The basic buffer zone extension radius represents the default extension width (e.g. 0.5 m) under critical density conditions. is the sensitivity amplification factor (e.g., 1.5 to 3.0), which is used to control the intensity of the extended response; is the nonlinear amplification power exponent (1.2~2.0 is recommended) to enhance the growth trend of the buffer width response in high-density areas;

[0086] This step converts the difference between the building density score and the set density threshold into the spatial expansion intensity of the fuzzy buffer, realizing a nonlinear adaptive control mechanism of "the higher the density, the wider the buffer", thereby improving the robustness of the Beidou positioning system in high-risk misjudgment areas.

[0087] Traverse all parking grid cells in the analysis area and perform fuzzy buffer zone expansion on their boundaries to construct a new adaptive fuzzy extended grid area. The geometric expansion rule of the adaptive fuzzy extended grid area is expressed as: ,in: Represents the morphological dilation operation (similar to the concept of structural dilation convolution); For is a two-dimensional dilated structuring element of radius (usually a circle or an eight-connected neighborhood); is the w-th parking grid area; is the fuzzy grid area after the w-th parking grid is expanded;

[0088] In the actual construction process, in order to prevent unnecessary intersections between adjacent raster buffers, a buffer zone boundary compression function is introduced to compress and adjust the overlapping areas. The adjustment expression is: ,in: is the center distance between adjacent grids; and is the buffer compression control coefficient (e.g. , ), used to avoid excessive expansion of high-density areas and resulting in buffer overlap; The actual edge buffer radius of the w-th parking grid after dynamic compression based on the interference relationship of adjacent grids;

[0089] This step spatially expands each parking grid area in a structured manner and adaptively adjusts the expansion radius based on the density relationship between adjacent parking spaces. This not only maintains the error tolerance for high-risk areas, but also avoids spatial misidentification interference between grids, thereby improving the stability and discrimination accuracy of the Beidou positioning system's spatial resolution.

[0090] When a vehicle's location is within the fuzzy edge buffer zone, the Beidou positioning system doesn't directly determine it as "overrunning" or "illegally parked." Instead, it triggers a secondary verification mechanism to improve accuracy and error tolerance. First, the Beidou positioning system uses data from the vehicle's IMU (Inertial Measurement Unit) sensor to extract the vehicle's current orientation (such as yaw angle or angle relative to geographic north). This orientation is then compared with the standard parking orientation preset for the parking space. If the angle between the two is within a reasonable range (e.g., less than 15°), the vehicle is considered parked in compliance with regulations. Second, the Beidou positioning system simultaneously uses the inertial navigation tracking unit to analyze the vehicle's continuous movement trajectory from entering the current area to its resting position, determining whether it exhibits a reasonable "lane-to-pull-to-park" behavior pattern. If the trajectory shows the vehicle entering the target space smoothly at a low speed without deviating from the main direction of the target space, it further supports compliance. By combining vehicle orientation and trajectory characteristics, the Beidou positioning system effectively eliminates misjudgments caused by positioning drift, ensuring more precise and intelligent judgment of parking behavior in critical locations within densely built-up areas.

[0091] By introducing intelligent prediction of building density based on three-dimensional urban semantic perception before the Beidou static grid judgment logic, and adaptively expanding the fuzzy buffer zone of the parking grid in high-density scenarios, and superimposing the secondary verification mechanism of vehicle body orientation and inertial navigation trajectory, the Beidou positioning system can dynamically offset the sub-meter positioning drift error caused by building complexes, and achieve high-tolerance judgment of critical parking boundary scenarios, thereby significantly reducing the misjudgment rate of "overlapping parking" and the risk of false triggering of automatic vehicle locking and other equipment, and improving the accuracy, safety and user experience of roadside parking management in complex urban environments.

[0092] The present invention provides Figure 2 The illustrated roadside parking space management system based on the Beidou positioning system includes a vehicle position recognition module, an analysis area generation module, a building boundary retrieval and mapping module, a building information map construction module, a building density prediction module, a buffer zone adaptive expansion module, and a secondary verification and determination module;

[0093] The vehicle location recognition module collects the vehicle's current location coordinates through the Beidou positioning system and combines it with the parking space location information database in the urban road parking management platform to identify the standard parking space number the vehicle is currently in.

