Road disease distribution visualization method and system

Abstract

The invention relates to the technical field of disease identification, in particular to a road disease distribution visualization method and system, and the method comprises the following steps: obtaining a road inspection image of a cross section structure layer, extracting edge points, recognizing direction offset and curvature mutation features, constructing a structure layer buffer graph, matching disease marks, and marking structure information. And calibrating passing attributes and adjusting positioning coordinates, constructing a coverage layer structure, counting multi-cycle disease changes and generating a path layer. According to the method, a disease distribution task of a multi-structure combined road is processed, a boundary buffer graph is constructed based on an edge direction vector and a structure cross-section coordinate interval, and the trans-boundary characteristic of a disease boundary between structure layers is dynamically identified in combination with the position of a curvature mutation point; the influence of structural dislocation, annotation distortion and periodic noise interference on layer expression is avoided, and graphic expression and trend evolution description of a road disease distribution result under a complex structure road are supported.

Description

Technical Field

[0001] This invention relates to the field of road defect identification technology, and in particular to a method and system for visualizing the distribution of road defects. Background Technology

[0002] The field of defect identification technology involves the automatic detection, classification, and recording of various defects appearing on the surfaces of infrastructure such as roads and bridges. The core aspects include multiple stages such as defect image acquisition, feature extraction, type determination, and information labeling. Relying on image recognition, computer vision, and pattern recognition, it enables the identification and classification management of different types of defects such as cracks, potholes, bulges, and subsidence. It is widely used in scenarios such as highway inspection and urban road maintenance. This technology systematically covers defect detection model construction, image preprocessing procedures, feature classification algorithms, detection accuracy improvement strategies, and visualization of detection results. Traditional road defect distribution visualization methods refer to a type of technical task that graphically presents the distribution of defects in road space after defect detection and identification. The main technical task is to spatially map and display different defect types and their corresponding locations based on the obtained defect detection results to assist subsequent management decisions. Traditional methods include using geographic information systems to overlay defect locations as icons on a two-dimensional map, locating linear paths based on vehicle driving trajectories and defect occurrence time records, distinguishing defect types using different graphic symbols, and providing detailed defect information using labeled tables or defect image thumbnails.

[0003] Existing technologies rely heavily on two-dimensional icon overlay and linear path positioning in the representation of disease distribution. The generation of illustrations is mainly based on fixed coordinate mapping, lacking a response mechanism to the spatial variation of disease boundaries in the context of structural layering. In cases involving overlapping structural layers or blurred disease edges, it is impossible to achieve floating correction of positioning boundaries, resulting in problems such as chaotic structural coverage and ambiguous disease attribution in the layer display. At the same time, no path recursion judgment logic has been established for the intersection of main paths and areas where repeated diseases occur, making it difficult to accurately express the trend of disease distribution changes over time. In multi-cycle inspection tasks, this can cause deviations in the judgment of disease cluster paths, affecting subsequent layer interpretation and inspection planning and deployment. Summary of the Invention

[0004] The purpose of this invention is to address the shortcomings of existing technologies by proposing a method and system for visualizing the distribution of road defects.

[0005] To achieve the above objectives, the present invention adopts the following technical solution: a method for visualizing the distribution of road defects, comprising the following steps: S1: Obtain road inspection images of cross-sectional structure layers, perform contour detection on the diseased areas, extract the set of edge point coordinates, and perform direction vector and curvature change detection on the edge points to identify edge points with directional offset and generate structure layer aligned boundary buffer graphics. S2: Based on the alignment boundary buffer graphic of the structure layer, obtain the covered image coordinate area, match the coordinates of the disease points marked in the map, determine whether they fall within the buffer graphic area, and obtain the disease boundary structure annotation record. S3: Call the original positioning coordinates of the disease points in the disease boundary structure annotation record, extract the passage priority field of the disease point and the adjacent nodes for comparison. If the passage priority of the node corresponding to the disease point is lower, it is determined to be a road disease, and the positioning coordinate set after passage attribute calibration is obtained. S4: Using the positioning coordinate set calibrated by the traffic attributes, extract the lane guidance number corresponding to each set of coordinates, obtain the centerline coordinate set and road width field, establish a lateral buffer graphic within the lane width range, and perform area merging on the positioning points falling within the buffer range to obtain the defect annotation overlay layer structure.

[0006] As a further embodiment of the present invention, the structural layer alignment boundary buffer graphic includes a structural layer number index, a set of directional offset edge points, a curvature abrupt change marker area, and a structural boundary buffer zone; the defect boundary structural annotation record includes defect point structural layer number, structural boundary coverage type, buffer matching marker, and boundary feature index; the location coordinate set after traffic attribute calibration includes a corrected coordinate point set, road defect markers, lane center guidance reference, and traffic priority adjustment mark; and the defect annotation overlay layer structure includes a lane centerline coordinate set, a lateral buffer range graphic, a region merging annotation boundary, and a lane guidance number identifier.

[0007] As a further aspect of the present invention, the step of obtaining the structural layer alignment boundary buffer pattern specifically includes: S111: Obtain the road inspection image of the cross-sectional structure layer, perform contour detection on the disease area, extract the set of edge point coordinates, calculate the change of direction vector of adjacent pixels of edge points and the change of angle between edge points, determine the continuity of edge direction and adjacent change trend, filter edge points with sudden angle changes or deviations, and generate a set of edge direction offset points. S112: Call the structural layer number and the transverse start and end coordinate interval in the road cross-section structure diagram, match the spatial position of the edge direction offset point set with the coordinate interval of the structural boundary line, calculate the angle between the edge point direction vector and the structural boundary line, compare the degree of angle difference, and generate the structural boundary offset point set. S113: Based on the distribution of the set of offset points of the structural boundary and the corresponding curvature change data, detect the position of the edge point at the abrupt change in the rate of curvature change, determine whether the spatial position is within the outer buffer of the structural boundary line, calculate the structural overlap distribution value of the edge offset points, and extract the spatially overlapping edge point region by combining the distribution with the spatial coverage relationship of the structural boundary buffer to obtain the structural layer aligned boundary buffer pattern.

[0008] As a further aspect of the present invention, the steps for obtaining the disease boundary structure annotation record are specifically as follows: S211: Based on the structural layer aligned boundary buffer graphic, obtain the coordinates of the marked disease points, identify the coordinate area of ​​the map sheet to which the disease point coordinates belong, call the coordinates of the structural layer boundary buffer graphic area, perform coordinate matching on the disease point coordinates respectively, determine whether the disease point is in any boundary buffer graphic area, filter the disease points that fall into the graphic area, and generate a set of disease points in the buffer graphic. S212: Call the set of disease points in the buffer graphic, call the structural layer number information and boundary type label, analyze the structural layer number corresponding to the disease point in the buffer graphic, extract the boundary type corresponding to the structural boundary, record the combination label relationship between the disease point number and the corresponding structural layer number and boundary type, and obtain the disease point structural label record set. S213: Call each combination relationship in the disease point structure label record set, combine the coordinates of the disease point, the structure layer number and the boundary type label value, calculate the disease boundary structure label value, bind the disease boundary structure label value with the corresponding disease point number, and obtain the disease boundary structure label record.

