Highway pavement abrasion state digital evaluation method and system

Abstract

The invention relates to the technical field of road recognition, in particular to a road pavement abrasion state digital evaluation method and system, and the method comprises the following steps: extracting the brightness and texture features of pavement partitions, recognizing synchronous anomalies, generating a candidate set, screening connectivity and brightness anomaly partitions, analyzing the gray scale and the structure, screening out an abnormal region, and evaluating the abrasion state of a road pavement. And evaluating a stress deviation calibration abrasion grade, and outputting a pavement repair linkage adjustment data table. According to the method, the brightness change and the texture track fluctuation are synchronously analyzed on the time axis, dynamic capture of the abrasion area is achieved, the consistency of the brightness gradient direction and the texture fitting vector and boundary connectivity judgment are combined, the recognition accuracy and coherence are improved, non-abrasion interference is eliminated, and the recognition accuracy is improved. In combination with a stress distribution trend and an original flatness curve deviation condition, an abrasion grade is accurately calibrated, a coverage range is reasonably adjusted in a repair priority sequence, abrasion identification precision and grading reliability are improved, and repair response pertinence and resource allocation rationality are enhanced.

Description

Technical Field

[0001] This invention relates to the field of road recognition technology, and in particular to a digital assessment method and system for the wear condition of highway pavement. Background Technology

[0002] Traditional digital assessment methods for road surface wear status refer to the process of classifying or quantifying the degree of road wear based on images of the road surface obtained by manual inspection or vehicle inspection devices, through manual observation or by using image grayscale histogram analysis, texture directionality measurement, and comparison of pixel differences on the road surface. It relies on preset wear classification standards and manually set image feature thresholds to complete the determination and statistics of road surface roughness, particle shedding, and density of fine cracks.

[0003] Existing technologies, in the process of wear assessment based on manual methods or simple image feature thresholds, lack temporal correlation analysis of changes in road surface brightness and texture fluctuations, which can easily lead to misjudgments under conditions of light changes or surface contamination. They also rely on manual experience in the identification of regional coherence and boundary features, resulting in unstable definition of the wear area. Furthermore, the analysis of grayscale distribution and structural features is limited in scope and cannot fully eliminate interference from non-wear factors. In the classification of wear levels, there is a lack of comprehensive correlation judgment with stress distribution and original smoothness, which can easily lead to grade assessment bias, thereby affecting the rationality of maintenance plans and the effectiveness of resource allocation. Summary of the Invention

[0004] The purpose of this invention is to overcome the shortcomings of the existing technology and to propose a digital evaluation method and system for road surface wear status.

[0005] To achieve the above objectives, the present invention adopts the following technical solution: a digital assessment method for road surface wear status, comprising the following steps: S1: Based on the detection partitions of the road surface, extract the brightness distribution curves and surface texture trajectories of the partitions, identify the synchronous abnormal intervals of brightness change amplitude and texture trajectory fluctuation on the time axis, extract the corresponding partition numbers and spatial coordinates, and generate a candidate set of wear areas. S2: Based on the candidate set of wear regions, extract the brightness gradient direction and texture fitting vector of the partition, analyze the consistency of their directions at the boundary of the partition, and combine the connectivity of the partition boundary to filter the partitions with connectivity and abnormal brightness gradient to form a suspected wear partition layer. S3: Based on the suspected wear partition layer, extract the gray-level distribution matrix and local structural features within the partition, analyze the gray-level uniformity and structural integrity, filter and match abnormal areas, and obtain a set of wear-linked partitions. S4: Based on the location number of the wear linkage zone set, analyze the road stress distribution trend corresponding to the zone, assess the degree of deviation from the original road smoothness curve, calibrate the wear level of the zone according to the deviation range, and output the road zone stress anomaly classification list.

[0006] As a further aspect of the present invention, the wear region candidate set includes brightness abrupt change partition numbers, texture fitting abnormal points, partition coordinate markers, and time series abnormality identifiers; the suspected wear partition layer includes brightness gradient abnormality partition identifiers, partition boundary connectivity units, partition boundary consistency blocks, and abnormal brightness distribution grids; the wear linkage partition set includes gray-scale abnormal continuous partitions, structurally incomplete areas, patch abrupt change overlapping areas, and linkage abnormality partition numbers; and the pavement partition stress anomaly classification list includes risk level labels, partition response deviation values, local stress anomaly indicators, and smoothness offset levels.

[0007] As a further aspect of the present invention, the step of obtaining the candidate set of wear regions specifically includes: S111: Based on the detection partitioning of the highway pavement surface, extract the brightness distribution curve and surface texture trajectory of the partition, compare and analyze the two types of data in the same partition, and obtain the trend value of brightness and texture difference. S112: Based on the brightness and texture difference trend value, identify the fluctuation amplitude value and the deviation value of the texture trajectory in the brightness change curve, superimpose the two types of values ​​in time periods, extract the time interval where the fluctuation exceeds the reference value and the deviation value exceeds the set threshold, and generate a high-frequency abnormal interval time period set. S113: For the set of high-frequency abnormal interval time periods, match the corresponding partition number and spatial coordinate information, extract the partition location where the signal occurred, and generate a candidate set of wear regions.