[0094] An analysis area generation module generates an analysis area with the center of the parking space as the center according to the center coordinates of the parking space and a preset analysis radius;

[0095] The building boundary retrieval and mapping module calls the connected urban 3D geographic information platform to retrieve all building boundary objects in the analysis area. For each building covering the analysis area, it maps it to the coordinate system of the parking space center according to the spatial relationship;

[0096] The building information map construction module graphically integrates the extracted building boundary information to generate a building information map;

[0097] The building density prediction module uses feature engineering technology to extract key features from the building information map that indicate the presence of high-density buildings in the analysis area. After in-depth analysis of the extracted features, the analyzed features are input into a convolutional neural network that has been trained on a large amount of data in advance. The convolutional neural network then predicts the building density within the analysis area.

[0098] The buffer zone adaptive expansion module, when the analysis area is identified as a densely built area, expands the fuzzy edge buffer zone around each grid and adaptively adjusts the width of the fuzzy edge buffer zone according to the density of the building group;

[0099] The secondary verification and judgment module does not immediately judge a vehicle as being parked in a buffer zone when its positioning point is within the buffer zone. Instead, it triggers the secondary verification mechanism, using vehicle orientation recognition and inertial navigation displacement tracking to determine whether the vehicle is parked reasonably.

[0100] An embodiment of the present invention provides a roadside parking space management method based on the Beidou positioning system, which is implemented through the above-mentioned roadside parking space management system based on the Beidou positioning system. The specific method and process of a roadside parking space management system based on the Beidou positioning system are detailed in the embodiment of the above-mentioned roadside parking space management method based on the Beidou positioning system, which will not be repeated here.

[0101] The above formulas are all dimensionless and numerical calculations. The formulas are obtained by collecting a large amount of data and performing software simulation to obtain the most recent real situation. The preset parameters in the formulas are set by technicians in this field according to actual conditions.

[0102] The above description is merely illustrative of certain exemplary embodiments of the present invention. It goes without saying that those skilled in the art will be able to modify the described embodiments in various ways without departing from the spirit and scope of the present invention. Therefore, the above drawings and description are illustrative in nature and should not be construed as limiting the scope of protection of the claims.

[0103] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of this application. Therefore, the scope of protection of this application should be based on the scope of protection of the claims.

Claims

1. A roadside parking space management method based on the Beidou positioning system, characterized in that: The following steps are involved: The vehicle's current location coordinates are collected through the Beidou positioning system and combined with the parking space location information database in the urban road parking management platform to identify the standard parking space number the vehicle is currently in. Based on the center coordinates of the parking space, an analysis area with the center of the parking space as the center is generated according to the preset analysis radius; The connected urban 3D geographic information platform is called to retrieve all building boundary objects in the analysis area. For each building covering the analysis area, it is spatially mapped to the coordinate system of the parking space center. The extracted building boundary information is graphically fused to generate a building information map. Feature engineering techniques are then used to extract key features from the building information map that reflect the presence of high-density building clusters in the analysis area. After in-depth analysis of the extracted features, these features are fed into a convolutional neural network, previously trained with a large amount of data, to predict the density of building clusters within the analysis area. When the analysis area is identified as a densely built-up area, a fuzzy edge buffer zone is extended around each grid, and the width of the fuzzy edge buffer zone is adaptively adjusted according to the density of the building complex. When the vehicle positioning point is within the buffer zone, it is not immediately judged as a straddle. Instead, a secondary verification mechanism is triggered, using vehicle orientation recognition and inertial navigation displacement tracking to determine whether the vehicle is parked reasonably.

2. A roadside parking space management method based on Beidou positioning system according to claim 1, characterized in that: Use the Beidou positioning system to identify the standard parking space number where the vehicle is located, and take the following steps: The Beidou high-precision positioning terminal installed on the vehicle collects the latitude and longitude coordinates of the vehicle's current location in real time, and uploads the location information to the urban road parking management platform; Calling a pre-built parking space location information database; Based on the spatial matching algorithm, the vehicle's current position is geometrically compared with the boundaries of all parking spaces to identify the standard parking space number where the vehicle is located. If the vehicle's positioning point falls in the overlapping area of ​​multiple parking space boundaries, the optimal judgment is made based on the principle of closest distance to accurately identify the actual parking space where the vehicle is parked.