[0009] As a further aspect of the present invention, the step of obtaining the positioning coordinate set after the passage attribute calibration is specifically as follows: S311: Based on the original positioning coordinates of the disease points in the disease boundary structure annotation record, extract the passage priority field values ​​of the node where the disease point is located and the adjacent nodes, match them using the node number, compare the passage priority field values ​​of the node where the disease point is located and the adjacent nodes, and if the passage priority field value of the node corresponding to the disease point is less than that of the adjacent node, then determine that the disease point is located at a road node and generate a road determination identifier value. S312: Based on the road determination identifier value, read the lane guidance number field from the corresponding defect record, retrieve and match the lane center point dataset according to the number, extract the lane center coordinate point group that matches the guidance number, and generate the lane center coordinate matching set; S313: Call the lane center coordinate matching set and the original positioning coordinates of the defect point, calculate the Euclidean distance value from the original coordinates to the corresponding lane center coordinate point group, introduce the difference of the traffic priority field and the number of adjacent nodes as extended variables, calculate the traffic adjustment distance value, and use the corresponding point as the repositioning coordinate point of the defect point to generate the positioning coordinate set after traffic attribute calibration.

[0010] As a further aspect of the present invention, the step of obtaining the disease annotation overlay layer structure specifically includes: S411: Based on the positioning coordinate set after the traffic attribute calibration, extract the lane guidance number associated with each set of positioning coordinates, and determine the corresponding centerline coordinate set and road width field in combination with the guidance number. According to the linear direction of the centerline coordinates and the corresponding road width field value, calculate the buffer width in the lateral direction of the centerline respectively, and generate the lateral buffer boundary range. S412: Call the lateral buffer boundary range, determine the position of all positioning coordinates within the coverage area, use the overlap relationship between the coordinate values ​​of the positioning points and the corresponding lane lateral buffer boundary to filter the set of positioning points falling into the corresponding range, classify and aggregate the positioning points in the same range, and generate the aggregated coordinate density value within the lane. S413: Based on the aggregated coordinate density value within the lane, match the spatial coverage of the existing disease labeling layer, use the guide number corresponding to the dense coordinate interval as the spatial index, perform overlap detection on the classified positioning point set and the disease labeling layer, determine the coordinate group and corresponding layer fragment with spatial overlap relationship, and obtain the disease labeling coverage layer structure.

[0011] As a further aspect of the present invention, the structural overlap distribution value measures the degree of coupling between the edge point direction, curvature and the boundary of adjacent structures. The disease boundary structure label value characterizes the spatial offset of the disease point relative to the structural layer boundary region and the coupling strength of the structural features. The traffic adjustment distance value represents the spatial offset intensity between the defect point and the center point of the candidate lane, and measures the degree of positioning matching of the defect point in the real-time traffic network.

[0012] As a further aspect of the present invention, the method further includes step S5: S5: Call the grid area encoding information in the disease annotation overlay layer structure, collect the coordinates of repeated disease records in multiple cycles in the grid, count the changes in the distribution of disease points in chronological order, filter the grids with increasing numbers in continuous cycles, connect adjacent grids in the direction of change to construct a continuous path, and obtain the periodic disease distribution path layer. The periodic disease distribution path layer includes a disease quantity increasing grid, a continuous periodic distribution trajectory, adjacent grid connection paths, and grid area coding statistics.

[0013] As a further aspect of the present invention, the steps for obtaining the periodic disease distribution path layer are specifically as follows: S511: Call the grid area coding information in the disease annotation overlay layer structure, collect the disease repeated record coordinates of multiple periods, sort the disease coordinates in the same grid according to the time field, count the number of disease coordinates appearing in the grid in each period, calculate the difference in the number of diseases between adjacent periods, obtain the grid area coding and quantity change information with positive difference, and generate a grid set with increasing disease quantity. S512: Based on the disease quantity increasing grid set, extract the spatial positional relationship between adjacent grids, calculate the coordinate offset of the grid center point in the disease quantity increasing direction in adjacent periods, and determine that grid pairs whose distance between adjacent grid center points in the offset direction is less than the grid side length threshold are connectable units. Connect grid pairs that meet the conditions to construct a continuous grid path and obtain a continuous disease offset path coordinate chain. S513: Based on the continuous disease offset path coordinate chain, aggregate and count the period to which the grid belongs in the path, use the period with increasing disease quantity in the path as the index, combine the corresponding path chains into a layer format, mark the path flow direction in time order, and generate a periodic disease distribution path layer.

[0014] The road defect distribution visualization system is used to execute the above-mentioned road defect distribution visualization method. The system includes: The structural edge extraction module obtains the road inspection image of the cross-sectional structural layer, extracts the set of disease edge points in the image, calculates the direction vector and curvature change between edge points, calls the structural layer number and the horizontal start and end coordinate interval in the road cross-sectional structural image, and generates a structural layer aligned boundary buffer graphic. The buffer matching module obtains the image coordinate region based on the alignment boundary buffer graphic of the structure layer, matches the coordinates of the disease points marked in the map, and determines whether the disease points fall within the buffer graphic region. If the match is successful, the structure layer number and boundary type are marked, and a disease boundary structure annotation record is generated. The structural layer annotation module calls the original positioning coordinates of the disease points in the disease boundary structural annotation record, extracts the passage priority field of the disease points and adjacent nodes and compares the priorities. If the passage priority of the disease point is lower, it is marked as a road disease, and the positioning coordinate set after the passage attribute is calibrated is obtained. The lane attribution calibration module obtains the lane guidance number based on the coordinates in the positioning coordinate set after the traffic attribute calibration, obtains the corresponding lane centerline coordinates and road width field, substitutes the disease point coordinates into the buffer range to determine whether they fall into the corresponding interval, and obtains the disease annotation overlay layer structure. The disease path construction module calls the disease annotation overlay layer structure, counts the coordinates of disease records in the same grid within the period, connects the center coordinates according to the adjacent relationship to construct a continuous path, and generates a periodic disease distribution path layer.

[0015] Compared with the prior art, the advantages and positive effects of the present invention are as follows: In this invention, by processing the disease distribution task of multi-structure combined roads, a boundary buffer graphic is constructed based on the edge direction vector and the cross-sectional coordinate interval of the structure. Combined with the location of curvature change point, the cross-boundary characteristics of the disease boundary between structural layers are dynamically identified. By introducing the structural layer number and the image spatial position, the structural belonging relationship of the disease point is established. The traffic priority field and the lane guidance number field are linked to perform spatial offset calibration on the disease coordinates, forming a disease location result with clear primary and secondary attributes. A disease path coherence expression mechanism based on time series grid is constructed, so that the repeated distribution trend of disease forms a spatial continuous layer. This can avoid the influence of structural misalignment, annotation distortion and periodic noise interference on the layer expression, effectively establish the correlation between structural layering logic and disease trend changes, and support the graphical expression and trend evolution description of road disease distribution results under complex road structures. Attached Figure Description