[0008] As a further aspect of the present invention, the step of obtaining the suspected wear-resistant partition layer specifically includes: S211: Based on the wear region candidate set, identify the brightness gradient direction and texture fitting vector of the partition, extract the projection trajectory of the two at the boundary of the partition, analyze the distribution number and aggregation degree of the boundary point in the partition, and obtain the partition boundary consistency map. S212: Based on the boundary consistency map of the partition, filter the boundary areas with a higher degree of aggregation than the average level, compare the spatial boundary of the overall road structure map, identify the continuous boundary clusters belonging to the same partition, and obtain the brightness gradient anomaly zoning within the road partition. S213: Call the brightness gradient anomaly zone within the road surface zone, perform integrated analysis on the zone boundary consistency, brightness gradient dispersion, grayscale distribution uniformity and texture fitting delay, calculate the brightness grayscale difference value, identify the wear high sensitivity response coefficient, and perform zone matching based on the response blocks in the layer to form a suspected wear zone layer.

[0009] As a further aspect of the present invention, the step of obtaining the wear-linked partition set specifically includes: S311: Based on the suspected wear partition layer, extract the gray-level distribution matrix and local structural features of the numbered partitions in the layer, align the data in the partitions with timestamps, identify the gray-level intensity fluctuation value and structural integrity offset, and obtain the local wear response feature set of the road surface. S312: Based on the local wear response feature set of the road surface, perform joint analysis on the gray-level uniformity and structural integrity within the partition, calculate the gray-level structural coupling feature value, filter the partition units of gray-level structural coupling degree in the layer, and establish a gray-level structural collaborative response spatial distribution map. S313: Call the grayscale structure collaborative response spatial distribution map, cluster the partitions in the coupled feature value layer that exceed the collaborative recognition benchmark, label the partition codes and coordinates corresponding to the continuous abnormal regions, and obtain the wear linkage partition set.

[0010] As a further aspect of the present invention, the steps for obtaining the pavement zone stress anomaly classification list are as follows: S411: Based on the position number of the wear linkage partition set, extract the partition stress distribution curve under the specified number, perform time uniform processing, identify the stress change per unit time, and obtain the partition stress abnormal change rate set. S412: Based on the set of abnormal stress change rates in the partitions, identify the stress distribution curve of the original flatness stage, compare the current stress change sequence with the reference curve, identify the stress deviation level of the partitions, extract and mark the partitions whose deviation level exceeds the warning upper limit, and obtain the set of partitions with sudden increase in deviation. S413: Based on the aforementioned deviation increase partition set, bind the deviation level value of each partition to the location number in the pavement structure spatial map, sort them according to risk level, and output a list of pavement partition stress anomaly classifications.

[0011] As a further aspect of the present invention, the method further includes step S5: S5: Call the pavement zone stress anomaly classification list, identify the corresponding number of the zone in the highway function map, retrieve the repair response unit list, compare the response level with the pavement protection priority sequence, filter the zone number that needs to adjust the response coverage, and output the pavement repair linkage adjustment data table. The road surface repair linkage adjustment data table includes the adjustment target zone number, response level adjustment parameters, protection priority comparison items, and linkage response trigger type.

[0012] As a further aspect of the present invention, the step of obtaining the road surface repair linkage adjustment data table specifically includes: S511: Call the pavement zone stress anomaly classification list, extract the zone number in the highway function map, map the zone risk level value with the area coordinate boundary, identify the zone information corresponding to the pavement protection level, and generate a pavement zone risk distribution map. S512: Based on the road zone risk distribution map, extract the repair response unit number and response level, match the zone risk level with the repair response level, identify the unit number with insufficient response coverage, and obtain the road zone response risk disconnect list. S513: Based on the pavement zone response risk disconnect list, extract the key zone numbers that need to improve the response coverage according to the level number in the pavement protection priority sequence, output the adjustment control parameters linked with the original repair unit in sequence, and output the pavement repair linkage adjustment data table.

[0013] The digital assessment system for road surface wear condition is used to execute the aforementioned digital assessment method for road surface wear condition. The system includes: The brightness monitoring module is based on the detection zones divided on the road surface. It compares the brightness change amplitude and texture trajectory fluctuation within the same time period, filters out the synchronous abnormal intervals of the two, extracts the zone number and spatial coordinates, summarizes the abnormal time period and zone number, and generates a candidate set of wear areas. Based on the wear region candidate set, the partition localization module identifies the consistency between the brightness gradient direction and the texture fitting vector, marks the partition boundary number, matches the overall road surface structure map, extracts the partition number range of the brightness gradient abnormal area, and establishes a suspected wear partition layer. Based on the suspected wear partition layer, the wear linkage module retrieves the gray-level distribution matrix and the continuous data sequence of local structural features in the region, judges the connectivity and structural integrity of gray-level anomaly boundaries, marks the partition numbers that meet the linkage threshold of both, and outputs the wear linkage partition set. The stress early warning module analyzes the stress distribution trend and the degree of deviation of the original smoothness curve of the corresponding partition based on the partition number of the wear linkage partition set, extracts the partition number of the deviation trend, completes the level identification according to the risk classification standard, and generates a list of abnormal stress classification of road partitions. The repair and optimization module, based on the pavement zone stress anomaly classification list, finds the corresponding position number of the risk level zone in the highway function map, retrieves the current repair response unit configuration list, compares the pavement protection priority with the current response level, filters the zone numbers that need to be updated, and outputs the pavement repair linkage adjustment data table.