3. The method for managing roadside parking spaces based on the Beidou positioning system according to claim 1, characterized in that: Map each building in the analysis area to the coordinate system of the parking space center according to its spatial relationship. This can be achieved by the following steps: Determine the geographic coordinates of the center point of the target parking space and set this point as the origin of the local coordinate system; Perform coordinate transformation on the coordinate data of the retrieved building boundary objects one by one, and calculate the relative displacement of each building vertex relative to the parking space center point, that is, convert the original absolute coordinates into local coordinate representation based on the parking space center through translation transformation; First unify the coordinate reference system, then perform mapping, and finally form a set of building boundary polygon data aligned with the center of the parking space.

4. The method for managing roadside parking spaces based on the Beidou positioning system according to claim 1, characterized in that: The process of graphically integrating the extracted building boundary information to generate a building information map is achieved through the following steps: Based on the building boundary data mapped to the parking space center coordinate system, the boundary polygons of each building are drawn on a unified two-dimensional plane, and the spatial outline is expressed in vector graphics format; Using the graphics processing library, all building boundary layers are overlaid and rendered to form a complete regional building information map. Different attribute layer information is assigned to different buildings to enhance the layer semantics. Calculate the auxiliary image information in the graphics and embed the auxiliary image information into the building information map in the form of pixel matrix, partitioned vector grid or raster encoding.

5. The method for managing roadside parking spaces based on the Beidou positioning system according to claim 1, characterized in that: Feature engineering techniques are used to extract key features from the building information map that reflect the presence of high-density buildings in the analysis area. These features include the overlap ratio of adjacent building boundaries on the projection plane and the concentration of direction vectors pointing from polygonal building edges to parking space centers. Within the analysis area, these extracted features are then deeply analyzed to generate boundary overlap ratio and polygon concentration ratio indices, which are used to quantify the building density in the analysis area. The analyzed boundary overlap rate index and polygon concentration rate index are used as feature vectors expressing the high-dimensional spatial structure and input into a convolutional neural network that has been trained with a large amount of data in advance. The convolutional neural network conducts an in-depth analysis of the input feature vectors and finally outputs a density index score that comprehensively reflects the density of building distribution in the analysis area. Based on the density index score, an intelligent prediction is made on the density of the building complex in the analysis area.

6. The method for managing roadside parking spaces based on the Beidou positioning system according to claim 5, characterized in that: The density index score generated by the convolutional neural network when predicting the building density in the analysis area is compared with the pre-set density threshold to identify the building-dense areas. The specific identification steps are as follows: If the density index score is greater than the preset density threshold, the analysis area is identified as a building-dense area; if the density index score is less than or equal to the preset density threshold, the analysis area is identified as a normal building area.

7. The method for managing roadside parking spaces based on the Beidou positioning system according to claim 6, characterized in that: When the analysis area is identified as a densely built area, a fuzzy edge buffer zone is extended around each grid, and the width of the fuzzy edge buffer zone is adaptively adjusted according to the density of the building group. The specific steps are as follows: According to the degree of deviation between the density index score output by the convolutional neural network and the pre-set density threshold, the building density exceeding standard ratio of the current analysis area is calculated. Based on the building density exceeding standard ratio, the basic expansion radius of the fuzzy edge buffer zone is calculated. The calculation expression is: ,in: The dense index score value output by the convolutional neural network; is the building density threshold; The building density exceeds the standard ratio, which reflects the degree to which the building density exceeds the building density threshold; Expand the radius of the basic buffer zone; is the sensitivity amplification factor, which is used to control the intensity of the extended response; It is a nonlinear amplification power exponent that enhances the growth trend of buffer width response in high-density areas; Traverse all parking grid cells in the analysis area and perform fuzzy buffer zone expansion on their boundaries to construct a new adaptive fuzzy extended grid area. The geometric expansion rule of the adaptive fuzzy extended grid area is expressed as: ,in: represents the morphological dilation operation; For is a two-dimensional dilated structural element with radius; is the w-th parking grid area; is the fuzzy grid area after the w-th parking grid is expanded; The buffer zone boundary compression function is introduced to compress and adjust the overlapping area. The adjustment expression is: ,in: is the center distance between adjacent grids; and To buffer the compression control coefficient and avoid excessive expansion of high-density areas resulting in buffer overlap; The actual edge buffer radius of the w-th parking grid after dynamic compression based on the interference relationship of adjacent grids.