[0016] Figure 1 This is a schematic diagram of the workflow of the present invention; Figure 2 This is a flowchart illustrating the process of obtaining the structural layer alignment boundary buffer pattern in this invention. Figure 3 This is a flowchart illustrating the process of obtaining the disease boundary structure annotation record in this invention; Figure 4 This is a flowchart illustrating the process of obtaining the location coordinate set after traffic attribute calibration in this invention. Figure 5 This is a flowchart illustrating the process of obtaining the disease annotation overlay layer structure in this invention; Figure 6 This is a flowchart illustrating the process of obtaining the periodic disease distribution path layer in this invention. Detailed Implementation

[0017] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0018] In the description of this invention, it should be understood that the terms "length," "width," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," and "outer," etc., indicating orientation or positional relationships, are based on the orientation or positional relationships shown in the accompanying drawings and are only for the convenience of describing the invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the invention. Furthermore, in the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0019] Please see Figure 1 This invention provides a technical solution, a method for visualizing the distribution of road defects, comprising the following steps: S1: Obtain the road inspection image of the cross-sectional structure layer, perform contour detection on the disease area, extract the set of edge point coordinates, and perform direction vector and curvature change detection on the edge points. Call the structural layer number and horizontal start and end coordinate interval in the road cross-sectional structure map. Based on the angle feature between the edge point direction vector and the structural boundary line, identify the edge points with directional deviation. When the curvature change position is close to the structural boundary, it is marked as a multi-layer interlacing area, and a structural layer aligned boundary buffer graphic is generated. S2: Based on the structural layer alignment boundary buffer graphic, obtain the covered image coordinate area, match the coordinates of the disease points marked in the map, determine whether they fall within the buffer graphic area, if the coverage relationship is satisfied, then mark the structural layer number and boundary type corresponding to the disease point, and obtain the disease boundary structure annotation record. S3: Call the original positioning coordinates of the disease point in the disease boundary structure annotation record, extract the passage priority field of the disease point and the adjacent nodes for comparison. If the passage priority of the node corresponding to the disease point is lower, it is determined to be a road disease. Read the lane guidance number field to obtain the lane center coordinates, reposition the coordinate position of the disease point, and obtain the positioning coordinate set after the passage attribute calibration. S4: Using the positioning coordinate set calibrated by the traffic attributes, extract the lane guidance number corresponding to each set of coordinates, obtain the centerline coordinate set and road width field, establish a lateral buffer graphic within the lane width range, and perform area merging on the positioning points falling within the buffer range to obtain the defect annotation overlay layer structure. S5: Call the grid area encoding information in the disease annotation overlay layer structure, collect the coordinates of repeated disease records in multiple cycles in the grid, count the changes in disease point distribution in chronological order, filter the grids with increasing numbers in continuous cycles, connect adjacent grids according to the direction of change to construct a coherent path, and obtain the periodic disease distribution path layer. The structural layer alignment boundary buffer graphic includes structural layer number index, direction offset edge point set, curvature change marker area, and structural boundary buffer zone. The defect boundary structure annotation record includes defect point structural layer number, structural boundary coverage type, buffer matching marker, and boundary feature index. The positioning coordinate set after traffic attribute calibration includes the corrected coordinate point set, road defect marker, lane center guidance reference, and traffic priority adjustment mark. The defect annotation cover layer structure includes lane centerline coordinate set, lateral buffer range graphic, area merging annotation boundary, and lane guidance number identifier. The periodic defect distribution path layer includes defect number increasing grid, continuous periodic distribution trajectory, adjacent grid connection path, and grid area coding statistics.

[0020] Please see Figure 2 The specific steps for obtaining the structural layer alignment boundary buffer pattern are as follows: S111: Obtain the road inspection image of the cross-sectional structure layer, perform contour detection on the disease area, extract the set of edge point coordinates, calculate the change of direction vector of adjacent pixels of edge points and the change of angle between edge points, determine the continuity of edge direction and adjacent change trend, filter edge points with sudden angle changes or deviations, and generate a set of edge direction offset points. The original image's pixel set is subjected to grayscale normalization, ensuring that pixel values ​​are concentrated between 0 and 255. Based on this, a 3×3 sliding window is used to perform edge gradient detection, and a preliminary edge point set is extracted based on the degree of gradient direction change. Edge segments with angle abrupt changes greater than 15° between consecutive pixels are divided into regions, constructing an initial contour set of the diseased area's edges. Single-point direction vector estimation is performed on the pixels within this contour segment. The direction vector is set as the unit vector connecting adjacent pixels, and the angle change is used as a parameter to determine edge continuity. In the detection samples, when an edge point's direction vector has an angle abrupt change greater than 12° within a 5-pixel window, it is marked as a direction offset point. Combining the edge morphology angle change process, the five-point center difference method is used to estimate the local difference in the angle formed by adjacent triangle sides. In the sample image, the edge angle change range is found to be between 0° and 27°. Those with changes greater than 18° are classified as non-consistent direction regions. The set of overlapping regions of direction vector abrupt changes and non-consistent angle changes is output to generate an edge direction offset point set.

[0021] S112: Call the structural layer number and the transverse start and end coordinate range in the road cross-section structure diagram, match the spatial position of the edge direction offset point set with the coordinate range of the structural boundary line, calculate the angle between the edge point direction vector and the structural boundary line, compare the degree of angle difference, and generate the structural boundary offset point set. For each edge point in the set of edge direction offset points, a structural boundary mapping operation is performed, projecting the coordinates of each edge point onto the horizontal coordinate interval corresponding to the structural boundary line segment recorded in the cross-sectional structural diagram, and recording the structural layer number. If the horizontal coordinate of the edge point is within the horizontal interval of the structural layer, and the distance between the vertical coordinate and the boundary line of that structural layer does not exceed 5 pixels, then the valid correspondence is considered to be established. For each edge point with an associated boundary line, the angle θ between its own direction vector and the direction of the structural boundary line segment is calculated, and the vector dot product formula is used. Angle judgment was performed. In the test image samples, edge points with an included angle θ greater than 25° were defined as areas with significant directional shifts. During this screening process, the angle change values ​​were recorded and summarized, as shown in Table 1. The angle differences in the summary results were mainly concentrated in the range of 18° to 36°. The set of edge points with significant angle shifts was merged to obtain the set of structural boundary shift points. Table 1: Statistics of Angles Between Direction Vectors and Structural Boundaries

[0022] As shown in Table 1, the directional angles between multiple edge points and the structural boundary lines are significantly greater than 25°. Based on the directional difference measurement standard, they are determined to be boundary offset points.