[0014] Compared with the prior art, the advantages and positive effects of the present invention are as follows: In this invention, dynamic capture of road wear areas is achieved by synchronously analyzing brightness changes and texture trajectory fluctuations on the time axis. By combining the consistency of the brightness gradient direction and the direction of the texture fitting vector, as well as the boundary connectivity judgment, the accuracy and coherence of wear area identification are improved. Through joint analysis of grayscale uniformity and structural integrity, non-wear interference areas are effectively eliminated. By combining the stress distribution trend and the deviation of the original smoothness curve, the wear level is accurately calibrated. The coverage is reasonably adjusted in the repair priority sequence, thereby improving the overall accuracy of wear area identification and the reliability of grade judgment, and enhancing the pertinence of road repair response and the rationality of resource allocation. Attached Figure Description

[0015] 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 candidate set of wear regions in this invention. Figure 3 This is a flowchart illustrating the process of obtaining the suspected wear-resistant partition layer in this invention. Figure 4 This is a flowchart illustrating the process of obtaining the wear linkage partition set in this invention; Figure 5 This is a flowchart illustrating the process of obtaining the pavement zone stress anomaly classification list in this invention. Figure 6 This is a flowchart illustrating the process of obtaining the road surface repair linkage adjustment data table in this invention. Detailed Implementation

[0016] 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.

[0017] 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.

[0018] Example 1: Please refer to Figure 1 This invention provides a technical solution: a digital assessment method for road surface wear condition, comprising the following steps: S1: Based on the detection partitions of the road surface, extract the brightness distribution curves and surface texture trajectories of the partitions, identify the synchronous abnormal intervals of brightness change amplitude and texture trajectory fluctuation on the time axis, extract the corresponding partition numbers and spatial coordinates, and generate a candidate set of wear areas. S2: Based on the candidate set of wear regions, extract the brightness gradient direction and texture fitting vector of the partition, analyze the consistency of their directions at the boundary of the partition, and combine the connectivity of the partition boundary to filter the partitions with abnormal brightness gradients and form a suspected wear partition layer. S3: Based on the suspected wear partition layer, extract the gray-level distribution matrix and local structural features within the partition, analyze the gray-level uniformity and structural integrity, filter and match abnormal areas, and obtain a set of wear-linked partitions. S4: Based on the location number of the wear linkage zone set, analyze the stress distribution trend of the corresponding road surface in the zone, assess the degree of deviation from the original road surface smoothness curve, calibrate the wear level of the zone according to the deviation range, and output the list of road surface zone stress anomaly classification. S5: Call the list of abnormal stress levels in road zoning, identify the corresponding number of the zoning in the highway functional map, retrieve the list of repair response units, compare the response level with the road protection priority sequence, filter the zoning numbers that need to adjust the response coverage, and output the road repair linkage adjustment data table.

[0019] The candidate set of wear areas includes brightness abrupt change partition numbers, texture fitting anomaly points, partition coordinate markers, and time series anomaly identifiers. The suspected wear partition layer includes brightness gradient anomaly partition identifiers, partition boundary connectivity units, partition boundary consistency blocks, and anomaly brightness distribution grids. The wear linkage partition set includes gray-scale anomaly continuous partitions, structurally incomplete areas, patch abrupt change overlapping areas, and linkage anomaly partition numbers. The pavement partition stress anomaly classification list includes risk level labels, partition response deviation values, local stress anomaly indicators, and smoothness offset levels. The pavement repair linkage adjustment data table includes adjustment target partition numbers, response level adjustment parameters, protection priority comparison items, and linkage response trigger types.

[0020] Please see Figure 2 The specific steps for obtaining the candidate set of wear regions are as follows: S111: Based on the detection partitioning of the highway pavement surface, extract the brightness distribution curve and surface texture trajectory of the partition, compare and analyze the two types of data in the same partition, and obtain the trend value of brightness and texture difference. For the completed detection zones on the highway surface, the image data acquired by the high-resolution image acquisition equipment is preprocessed. The continuous highway surface image is divided into a series of square detection zones with sides of 2 meters. For example, a 1000-meter-long road segment will be divided into 500 independent detection zones, each with a unique number and spatial coordinates. For the detection zone numbered "DN001", the brightness distribution curve and surface texture trajectory obtained during continuous monitoring are extracted. The brightness distribution curve is obtained by recording the average pixel grayscale value at every 0.2-meter sampling point within the zone. For example, 10 uniform sampling points are selected along the length of the "DN001" zone to obtain the brightness average value sequence (grayscale value range 0-255). The surface texture trajectory is obtained by calculating the texture roughness of the local area of ​​the image at the same sampling point (e.g., based on...). The contrast eigenvalues ​​of the gray-level co-occurrence matrix are obtained and normalized to a range comparable to the brightness value (e.g., 0-255). The normalized texture roughness sequence is as follows: Comparative analysis is performed on two types of data within the same partition. Specifically, the absolute difference between the brightness value and the normalized texture value of each corresponding sampling point is calculated. For example, for the first sampling point, the absolute difference is |120-51|=69, for the second sampling point, the absolute difference is |125-61|=64, and so on, to obtain a series of absolute differences. The differences are averaged. For example, for the “DN001” partition, the sum of the absolute differences of its 10 sampling points is 69+64+74+38+8+7+22+43+58+78=461, and the trend value of brightness and texture difference is 461 / 10=46.1. This value reflects the overall deviation of brightness and texture features within the partition, as shown in Table 1. Table 1: Brightness and Texture Sampling and Difference Values ​​of DN001 Detection Zones: ; As shown in Table 1, by comparing the brightness and normalized texture data point by point within the detection partition "DN001" and calculating the average absolute difference, the trend value of the brightness and texture difference of this partition was obtained as 46.1.