8. The method for managing roadside parking spaces based on the Beidou positioning system according to claim 5, characterized in that: The specific steps for generating the boundary overlap ratio index after in-depth analysis of the overlap ratio of adjacent building boundary lines on the projection plane in the building information diagram within the analysis area are as follows: In the analysis area, all adjacent building pairs Perform boundary overlap detection to identify all overlapping boundary line segments. , for each pair of overlapping boundary segments, obtain their overlapping length, and combine their spatial spacing to calculate the overlapping weight factor of the boundary segment pair. The calculation expression is: ,in: is the length of the overlapping part of the boundary segments; For buildings and The Euclidean distance of the centroid; is a small constant; Boundary segment pair The angle between the direction vectors of is the directional consistency factor, ; is the overlap weight factor of the boundary segment pair; The overlap weight factors of all adjacent building pairs that meet the conditions Perform nonlinear fusion processing and calculate the boundary overlap index of the analysis area. The calculation expression is: ,in: is the boundary overlap rate indicator; Used to analyze all buildings in the area that meet the adjacency conditions. Perform traversal; Indicates that in the building Count all the boundary line segments that actually overlap ; is the structural trend amplification coefficient used to control the nonlinear growth rate.

9. The method for managing roadside parking spaces based on the Beidou positioning system according to claim 5, characterized in that: The specific steps for generating the polygon concentration ratio index by performing an in-depth analysis of the concentration of direction vectors pointing from the edges of polygonal buildings to the centers of parking spaces in the building information map within the analysis area are as follows: In the local coordinate system with the center of the parking space as the origin, edge vectors are extracted for all polygonal buildings in the analysis area. For each boundary segment, its unit direction vector is constructed, and a unit vector is constructed from the midpoint of the boundary segment to the center of the parking space. The degree of alignment between the two directions is calculated using the following formula: ,in: is the unit direction vector of the nth boundary segment; The unit direction vector from the midpoint of the nth boundary segment to the center of the parking space; Represents a vector and vector The dot product operation; Indicates the consistency weight value of the direction between the nth boundary segment and the parking space center; After obtaining the consistency weight values ​​of all boundary segments and the parking space center direction, a normalized entropy evaluation is performed based on the sparsity of the direction alignment weights to construct a polygon concentration index. The constructed expression is: , where: N is the total number of all boundary segments in the analysis area; Represents the energy normalized distribution of each directional consistency weight value, ; It is the polygon concentration index of the final output.

10. A roadside parking space management system based on the Beidou positioning system, used to implement the roadside parking space management method based on the Beidou positioning system according to any one of claims 1 to 9, characterized in that: It includes a vehicle location recognition module, an analysis area generation module, a building boundary retrieval and mapping module, a building information map construction module, a building density prediction module, a buffer zone adaptive expansion module, and a secondary verification and judgment module; The vehicle location recognition module collects the vehicle's current location coordinates through the Beidou positioning system and combines it with the parking space location information database in the urban road parking management platform to identify the standard parking space number the vehicle is currently in. An analysis area generation module generates an analysis area with the center of the parking space as the center according to the center coordinates of the parking space and a preset analysis radius; The building boundary retrieval and mapping module calls the connected urban 3D geographic information platform to retrieve all building boundary objects in the analysis area. For each building covering the analysis area, it maps it to the coordinate system of the parking space center according to the spatial relationship; The building information map construction module graphically integrates the extracted building boundary information to generate a building information map; The building density prediction module uses feature engineering technology to extract key features from the building information map that indicate the presence of high-density buildings in the analysis area. After in-depth analysis of the extracted features, the analyzed features are input into a convolutional neural network that has been trained on a large amount of data in advance. The convolutional neural network then predicts the building density within the analysis area. The buffer zone adaptive expansion module, when the analysis area is identified as a densely built area, expands the fuzzy edge buffer zone around each grid and adaptively adjusts the width of the fuzzy edge buffer zone according to the density of the building group; The secondary verification and judgment module does not immediately judge a vehicle as being parked in a buffer zone when its positioning point is within the buffer zone. Instead, it triggers the secondary verification mechanism, using vehicle orientation recognition and inertial navigation displacement tracking to determine whether the vehicle is parked reasonably.

Images