[0023] S113: Based on the distribution of the structural boundary offset point set and the corresponding curvature change data, detect the position of the edge point at the point of abrupt change in the rate of curvature change, and determine whether the spatial position is within the outer buffer zone of the structural boundary line, using the formula: ; Calculate the structural overlap distribution value of the edge offset points, combine the distribution with the spatial coverage relationship of the structural boundary buffer, extract the spatially overlapping edge point regions, and obtain the structural layer alignment boundary buffer pattern; in, Represents the distribution value of structural overlap. Representing the The direction vector angle of each edge point This represents the angular direction of the structural boundary line corresponding to the location. Represents the rate of change of curvature at the edge point. This represents the shortest distance from an edge point to the structural boundary line. This represents the total number of structural boundary offset points. Formula calculation logic: By introducing the angle difference of edge direction vectors Curvature change value Perpendicular distance from edge point to structural boundary A weighted integral index reflecting the overlapping characteristics of the structure is established, and three data acquisitions are performed for each edge offset point: the angle difference between its own direction and the direction of the structural boundary is calculated through the vector inner product relationship. This reflects the magnitude of local edge offset; the curvature change value of the edge points is estimated based on the three-point difference method. This represents the local deformation trend of the edge line; the vertical distance is obtained using the point-to-line distance formula. To measure the degree to which the edge point is close to the structural boundary, the square root of the sum of the curvature and the square of the distance is used to generate the structural offset strength value. This value is then multiplied by the direction angle difference to form a coupling term, which represents the degree of structural offset at a single point. The average of the coupling values ​​at the offset points is then summed to obtain the structural overlap distribution value. The structural overlap distribution value is used to measure the degree of coupling between the direction and curvature of the edge point and the boundary of the adjacent structure. It is a numerical index that reflects the spatial overlap characteristics of the disease edge and the structural layer distribution. The larger the value, the more significant the edge direction shift and the higher the degree of spatial proximity to the structural boundary area. It can be used to help identify the location of multi-layer structure intersection or abnormal boundary response. Extract its corresponding curvature change value Curvature calculation is based on the three-point difference method under curve fitting, through edge points. The curvature was calculated from the change in normal angle. In the measured example, the curvature change at the edge points was concentrated between 0.05 and 0.62. The shortest vertical distance from the edge point to the structural boundary line was further extracted. Numerical calculations were performed using the projection formula from a point to a line, and the results were... Values ​​range from 0.3 to 6.4 pixels; combined with edge point orientation angles. Angle with structural boundary line The difference; In the formula, In this embodiment, the total number of edge points in the set of structural boundary offset points is [number]. These correspond to the numbers E107, E213, E315, and E502, respectively. Substitute the parameter values ​​into the formula and perform the following calculation: Point E107: ; Point E213: ; Point E315: ; Point E502: ; Calculate each item separately: ; ; ; ; Summarize and average: ; The results show that the structural overlap distribution value is 142.74. This value is combined with the position range of the structural boundary buffer zone to perform spatial overlap matching. The matching condition is that if the projection range of the offset point falls within the strip-shaped interval 3 pixels outside the boundary line, it is marked as an overlapping area, and the structural layer is aligned with the boundary buffer pattern.

[0024] Please see Figure 3 The specific steps for obtaining the disease boundary structure annotation record are as follows: S211: Based on the structural layer aligned boundary buffer graphic, obtain the coordinates of the marked disease points, identify the coordinate area of ​​the map sheet to which the disease point coordinates belong, call the coordinates of the structural layer boundary buffer graphic area, perform coordinate matching on the disease point coordinates respectively, determine whether the disease point is in any boundary buffer graphic area, filter the disease points that fall into the graphic area, and generate a set of disease points in the buffer graphic. The map sheets are numbered and labeled, and the structural layer boundary buffer graphic is vectorized using a spatial reference system. The coordinate set of the disease points is extracted from the original vector map sheets. The coordinates of the disease points are analyzed, and their relative position coordinate coefficients in the current map sheet are determined to ensure accurate matching with the structural layer boundary buffer graphic. For example, map sheet number T001 has a disease point P1 with coordinates (152.3, 78.5). This value is used to calculate the spatial intersection with the polygon coordinate set in the boundary buffer graphic to determine if it is within the structural layer boundary graphic. In practice, the buffer graphic consists of several closed polylines. The determination is made by checking if the disease point satisfies the point-in-pol... The ygon relationship is used for matching. If the disease point P1 matches the buffer graphic number S005, its position status is recorded as "internal". Otherwise, it is recorded as "external". In the batch processing, each disease point is matched and judged by a point-by-point screening method. Boolean values ​​are used to indicate whether it falls within any boundary graphic. The set of disease points with the status value of "internal" is filtered out and its index value under the map sheet number, disease point ID and structural layer boundary number is recorded. A total of 25 disease points are extracted under map sheet T001. 17 of them fall within the area of ​​buffer graphic numbers S001 to S006. After comparison, they are respectively assigned to their respective graphics to obtain the set of disease points within the buffer graphic.

[0025] S212: Call the coordinate values ​​of the disease points in the disease point set within the buffer graphic, call the structural layer number information and boundary type label, analyze the structural layer number corresponding to the disease point within the buffer graphic, extract the boundary type corresponding to the structural boundary, record the combination label relationship between the disease point number and the corresponding structural layer number and boundary type, and obtain the disease point structural label record set. Obtain the corresponding map sheet number, point number, and coordinate position index, and simultaneously read the structural layer number data and boundary type code indicated by the corresponding buffer graphic number. During the operation, a multi-field index mapping relationship needs to be established to match the structural layer number with the boundary type one by one. If the defect point P1 is located in the boundary graphic S005 corresponding to the structural layer number L03, and the boundary type number is set to BT02, then a combined index P1-L03-BT02 can be constructed. During batch processing, this operation is performed for each defect point, and the combined triplet result is stored in a structured manner. The defect point number adopts a sequential coding method, the structural layer number is the number defined in the design map sheet, and the boundary type adopts a standardized dictionary coding method. "BT01" represents the cut boundary, "BT02" represents the joint boundary, etc. The combined result is sorted in ascending order by the defect point number and recorded as a structural mark index table. The fields of the structural mark index table are shown in Table 2. Table 2: Record of Structural Markings of Disease Points

[0026] Table 2 lists the storage format of the structural marker index. Each record corresponds to a unique disease point. This structure is used as the basis for calculating the subsequent structural boundary label values, forming a disease point structural marker record set.

[0027] S213: Call upon each combination relationship in the disease point structure marker record set, combining the coordinates of the disease point location, the structure layer number, and the boundary type label value, using the following formula: ; Calculate the disease boundary structure annotation value, and bind the disease boundary structure annotation value with the corresponding disease point number to obtain the disease boundary structure annotation record; in, Representative disease points Corresponding structural layer The labeling value of the disease boundary structure, Representative disease points In structural layer The coordinates of the center of the boundary envelope region on the upper boundary. Representative disease points The coordinates themselves are in the structural layer Matching coordinate values ​​in Indicates the location of the disease In structural layer The extension width value of the middle boundary region, Indicates the location of the disease Boundary graphics in structural layers Vertical offset value, Disease location Structural layer The Middle Item associated boundary line density value, The number of boundary lines; Formula calculation logic: Based on the coordinates of the disease point 'a' and its structural layer Geometric center coordinates of the middle boundary envelope region Perform coordinate difference calculations and obtain the spatial offset value in absolute form, then read the extension width of the corresponding boundary region in structure layer b. With vertical offset value Square both factors separately, add them together, and then take the square root to form the comprehensive structural scale factor of the disease point under that structural layer. The number of boundary lines M is counted, and the density factor corresponding to the k-th boundary line is obtained sequentially. ,Will Summing and adding 1 as an adjustment term is used to standardize and correct the error caused by the boundary distribution density. Multiplying the offset value by the comprehensive structural scale and then dividing by the density adjustment term yields the labeled value of the disease boundary structure. ; The disease boundary structure annotation value is used to characterize the spatial offset of the disease point relative to the boundary area of ​​the structural layer and the coupling strength of the structural features. This value comprehensively considers factors such as geometric offset, boundary scale and boundary line distribution interference, and serves as a key quantitative indicator for subsequent structural attribute analysis. Perform boundary structure annotation calculations for each structural marker combination and read the coordinates of the defect point 'a'. And read the coordinates of the geometric center of its boundary envelope region. Calculate the Euclidean distance between the two and obtain the offset value through absolute value operation. Then read the extension width value of the boundary region in structure layer b. and vertical offset value By combining the two, the sum of squares is calculated and the square root is taken to obtain the synthetic geometric scale. When calculating the boundary line interference influence factor, the statistical structural layer is used. The number of boundary lines is M, and the density factor of each boundary line is calculated. The total interference value is obtained by summation. ; Taking point P1 as an example, its boundary center coordinates Coordinates of the disease point Calculate the Euclidean offset value. The absolute value is 1.836, and the extension width is... m, vertical offset value m, then the square root term is The number of boundary lines M is 4, and its density value is The sum is 0.54, so the calculation is as follows: ; The results show that the disease boundary structure annotation value of the disease point P1 on the structural layer L03 is 2.946. By constructing the annotation record value and binding the original structural mark index combination, the disease boundary structure annotation record is obtained.