[0021] S112: Based on the trend value of brightness and texture difference, identify the fluctuation amplitude value and the deviation value of texture trajectory in the brightness change curve, superimpose the two types of values ​​in time periods, extract the time interval where the fluctuation exceeds the benchmark value and the deviation value exceeds the set threshold, and generate a set of high-frequency abnormal interval time periods. Based on the continuously acquired raw brightness and texture data streams within the detection zones, the fluctuation amplitude values ​​in the brightness variation curve and the deviation values ​​of the texture trajectory are identified. The fluctuation amplitude value of the brightness variation curve is obtained by calculating the standard deviation of the brightness grayscale values ​​within each detection zone using a fixed-size sliding window (e.g., containing 4 sampling points). For example, for the "DN001" zone, the brightness fluctuation amplitude value in a certain window is 13.04. The texture trajectory deviation value is obtained by calculating the average absolute difference between the original texture value and the texture smoothing mean within the same sliding window. For example, the texture deviation value in a certain window is 5.8. The two types of values ​​are overlaid over time periods, that is, the brightness fluctuation amplitude value sequence and the texture deviation value sequence are aligned in temporal or spatial order. For each corresponding time period or spatial point, a judgment is made, and fluctuations exceeding the baseline are extracted. The time intervals where the brightness fluctuation baseline value and the deviation value exceed the set threshold are defined as follows: the brightness fluctuation baseline value is set to 8.0, which is determined by statistical analysis of brightness fluctuation data of a large number of historical healthy road sections and taking its 95th percentile; the texture deviation threshold value is set to 5.0, which is determined by analyzing expert-annotated data on the texture deviation of road surfaces of different wear levels and identifying the minimum texture deviation value at which wear begins to be significant. When judging, if the brightness fluctuation value of the current window is greater than 8.0 and the texture deviation value is greater than 5.0, the time interval is marked as abnormal, as shown in Table 2. Subsequently, the time intervals marked as abnormal are continuously detected and counted. If the number of abnormal intervals or the total length exceeds the preset proportion (for example, the total length of abnormal intervals exceeds 10% of the total length of the partition), a high-frequency abnormal interval time period set is generated. Table 2: Judgment of brightness fluctuation and texture deviation in a certain detection zone: ; As shown in Table 2, sampling points 4 and 5 were marked as abnormal intervals because they simultaneously met the conditions that the brightness fluctuation exceeded the reference value and the texture deviation exceeded the set threshold.

[0022] S113: For the time period set of high-frequency abnormal intervals, match the corresponding partition number and spatial coordinate information, extract the partition location where the signal occurred, and generate a candidate set of wear areas; For each high-frequency anomaly interval time period, the corresponding partition number and spatial coordinate information are matched one by one. For example, if the "DN001" partition detects a high-frequency anomaly interval time period within the 0 to 100 meter range of the road surface, the partition number "DN001" and its precise spatial coordinate range on the road surface map will be extracted, such as longitude 118.000000, latitude 30.000000 to longitude 118.001000, latitude 30.000000. Subsequently, the partition location where the signal occurred is precisely extracted, that is, the road surface area where the high-frequency anomaly signal actually occurred is marked. For example, if there is a high-frequency anomaly at 10 to 15 meters and 25 to 30 meters of the "DN001" partition, the coordinate information of the specific road segment is precisely extracted and recorded, not just the coordinates of the entire partition. The partition locations with high-frequency anomaly signals are collected to generate a wear area candidate set. This candidate set contains the precise geographical location and unique identifier of all road surface areas with wear.

[0023] Please see Figure 3 The specific steps to obtain the suspected worn partition layer are as follows: S211: Based on the candidate set of wear regions, identify the brightness gradient direction and texture fitting vector of the partition, extract the projection trajectory of the two at the boundary of the partition, analyze the distribution number and aggregation degree of the boundary point in the partition, and obtain the partition boundary consistency map. Based on the candidate set of worn regions, the boundary features of the candidate regions are further analyzed in detail to identify the brightness gradient direction and texture fitting vector of the partition. The brightness gradient direction is determined by calculating the rate of change of gray values ​​of pixels at the edge of the partition image, while the texture fitting vector is obtained by analyzing the dominant texture direction of local areas within the partition. Subsequently, the projection trajectories of the brightness gradient direction and texture fitting vector at the boundary of the partition are extracted. Specifically, sampling points are selected at fixed intervals (e.g., every 0.1 meters) on the boundary line between each partition in the candidate set of worn regions and adjacent partitions or background regions. The projection values ​​of the brightness gradient vector and the boundary line normal direction, as well as the projection values ​​of the texture fitting vector and the boundary line normal direction at each sampling point are calculated. For example, if the projection of the brightness gradient vector in the boundary normal direction is... If the shadow value is greater than 10 (unit: grayscale / pixel) and the projection value of the texture fitting vector in the boundary normal direction is greater than 0.2 (normalized value), then the sampling point is considered a "signal point". Then, the distribution quantity and aggregation degree of the boundary points in the partition are analyzed. The distribution quantity refers to the total number of sampling points that meet the above "signal point" conditions on a partition boundary line. For example, if there are 50 sampling points on a 5-meter-long boundary line, of which 35 are signal points, the distribution quantity is 35. The aggregation degree refers to the concentration of signal points on the boundary. It is measured by calculating the ratio of the length of the longest continuous signal point segment to the total length of all signal points. For example, if 30 of the 35 signal points form a continuous block with a total length of 3 meters, and the remaining 5 are scattered, the aggregation degree is 85.7%, thus obtaining the partition boundary consistency map.