[0028] Please see Figure 4 The specific steps for obtaining the location coordinate set after traffic attribute calibration are as follows: S311: Based on the original location coordinates of the disease points in the disease boundary structure annotation record, extract the passage priority field values ​​of the node where the disease point is located and the adjacent nodes, use the node number for matching, compare the passage priority field values ​​of the node where the disease point is located and the adjacent nodes, if the passage priority field value of the node corresponding to the disease point is less than that of the adjacent node, then determine that the disease point is located at a road node and generate a road judgment identifier value. The node number corresponding to the defective point and its upstream and downstream neighboring node numbers are extracted. The corresponding traffic priority field value is retrieved from the road network topology and used as the basis for subsequent processing to classify the traffic priority. The traffic priority field value is set as a discrete integer, ranging from 1 to 5, with smaller values ​​indicating higher traffic priority. The priority is assigned as follows: main roads 1, secondary roads 2, roads 3, branch roads 4, and internal passages 5. Based on the node number, the traffic priority field value of the node containing the defective point is compared pairwise with the traffic priority field values ​​of nodes in its neighboring node set to determine the defect. The condition for determining whether a node value is greater than the minimum value of its neighboring nodes is used. If the traffic priority field value of the node where the defect point is located is not the minimum value in the set of traffic priority field values ​​of its neighboring nodes, then it is determined to be a road defect point. The traffic priority field value of the node corresponding to defect point A is set to 4, and the field values ​​corresponding to the adjacent node numbers are 2, 3, and 3 respectively. Since 4 is greater than the minimum value 2, A is determined to be a road node, and a road determination identifier value is generated. Its two values ​​are defined as 0 for non-road and 1 for road. The road determination identifier value can be generated by comparing the structural annotation record examples listed in Table 3. Table 3: Examples of Boundary Structure Annotations

[0029] As shown in Table 3, the defect point A001 is identified as a road defect with a corresponding value of 1, while the other points do not meet the criteria and have a corresponding value of 0.

[0030] S312: Based on the road determination identifier value, read the lane guidance number field from the corresponding defect record, retrieve and match the lane center point dataset according to the number, extract the lane center coordinate point group that matches the guidance number, and generate the lane center coordinate matching set; By reading the lane guidance number field corresponding to the defect point number in the defect annotation record, and using this number as a key index, the corresponding lane number group is retrieved in the lane centerline information database. The set of lane center point coordinates corresponding to this group is then extracted. The lane guidance number uses a structured numbering method, such as "L12" or "L45". The lane center points are equidistant sampling point groups divided according to the actual road, with each group containing approximately 10 to 20 coordinate points. Each point is represented by two-dimensional coordinates. The system is composed of a metric coordinate system with a sampling interval of 1.0 meter. In actual execution, if the guide number of the defect point A001 is set as L12, the coordinate array group corresponding to L12 will be called, which contains the point sequence P1 to P20. The point sequence is as follows: P1=(198.2, 412.5), P2=(199.2, 412.6), ..., P20=(217.0, 413.9). This array is assembled into a lane center coordinate matching set for subsequent use. To enhance matching accuracy, each set of center coordinate points must ensure continuity and stable coordinate changes. The maximum Euclidean distance between a single point and its adjacent points must not exceed 1.5 meters. This generates the lane center coordinate matching set.

[0031] S313: Using the lane center coordinate matching set and the original location coordinates of the defect points, calculate the Euclidean distance from the original coordinates to the corresponding lane center coordinate point group. Introduce the traffic priority field difference and the number of adjacent nodes as extended variables, using the following formula: ; Calculate the passage adjustment distance value, use the corresponding point as the disease point to relocate the coordinate point, and generate the positioning coordinate set after passage attribute calibration; in, Adjust the distance value for passage. The x-coordinate represents the original location coordinates of the disease point. The x-coordinate represents the center point of the lane. The vertical coordinate represents the original location coordinates of the diseased point. The ordinate represents the center point of the lane. The priority field value represents the path of the node where the defective point is located. The average value of the passage priority field representing adjacent nodes. Represents the number of adjacent nodes; Formula calculation logic: Using Euclidean distance as the basic indicator of spatial offset, the two-dimensional straight-line distance between the original coordinates of the fault point and the center coordinates of each lane is calculated to measure the deviation in physical location. The absolute value of the difference between the traffic priority field values ​​of the node where the fault point is located and its adjacent nodes is introduced to reflect the degree of difference between traffic structure levels. This difference is then added to the spatial distance value to form the total offset value. To avoid inconsistent offset intensity caused by differences in the number of node connections, the above total value is divided by a factor of 1 (the number of adjacent nodes plus 1) to normalize the complexity of the traffic environment. The output is... The value can comprehensively reflect the degree of matching between the road traffic structure and the physical spatial location of the defect points; The smaller the value, the closer the center coordinate point is to the disease point in terms of spatial location and passage structure, making it suitable as a positioning point after passage attribute calibration. The traffic adjustment distance value represents the weighted comprehensive value of the spatial offset intensity between the defect point and the center point of each candidate lane and the difference in traffic structure level. It measures the degree of reasonable positioning and matching of the defect point in the actual traffic network. By integrating spatial distance and traffic structure differences and through weighted normalization, the overall deviation of each center point relative to the defect point is obtained. Calculate the Euclidean distance between the defect point and each point in the coordinate set of the lane center. For example, if the original coordinates of the defect point are... Lane center point The corresponding Euclidean distance is Meters, call the priority field value of the node where the disease point is located. Average passage priority field value of adjacent nodes Node adjacency count ; Substituting into the formula, we get: ; Calculated from the center point The values ​​are sorted, and the coordinate point corresponding to the minimum value is selected as the new location of the disease point. This value is one of the elements of the location coordinate set after the passage attribute calibration. The advantage of this formula lies in its ability to couple the location of disease points with the complexity of the traffic path structure by introducing the absolute value of the difference in the traffic priority field, using weighted division based on the number of adjacent nodes, and combining this with the Euclidean distance term. This enhances the matching strength between disease points and the actual traffic structure, thereby improving the rationality of their location. To weight and adjust the distance values, the matching degree between the disease point and each candidate center point is measured; Table 4: Example table of value calculation

[0032] Table 4 lists the different center points The calculation process shows that P2 corresponds to... The value is the smallest, so its coordinates are selected as the positioning coordinates after the disease point is calibrated. This result shows that the original positioning of the disease point deviates from the main traffic direction in the traffic path structure, and has the smallest weighted path offset from point P2 in the spatial structure, so it is used as the corrected reference point.

[0033] Please see Figure 5 The specific steps for obtaining the disease annotation overlay layer structure are as follows: S411: Based on the positioning coordinate set after traffic attribute calibration, extract the lane guidance number associated with each set of positioning coordinates, and determine the corresponding centerline coordinate set and road width field in combination with the guidance number. According to the linear direction of the centerline coordinates and the corresponding road width field value, calculate the buffer width in the lateral direction of the centerline respectively, and generate the lateral buffer boundary range. Extract the lane guidance number associated with the location point. This process can be achieved using GIS spatial analysis software such as ArcGIS or PostGIS plugins. During this process, each location point is spatially matched with the road layer to determine whether it falls within any lane buffer zone. The lane number is determined by spatial inclusion judgment. The "ST-Contains" function can be used to detect whether the point is located within the buffer polygon, obtaining the matching relationship between the location point and the Lane-ID. Based on the guidance number, the corresponding centerline coordinate set is extracted from the road database. This coordinate set is a polyline segment composed of a set of ordered points, representing the actual shape of the road. The road width field corresponding to the lane needs to be extracted. If the width field value of a certain main road is set to 3.75 meters, then its buffer zone in the lateral direction is half the width and extends to both sides. Considering the existence of road offset or positioning error, the buffer width can be set to 0.5 to 0.6 times the width, generating the lateral buffer boundary range.

[0034] S412: Call the lateral buffer boundary range, determine the position of all positioning coordinates within the coverage area, use the overlap relationship between the coordinate value of the positioning point and the corresponding lane lateral buffer boundary to filter the set of positioning points falling into the corresponding range, classify and aggregate the positioning points in the same range, and generate the aggregated coordinate density value within the lane. The location of the coordinate points within the coverage area is determined by detecting the overlap between each point and the buffer area. An algorithm for determining if a point is within a polygon can be used, such as the ray method. When point Q is within a certain area, a ray is emitted from point Q in any direction and its intersection with each side of the polygon is detected. If the number of intersections is odd, point Q is within the polygon area; otherwise, it is not. After completing the overlap determination, the overlapping positioning points are grouped according to their guide numbers, and the points within each group are aggregated. The aggregation process can be based on dividing the area into smaller intervals based on length units. For example, each lane buffer zone can be divided into several 10-meter segments along the centerline. The point density within each segment is calculated. The aggregation density represents the frequency of vehicle traffic or the concentration of positioning data in the area. If there are 26 points in the third segment of a lane, the density of that segment can be recorded as 26 points / 10 meters. This method obtains the distribution of density changes along the lane and generates the aggregated coordinate density value within the lane.

[0035] S413: Based on the aggregated coordinate density value within the lane, match the spatial coverage of the existing disease labeling layer, use the guide number corresponding to the dense coordinate interval as the spatial index, perform overlap detection on the classified positioning point set and the disease labeling layer, determine the coordinate group and corresponding layer fragment with spatial overlap relationship, and obtain the disease labeling coverage layer structure. Using each densely populated area as the analysis target, spatial matching is performed against existing disease labeling layers. The guide number corresponding to each high-density distribution segment needs to be determined based on the aggregated density value, and a spatial index needs to be established. Vector tile areas corresponding to these numbers are selected from the disease layer. Geometric operation functions such as "ST-Intersects" or "ST-Overlap" are used to perform spatial overlay analysis on the coordinate points and disease labeling layers. It is determined whether the aggregated point set geometrically overlaps with the disease area. If overlap exists, the corresponding coordinate point set and disease layer fragment are extracted. A disease area with the number D-007 is defined in a certain layer D, with its spatial boundary containing several points. If multiple consecutive points in the location point set P are located within the range of D-007, and the density value is greater than a preset threshold such as 20 points / 10 meters, then this aggregated segment can be considered to have a spatial overlap relationship with the disease layer. Data pairs with spatial coverage relationships are selected, and the structural information of the disease area layer is extracted, such as its lane, disease type, and layer labeling attributes. This prepares the data foundation for subsequent layer structure updates and data integration, obtaining the structure of the disease labeling overlay layer.

[0036] Please see Figure 6 The specific steps for obtaining the distribution path layer of periodic diseases are as follows: S511: Call the grid area coding information in the disease annotation overlay layer structure, collect the disease repeated record coordinates of multiple periods, sort the disease coordinates in the same grid according to the time field, count the number of disease coordinates in the grid in each period, calculate the difference in the number of diseases between adjacent periods, obtain the grid area coding and quantity change information with positive difference, and generate a grid set with increasing disease quantity. Multiple periods of repeated disease records need to be collected sequentially. Specific operations include using database query languages ​​or geographic information system tools (such as PostGIS, ArcGIS, etc.) to extract multi-period inspection data, classifying and sorting disease records according to the "grid code" and "collection time" fields, extracting disease records for January, March, and May 2024, and mapping the records to their respective grid areas. For example, grid G101 recorded disease 2 times in January 2024 and 4 times in March 2024. After extracting the records, the coordinates of the records in each grid are then analyzed. The system sorts the disease coordinates within the same grid according to their cycles to facilitate subsequent difference calculations. It counts the number of disease coordinates in the grid periodically. For example, in grid G101, there are 2 disease points in January 2024 and 4 disease points in March, resulting in a difference of 4-2=2. This difference is stored in the "Cycle Difference Table." The system iterates through the grid, judging each pair of adjacent cycle differences. When the difference is greater than zero (ΔN>0), it records the grid code and its difference. It then summarizes the grid codes with positive differences and their disease quantity changes, generating a grid set with increasing disease quantity.