[0024] S212: Based on the boundary consistency map of the zones, filter the boundary areas with a higher degree of aggregation than the average level, compare with the spatial boundary of the overall road structure map, identify the continuous boundary clusters belonging to the same zone, and obtain the brightness gradient anomaly zoning within the road zone. Based on the boundary consistency map, boundary areas with a aggregation degree higher than the average level are selected. The average level is calculated as the arithmetic mean of the aggregation degree values ​​of the boundary areas of all wear area candidate sets. For example, if the average aggregation degree of all analyzed boundaries is 60%, then boundary areas with an aggregation degree of 60% or higher will be selected (e.g., boundaries with an aggregation degree of 75% or 80%). These areas represent regions with relatively clear wear boundary characteristics. The selected boundary areas are compared with the spatial boundaries of the overall pavement structure map, which includes inherent structures such as lane lines, shoulder lines, and construction joints. This comparison operation is achieved through spatial overlay analysis, with the aim of eliminating or adjusting areas that highly overlap with known structural boundaries. For example, if a boundary area with a high aggregation degree overlaps with a clearly defined lane line... If there is a perfect overlap, the area is not caused by wear and will be further reviewed or excluded. Identify continuous and concentrated boundary areas belonging to the same zone. This means merging the selected boundary areas that do not highly overlap with known structural boundaries and are spatially connected to form continuous lines, and ensuring that the continuous areas are completely located within the same wear area candidate zone. For example, if there are two highly concentrated and non-structural boundary line segments in zone "DN001", with lengths of 5 meters and 7 meters respectively, and they are spatially connected end to end, they will be identified as a 12-meter-long continuous concentrated area. Finally, the brightness gradient anomaly zoning within the road surface zone is obtained. This zoning is a precisely delineated geometric boundary of the suspected wear area, which can reflect the actual distribution range of wear phenomena on the road surface.

[0025] S213: Invoke the brightness gradient anomaly zoning within the road surface zones, and perform integrated analysis on zone boundary consistency, brightness gradient dispersion, grayscale distribution uniformity, and texture fitting delay, using the following formula: ; Calculate the brightness and grayscale difference values, identify the wear high sensitivity response coefficient, and perform partition matching based on the response blocks in the layer to form a suspected wear partition layer; Where R represents the brightness-to-grayscale difference value, D represents the brightness gradient dispersion value within the road surface zone, and E represents the discretized value of the grayscale distribution uniformity within the road surface zone. The value represents the brightness of the j-th texture fitting delay measurement sample, and m is the total number of texture fitting delay measurement samples. The system calls up the brightness gradient anomaly zones within the road surface partition. The brightness gradient dispersion D represents the degree of dispersion in the direction of the brightness gradient distribution within the anomaly zone. This is obtained by calculating the standard deviation of the brightness gradient amplitude of all pixels within the zone. For example, in a certain anomaly zone, the calculated standard deviation of the brightness gradient amplitude is 25.0 (unit: grayscale / pixel, standard deviation). The grayscale distribution uniformity discretization value E represents the degree of non-uniformity in the distribution of pixel grayscale values ​​within the anomaly zone. This is obtained by calculating the standard deviation of the pixel grayscale values ​​within the zone and normalizing it. For example, if the standard deviation of the pixel grayscale values ​​within the anomaly zone is 60, it is normalized to 6.0 (e.g., divided by 10). The brightness values ​​of the texture fitting delayed measurement samples are then used. This represents the brightness values ​​at m key sampling points where the texture fitting model identifies texture anomalies (e.g., the model's predicted texture feature values ​​deviate from the actual measured values ​​by more than a preset threshold). For example, a total of m points were identified within this anomaly region. Typical regions with texture fitting delay were identified, and their corresponding average brightness values ​​were collected. (Grayscale value), the parameter is calculated using the following formula: The calculation is performed to determine the brightness-to-grayscale difference value R, where R represents the brightness-to-grayscale difference value, D represents the value of brightness gradient dispersion within the road surface zone, and E represents the discretized value of grayscale distribution uniformity within the road surface zone. The value represents the brightness of the j-th texture fitting delay measurement sample, and m is the total number of texture fitting delay measurement samples. The advantage of the formula lies in combining the sum of the absolute values ​​of the brightness gradient discretization D and the gray-level distribution uniformity discretization value E. This method captures the dual anomalies in image brightness and grayscale distribution caused by road surface wear, and simultaneously correlates the combined anomaly level with the sum of brightness from multiple texture-fitted delayed measurement samples. Multiplication further enhances the response intensity to wear signals; the numerator directly reflects the cumulative effect of wear, while the denominator... The consistency or complementarity of these two types of anomalies is measured by the difference between the absolute values ​​of D and E. When the values ​​of D and E are close, the denominator is smaller, which makes the R value increase accordingly. This indicates that the brightness gradient anomaly and the gray scale distribution anomaly are significant at the same time and corroborate each other, which greatly improves the recognition sensitivity of the wear area. Adding 1 to the denominator ensures the stability of the operation and avoids division by zero error. This precise calculation structure enables the R value to be used as a high-sensitivity response coefficient for wear, realizing high-precision recognition and quantification of road wear areas. Based on the above parameter values, the calculation is performed first: ; Then calculate the molecular weight. Then calculate the denominator. ; Substitute the calculation result into the formula: ; The results indicate that for the currently analyzed road surface anomaly zones, the brightness grayscale difference value R is approximately 24.90. This value is identified as a high-sensitivity wear response coefficient. When the R value is greater than 20, it indicates that the area has an extremely high probability of wear. This high-sensitivity response coefficient identifies response blocks in the layer. For example, areas with an R value exceeding 20 are marked as high-risk response blocks. Subsequently, based on the spatial location and shape of the response blocks, partition matching is performed, that is, the blocks are associated with predefined road surface detection partitions. For example, if a response block mainly covers most of the area of ​​the "DN001" partition, then the "DN001" partition will be identified as a suspected wear partition, ultimately forming a suspected wear partition layer. This layer displays all areas on the road surface suspected of having wear in an intuitive geographic information form.

[0026] Please see Figure 4 The specific steps for obtaining the wear-related partition set are as follows: S311: Based on the suspected wear partition layer, extract the gray-level distribution matrix and local structural features of the numbered partitions in the layer, align the data in the partitions with timestamps, identify the gray-level intensity fluctuation value and structural integrity offset, and obtain the local wear response feature set of the road surface. Based on a suspected wear-affected partitioning layer, the gray-level distribution matrix is ​​a two-dimensional matrix generated by statistically analyzing the gray-level values ​​of all pixels within each suspected wear-affected partition. Local structural features are obtained by extracting features such as edge intensity, corner density, and line direction from the image within the partition. For example, edges are detected and the number of edge pixels per unit area is counted. Subsequently, the data within the partition is time-stamp aligned, that is, the gray-level distribution matrix and local structural feature data within the same partition collected at different time points are precisely synchronized in time. The gray-level intensity fluctuation value and structural integrity offset are identified. The gray-level intensity fluctuation value is calculated by measuring the gray-level distribution matrix and local structural feature data of the same partition at different time points. The variance of the corresponding pixel grayscale values ​​in the grayscale distribution matrix is ​​obtained. The structural integrity offset is calculated by comparing the differences in local structural features collected at different time points. For example, by comparing the number of matching points and their average distance of the structural feature descriptor at different time points, the offset is higher if the number of matching points is small and the distance is large. Finally, the local wear response feature set of the road surface is obtained. For example, for the "DN001" partition, this feature set contains {(time point T1, grayscale intensity fluctuation value: 5.2, structural integrity offset: 0.15), (time point T2, grayscale intensity fluctuation value: 7.8, structural integrity offset: 0.25)}.

[0027] S312: Based on the local wear response characteristic set of the road surface, a joint analysis of gray-level uniformity and structural integrity within the zone is performed, using the following formula: ; Calculate the gray-scale structure coupling characteristic value, filter the partition units of gray-scale structure coupling degree in the layer, and establish a gray-scale structure cooperative response spatial distribution map; Where H represents the gray-scale structure coupling characteristic value, This represents the gray value of the i-th region unit. The mean of the grayscale values. This represents the structural integrity value of the i-th region unit. The mean value represents the structural integrity value, and N represents the total number of region units; Based on the local wear response characteristic set of the road surface, the gray-scale structure coupling characteristic value H is calculated. The advantage of the formula is that it considers the gray-scale values ​​simultaneously. The degree of deviation and structural integrity value The formula calculates the degree of deviation from the average value and takes the square root of the sum of squares. This effectively quantifies the synergistic changes in grayscale and structural features caused by pavement wear. When pavement wears, its grayscale value becomes uneven, and at the same time, the structure is damaged or deformed. These two types of features do not change independently, but are related and deteriorate synchronously. This formula can sensitively capture this coupled deterioration phenomenon by calculating the sum of squares of the deviations of each from its average value and averaging them. That is, when both grayscale changes and structural changes are significant and occur synergistically, the H value will increase significantly, thus providing a comprehensive and accurate indicator to assess the severity of wear. To calculate H, a suspected wear partition is defined. Each tiny region unit has a grayscale value. (0-255) and structural integrity value (0-1, 1 represents complete) as follows: Unit 1: , Unit 2: , Unit 3: , Unit 4: , Unit 5: , ; First, calculate the mean of the grayscale values. and the mean of structural integrity values : ; ; Then calculate each unit. and : Unit 1: , ; Unit 2: , ; Unit 3: , ; Unit 4: , ; Unit 5: , ; Calculate the sum of all terms: ; Substitute into the formula to calculate H: ; The results indicate that the gray-scale structural coupling characteristic value within the partition is approximately 13.68. This value reflects the degree of synergistic degradation between gray-scale uniformity and structural integrity. Subsequently, partition units in the filter layer whose gray-scale structural coupling exceeds the synergistic identification benchmark are identified. The synergistic identification benchmark value can be set by statistically analyzing the H values ​​of a large number of normal road surface areas and taking the 95th percentile as the threshold. For example, by calculating the H values ​​of 1000 normal road surface units, it was found that they are generally lower than 10. Therefore, the synergistic identification benchmark value is set to 10. If the calculated H value (e.g., 13.68) is greater than this benchmark value of 10, the partition unit is considered to have an abnormal gray-scale structural synergistic response. A spatial distribution map of gray-scale structural synergistic response is established. This map visualizes all partition units on the road surface whose H values ​​exceed the benchmark.