[0037] S512: Based on the grid set with increasing disease quantity, extract the spatial positional relationship between adjacent grids, calculate the coordinate offset of the grid center point in the direction of increasing disease quantity in adjacent periods, and determine that grid pairs with a distance between adjacent grid center points in the offset direction less than the grid side length threshold are connectable units. Connect grid pairs that meet the conditions to construct a continuous grid path and obtain a continuous disease offset path coordinate chain. The spatial relationship between adjacent grids needs to be extracted. During the operation, spatial analysis is performed using the center point coordinates of the grids. For example, the center point of each 100m × 100m grid can be calculated from its boundary coordinates. If the boundary of grid G205 is (400, 500) - (500, 600), then the center is (450, 550). After calculating the center points of the increasing grids, it is determined whether adjacent grids are spatially connected. A 4-adjacency or 8-adjacency method is used for comparison. If the difference in center point coordinates between G205 and G206 does not exceed the grid side length threshold, they are considered spatially adjacent. The increasing direction of the number of diseases is calculated based on the overlap of disease coordinates in each cycle. Once the centroid of the fault is determined, if the centroid of the fault coordinates moves from (450, 550) to (460, 570) between periods t1 and t2, the offset is (10, 20). The angle between this offset direction and the direction of the line connecting the center points of adjacent grids is compared. If the angle θ is less than 45 degrees and the distance between the two points is less than 120 meters, the grid pair is determined to be connectable. A connecting unit is set as G205→G206. The traversal connection operation is continued, and grid pairs that meet the conditions are connected to form a path chain, such as G205→G206→G210. The continuous center point coordinate chain is recorded as the fault offset path, and the continuous fault offset path coordinate chain is obtained.

[0038] S513: Based on the coordinate chain of the continuous disease offset path, aggregate and count the period to which the grid in the path belongs, use the period in which the number of diseases in the path increases as the index, combine the corresponding path chains into a layer format, mark the path flow direction in time order, and generate a periodic disease distribution path layer. When aggregating and statistically analyzing the cycles of grids within a path, it is necessary to extract the incrementing cycle number of each grid in the path. For example, in the path G310→G311→G315, the disease growth in G310 occurs in January 2025, G311 in March 2025, and G315 in May 2025. These three cycles are combined to generate the complete time series for the path. The paths are then grouped and aggregated according to their start and end cycles to form a disease path cycle index table. Each path chain is converted into a layer format, with each path in the layer recorded as a vector line segment, accompanied by attribute fields such as "path". The path is set as follows: ID, start and end period, number of periods, total path length, number of grids involved, etc. Path P003 is set with a start period of 202501, an end period of 202505, 3 grids involved, and a total length of approximately 450 meters. After generating a vector path by combining the coordinate sequence of the path center point, directional arrow markers are superimposed on the path in chronological order using the GIS style setting function to indicate the time flow of the disease path. The arrow indicators are set as: G310→G311→G315, for subsequent loading for visualization or for analyzing and processing the periodic disease distribution trend, generating a periodic disease distribution path layer.

[0039] The road defect distribution visualization system is used to execute the above-mentioned road defect distribution visualization method. The system includes: The structural edge extraction module obtains the road inspection image of the cross-sectional structural layer, extracts the set of disease edge points in the image, calculates the direction vector and curvature change between edge points, calls the structural layer number and the horizontal start and end coordinate interval in the road cross-sectional structural image, and generates a structural layer aligned boundary buffer graphic. The buffer matching module obtains the image coordinate region based on the structural layer alignment boundary buffer graphic, matches the coordinates of the disease points marked in the map, and determines whether the disease points fall within the buffer graphic region. If the match is successful, the structural layer number and boundary type are marked, and a disease boundary structure annotation record is generated. The structural layer annotation module calls the original positioning coordinates of the disease points in the disease boundary structural annotation record, extracts the passage priority field of the disease points and adjacent nodes and compares the priorities. If the passage priority of the disease point is lower, it is marked as a road disease, and the positioning coordinate set after the passage attribute is calibrated is obtained. The lane attribution calibration module obtains the lane guidance number based on the coordinates in the positioning coordinate set after traffic attribute calibration, obtains the corresponding lane centerline coordinates and road width field, substitutes the disease point coordinates into the buffer range to determine whether they fall into the corresponding interval, and obtains the disease annotation overlay layer structure. The disease path construction module calls the disease annotation overlay layer structure, counts the coordinates of disease records in the same grid within the period, connects the center coordinates according to the adjacent relationship to construct a continuous path, and generates a periodic disease distribution path layer.

[0040] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention in any other way. Any person skilled in the art may make changes or modifications to the above-disclosed technical content to create equivalent embodiments that can be applied to other fields. However, any simple modifications, equivalent changes, and modifications made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the protection scope of the present invention.

Claims

1. A method for visualizing the distribution of road defects, characterized in that, Includes the following steps: S1: Obtain road inspection images of cross-sectional structure layers, perform contour detection on the diseased areas, extract the set of edge point coordinates, and perform direction vector and curvature change detection on the edge points to identify edge points with directional offset and generate structure layer aligned boundary buffer graphics. S2: Based on the alignment boundary buffer graphic of the structure layer, obtain the covered image coordinate area, match the coordinates of the disease points marked in the map, determine whether they fall within the buffer graphic area, and obtain the disease boundary structure annotation record. S3: Call the original positioning coordinates of the disease points in the disease boundary structure annotation record, extract the passage priority field of the disease point and the adjacent nodes for comparison. If the passage priority of the node corresponding to the disease point is lower, it is determined to be a road disease, and the positioning coordinate set after passage attribute calibration is obtained. S4: Using the positioning coordinate set calibrated by the traffic attributes, extract the lane guidance number corresponding to each set of coordinates, obtain the centerline coordinate set and road width field, establish a lateral buffer graphic within the lane width range, and perform area merging on the positioning points falling within the buffer range to obtain the defect annotation overlay layer structure.

2. The method for visualizing the distribution of road defects according to claim 1, characterized in that, The structural layer alignment boundary buffer graphic includes a structural layer number index, a set of directional offset edge points, a curvature abrupt change marker area, and a structural boundary buffer zone. The defect boundary structure annotation record includes defect point structural layer number, structural boundary coverage type, buffer matching marker, and boundary feature index. The location coordinate set after traffic attribute calibration includes a corrected coordinate point set, road defect markers, lane center guidance reference, and traffic priority adjustment mark. The defect annotation overlay layer structure includes a lane centerline coordinate set, a lateral buffer range graphic, a region merging annotation boundary, and a lane guidance number identifier.

3. The method for visualizing the distribution of road defects according to claim 1, characterized in that, The specific steps for obtaining the alignment boundary buffer pattern of the structural layer are as follows: S111: Obtain the road inspection image of the cross-sectional structure layer, perform contour detection on the disease area, extract the set of edge point coordinates, calculate the change of direction vector of adjacent pixels of edge points and the change of angle between edge points, determine the continuity of edge direction and adjacent change trend, filter edge points with sudden angle changes or deviations, and generate a set of edge direction offset points. S112: Call the structural layer number and the transverse start and end coordinate interval in the road cross-section structure diagram, match the spatial position of the edge direction offset point set with the coordinate interval of the structural boundary line, calculate the angle between the edge point direction vector and the structural boundary line, compare the degree of angle difference, and generate the structural boundary offset point set. S113: Based on the distribution of the set of offset points of the structural boundary and the corresponding curvature change data, detect the position of the edge point at the abrupt change in the rate of curvature change, determine whether the spatial position is within the outer buffer of the structural boundary line, calculate the structural overlap distribution value of the edge offset points, and extract the spatially overlapping edge point region by combining the distribution with the spatial coverage relationship of the structural boundary buffer to obtain the structural layer aligned boundary buffer pattern.

4. The method for visualizing the distribution of road defects according to claim 3, characterized in that, The specific steps for obtaining the disease boundary structure annotation record are as follows: S211: Based on the structural layer aligned boundary buffer graphic, obtain the coordinates of the marked disease points, identify the coordinate area of ​​the map sheet to which the disease point coordinates belong, call the coordinates of the structural layer boundary buffer graphic area, perform coordinate matching on the disease point coordinates respectively, determine whether the disease point is in any boundary buffer graphic area, filter the disease points that fall into the graphic area, and generate a set of disease points in the buffer graphic. S212: Call the set of disease points in the buffer graphic, call the structural layer number information and boundary type label, analyze the structural layer number corresponding to the disease point in the buffer graphic, extract the boundary type corresponding to the structural boundary, record the combination label relationship between the disease point number and the corresponding structural layer number and boundary type, and obtain the disease point structural label record set. S213: Call each combination relationship in the disease point structure label record set, combine the coordinates of the disease point, the structure layer number and the boundary type label value, calculate the disease boundary structure label value, bind the disease boundary structure label value with the corresponding disease point number, and obtain the disease boundary structure label record.

5. The method for visualizing the distribution of road defects according to claim 4, characterized in that, The specific steps for obtaining the location coordinate set after the accessibility attribute calibration are as follows: S311: Based on the original positioning coordinates of the disease points in the disease boundary structure annotation record, extract the passage priority field values ​​of the node where the disease point is located and the adjacent nodes, match them using the node number, compare the passage priority field values ​​of the node where the disease point is located and the adjacent nodes, and if the passage priority field value of the node corresponding to the disease point is less than that of the adjacent node, then determine that the disease point is located at a road node and generate a road determination identifier value. S312: Based on the road determination identifier value, read the lane guidance number field from the corresponding defect record, retrieve and match the lane center point dataset according to the number, extract the lane center coordinate point group that matches the guidance number, and generate the lane center coordinate matching set; S313: Call the lane center coordinate matching set and the original positioning coordinates of the defect point, calculate the Euclidean distance value from the original coordinates to the corresponding lane center coordinate point group, introduce the difference of the traffic priority field and the number of adjacent nodes as extended variables, calculate the traffic adjustment distance value, and use the corresponding point as the repositioning coordinate point of the defect point to generate the positioning coordinate set after traffic attribute calibration.

6. The method for visualizing the distribution of road defects according to claim 5, characterized in that, The specific steps for obtaining the disease annotation overlay layer structure are as follows: S411: Based on the positioning coordinate set after the traffic attribute calibration, extract the lane guidance number associated with each set of positioning coordinates, and determine the corresponding centerline coordinate set and road width field in combination with the guidance number. According to the linear direction of the centerline coordinates and the corresponding road width field value, calculate the buffer width in the lateral direction of the centerline respectively, and generate the lateral buffer boundary range. S412: Call the lateral buffer boundary range, determine the position of all positioning coordinates within the coverage area, use the overlap relationship between the coordinate values ​​of the positioning points and the corresponding lane lateral buffer boundary to filter the set of positioning points falling into the corresponding range, classify and aggregate the positioning points in the same range, and generate the aggregated coordinate density value within the lane. S413: Based on the aggregated coordinate density value within the lane, match the spatial coverage of the existing disease labeling layer, use the guide number corresponding to the dense coordinate interval as the spatial index, perform overlap detection on the classified positioning point set and the disease labeling layer, determine the coordinate group and corresponding layer fragment with spatial overlap relationship, and obtain the disease labeling coverage layer structure.

7. The method for visualizing the distribution of road defects according to claim 6, characterized in that, The structural overlap distribution value measures the degree of coupling between the edge point direction, curvature and the boundary of adjacent structures. The disease boundary structure label value characterizes the spatial offset of the disease point relative to the structural layer boundary region and the coupling strength of the structural features. The traffic adjustment distance value represents the spatial offset intensity between the defect point and the center point of the candidate lane, and measures the degree of positioning matching of the defect point in the real-time traffic network.

8. The method for visualizing the distribution of road defects according to claim 1, characterized in that, The method further includes step S5: S5: Call the grid area encoding information in the disease annotation overlay layer structure, collect the coordinates of repeated disease records in multiple cycles in the grid, count the changes in the distribution of disease points in chronological order, filter the grids with increasing numbers in continuous cycles, connect adjacent grids in the direction of change to construct a continuous path, and obtain the periodic disease distribution path layer. The periodic disease distribution path layer includes a disease quantity increasing grid, a continuous periodic distribution trajectory, adjacent grid connection paths, and grid area coding statistics.

9. The method for visualizing the distribution of road defects according to claim 8, characterized in that, The specific steps for obtaining the periodic disease distribution path layer are as follows: S511: Call the grid area coding information in the disease annotation overlay layer structure, collect the disease repeated record coordinates of multiple periods, sort the disease coordinates in the same grid according to the time field, count the number of disease coordinates appearing in the grid in each period, calculate the difference in the number of diseases between adjacent periods, obtain the grid area coding and quantity change information with positive difference, and generate a grid set with increasing disease quantity. S512: Based on the disease quantity increasing grid set, extract the spatial positional relationship between adjacent grids, calculate the coordinate offset of the grid center point in the disease quantity increasing direction in adjacent periods, and determine that grid pairs whose distance between adjacent grid center points in the offset direction is less than the grid side length threshold are connectable units. Connect grid pairs that meet the conditions to construct a continuous grid path and obtain a continuous disease offset path coordinate chain. S513: Based on the continuous disease offset path coordinate chain, aggregate and count the period to which the grid belongs in the path, use the period with increasing disease quantity in the path as the index, combine the corresponding path chains into a layer format, mark the path flow direction in time order, and generate a periodic disease distribution path layer.

10. A road defect distribution visualization system, characterized in that, The system is used to implement the road defect distribution visualization method according to any one of claims 1-9, the system comprising: The structural edge extraction module obtains the road inspection image of the cross-sectional structural layer, extracts the set of disease edge points in the image, calculates the direction vector and curvature change between edge points, calls the structural layer number and the horizontal start and end coordinate interval in the road cross-sectional structural image, and generates a structural layer aligned boundary buffer graphic. The buffer matching module obtains the image coordinate region based on the alignment boundary buffer graphic of the structure layer, matches the coordinates of the disease points marked in the map, and determines whether the disease points fall within the buffer graphic region. If the match is successful, the structure layer number and boundary type are marked, and a disease boundary structure annotation record is generated. The structural layer annotation module calls the original positioning coordinates of the disease points in the disease boundary structural annotation record, extracts the passage priority field of the disease points and adjacent nodes and compares the priorities. If the passage priority of the disease point is lower, it is marked as a road disease, and the positioning coordinate set after the passage attribute is calibrated is obtained. The lane attribution calibration module obtains the lane guidance number based on the coordinates in the positioning coordinate set after the traffic attribute calibration, obtains the corresponding lane centerline coordinates and road width field, substitutes the disease point coordinates into the buffer range to determine whether they fall into the corresponding interval, and obtains the disease annotation overlay layer structure. The disease path construction module calls the disease annotation overlay layer structure, counts the coordinates of disease records in the same grid within the period, connects the center coordinates according to the adjacent relationship to construct a continuous path, and generates a periodic disease distribution path layer.

Images