[0028] S313: Call the grayscale structure co-response spatial distribution map, cluster the partitions in the coupled feature value layer that exceed the co-identification benchmark, label the partition codes and coordinates corresponding to the continuous abnormal areas, and obtain the wear linkage partition set; The grayscale structure collaborative response spatial distribution map is invoked to cluster the partitions in the coupling feature value layer that exceed the collaborative identification benchmark. The clustering operation is achieved by analyzing the coupling feature values ​​and spatial connectivity of adjacent partitions. For example, by setting the minimum number of neighborhood points and the maximum neighborhood radius, multiple partition units whose coupling feature values ​​are all greater than the collaborative identification benchmark value (e.g., 10) and are spatially close to each other are clustered into a continuous wear region. For example, if the coupling feature values ​​of partitions "DN001", "DN002", and "DN003" are 13.68, 14.12, and 12.95, respectively, and they are continuously adjacent on the road surface, then these three partitions will be clustered into a larger continuous anomaly region. Subsequently, the partition codes and coordinates corresponding to the continuous anomaly region are labeled, and the boundary information of the wear region formed by clustering is accurately recorded. Finally, a wear linkage partition set is obtained, which contains all spatially continuous wear regions identified by the grayscale structure collaborative response mechanism.

[0029] Please see Figure 5 The specific steps for obtaining the list of abnormal stress levels in road sections are as follows: S411: Based on the location number of the wear linkage zone set, extract the stress distribution curve of the zone under the specified number, perform time uniform processing, identify the stress change per unit time, and obtain the zone stress abnormal change rate set. Based on the location number of the wear linkage partition set, the stress distribution curve of the partition under the specified number is first extracted. For example, for the linkage partition numbered "DN001-DN003" in the set, the real-time stress data collected by embedded sensors or dynamic weighing in the past year is retrieved to form a stress value sequence per unit time (e.g., per hour). Time unification processing is then performed, that is, data from different sensors or different acquisition frequencies are resampled or interpolated to ensure that all stress data maintain a consistent step size on the time axis. Subsequently, the stress change per unit time is identified, specifically by calculating the difference between stress values ​​at adjacent time points. For example, if the stress value of a certain partition is 1.2 MPa at time T0 and 1.5 MPa at time T1, then the stress change per unit time is 0.3 MPa. Finally, the partition stress anomaly change rate set is obtained.

[0030] S412: Based on the set of abnormal stress change rates in the partitions, identify the stress distribution curve of the original flatness stage, compare the current stress change sequence with the reference curve, identify the stress deviation level of the partitions, extract and mark the partitions whose deviation level exceeds the warning upper limit, and obtain the set of partitions with sudden increase in deviation. Based on the set of abnormal stress change rates in different zones, the stress distribution curve of the original smoothness stage is identified. This curve represents the normal stress response pattern of the pavement when no significant wear or damage has occurred. For example, stress monitoring data within three months after the pavement is newly built or after the most recent major overhaul is retrieved and its average stress change curve is constructed as a reference. Subsequently, the current stress change sequence is compared with the reference curve. By calculating the root mean square error or correlation coefficient between the current zone stress change sequence and the original smoothness stage reference curve, the degree of deviation is quantified, and the zone stress deviation level is identified. The deviation levels are divided into: slight deviation (root mean square error less than 0.2 MPa), moderate deviation (root mean square error between 0.2 MPa and 0.6 MPa), and severe deviation (root mean square error greater than or equal to 0.6 MPa). Zones with deviation levels exceeding the upper warning limit are extracted and marked. The upper warning limit is set to a root mean square error of 0.6 MPa. For example, if the root mean square error of a zone is 0.7 MPa, then the zone is marked as having a deviation level exceeding the upper warning limit. Finally, the set of zones with sudden increases in deviation is obtained.

[0031] S413: Based on the deviation increase partition set, bind the deviation level value of each partition to the location number in the pavement structure space map, sort them according to risk level, and output the pavement partition stress anomaly classification list. Based on the deviation abrupt increase partition set, the deviation level value of each partition is bound to the location number in the pavement structure spatial map. The pavement structure spatial map provides the precise geographical location and topological relationship of each pavement partition. The binding operation involves attaching the deviation level as an attribute to the corresponding partition object in the pavement structure spatial map. Subsequently, all partitions in the deviation abrupt increase partition set are sorted according to their stress deviation level from high to low, based on their risk level. The higher the level value, the higher the risk ranking. For example, if the root mean square error of partition "DN005" is 0.9 MPa, then... The root mean square error of “N001” is 0.7MPa. Therefore, the risk level of “DN005” is higher than that of “DN001”. Output the pavement zone stress anomaly classification list. For example, the pavement zone stress anomaly classification list includes: {(zone number: DN005, spatial coordinates: [mileage 120m, mileage 130m], risk level: severe deviation, root mean square error: 0.9MPa), (zone number: DN001, spatial coordinates: [mileage 10m, mileage 15m], risk level: severe deviation, root mean square error: 0.7MPa)}.

[0032] Please see Figure 6 The specific steps for obtaining the road surface repair linkage adjustment data table are as follows: S511: Call the list of stress anomalies in road zones, extract the zone number in the highway functional map, map the zone risk level value to the regional coordinate boundary, identify the zone information corresponding to the road protection level, and generate a road zone risk distribution map. The process involves calling the pavement zoning stress anomaly classification list. First, the zoning numbers within the highway functional map are extracted. The highway functional map contains information such as highway grade, traffic volume, and design life. The extraction operation associates the zoning numbers in the stress anomaly classification list with the corresponding zoning numbers in the highway functional map. Subsequently, the zoning risk level values ​​are mapped to the regional coordinate boundaries. This involves spatially mapping the risk level values ​​in the stress anomaly classification list to the precise geographic coordinate boundaries of the zoning, making it visually displayable in the geographic information system. This identifies the zoning information corresponding to the pavement protection level. The pavement protection level is preset based on factors such as highway grade and traffic volume in the highway functional map. For example, a Class I highway corresponds to the highest protection level. The mapped zoning risk level is compared with the corresponding pavement protection level to identify pavement zoning that requires key protection. For example, if zoning "DN001" is a Class I highway and its stress risk level is "severe deviation," it is identified as an anomaly zoning with a high protection level. Finally, a pavement zoning risk distribution map is generated.

[0033] S512: Based on the pavement zone risk distribution map, extract the repair response unit number and response level, match the zone risk level with the repair response level, identify the unit number with insufficient response coverage, and obtain the pavement zone response risk disconnect list. Based on the road surface zone risk distribution map, the repair response unit number and response level are first extracted. The repair response unit is a pre-defined repair scheme module for road sections with different risk levels. Each module has a unique number and corresponding response level. For example, "RRU001" corresponds to "light maintenance" and "RRU002" corresponds to "moderate repair". The extraction operation selects the response unit that matches the protection level of each zone in the risk distribution map from the predefined repair scheme library. The risk level of the zone is matched with the repair response level. By comparing the actual risk level of the zone in the risk distribution map with the repair level that the currently recommended repair response unit can cover, for example, if the actual risk level of zone "DN001" is 0.7 MPa (severe deviation), while the currently matched repair unit "RRU002" can only cover 0.5 MPa (moderate deviation), there is insufficient response. The unit number with insufficient response coverage is identified, that is, those road zones whose actual risk level is higher than the coverage capacity of the existing repair response units are identified, and finally a list of road surface zone response risk disconnection is obtained.

[0034] S513: Based on the list of risk disconnection in pavement zone response, extract the key zone numbers that need to be improved in response coverage according to the level number in the pavement protection priority sequence, output the adjustment control parameters linked with the original repair unit in sequence, and output the pavement repair linkage adjustment data table. Based on the pavement zoning response risk disconnect list, and according to the level number in the pavement protection priority sequence, the key zoning numbers that need to improve response coverage are extracted. The pavement protection priority sequence is a predefined list of repair priorities determined based on factors such as highway importance, traffic volume, and cost-effectiveness. By comparing the zoning in the disconnect list, the zoning that needs to be focused on is extracted in order from high priority to low priority. Then, the adjustment control parameters linked with the original repair unit are output in sequence. For each key zoning that needs to improve response coverage, adjustment suggestions for its current matching repair unit are output. For example, for the repair unit "RRU002" of zoning "DN001", the suggested adjustment control parameters include: upgrading the repair material grade, increasing the construction depth, or expanding single-point repair to regional repair. Finally, the pavement repair linkage adjustment data table is output. The pavement repair linkage adjustment data table is used to integrate and match high-risk or insufficiently responsive pavement zones with corresponding repair plans and adjustment measures. Based on the risk level of the zone, highway protection priority, and existing repair unit capabilities, it clearly indicates which zones need to increase repair coverage and provides specific adjustment suggestions, such as upgrading repair material grades, increasing construction depth, or expanding from local repair to regional repair. This provides decision-making departments with accurate and executable repair instructions, enabling efficient allocation and linkage management of maintenance resources.

[0035] The digital assessment system for road surface wear condition is used to execute the above-mentioned digital assessment method for road surface wear condition. The system includes: The brightness monitoring module is based on the detection zones divided on the road surface. It compares the brightness change amplitude and texture trajectory fluctuation within the same time period, filters out the synchronous abnormal intervals of the two, extracts the zone number and spatial coordinates, summarizes the abnormal time period and zone number, and generates a candidate set of wear areas. The partition localization module identifies the consistency between the brightness gradient direction and the texture fitting vector based on the wear area candidate set, marks the partition boundary number, matches the overall road surface structure map, extracts the partition number range of the brightness gradient abnormal area, and establishes a suspected wear partition layer. The wear linkage module is based on the suspected wear partition layer. It retrieves the continuous data sequence of gray-level distribution matrix and local structural features in the region, judges the connectivity and structural integrity of gray-level anomaly boundaries, marks the partition number that meets the linkage threshold of both, and outputs the wear linkage partition set. The stress early warning module analyzes the stress distribution trend and the degree of deviation of the original smoothness curve of the corresponding zone based on the zone number of the wear linkage zone set, extracts the zone number of the deviation trend, completes the level identification according to the risk classification standard, and generates a list of abnormal stress classification of road zone. The repair and optimization module is based on the list of stress anomalies in pavement zones. It finds the corresponding position number of the risk level zone in the highway function map, retrieves the current repair response unit configuration list, compares the pavement protection priority with the current response level, filters the zone numbers that need to be updated, and outputs the pavement repair linkage adjustment data table.

[0036] 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 digital assessment method for road surface wear condition, characterized in that, Includes the following steps: S1: Based on the detection partitions of the road surface, extract the brightness distribution curves and surface texture trajectories of the partitions, identify the synchronous abnormal intervals of brightness change amplitude and texture trajectory fluctuation on the time axis, extract the corresponding partition numbers and spatial coordinates, and generate a candidate set of wear areas. S2: Based on the candidate set of wear regions, extract the brightness gradient direction and texture fitting vector of the partition, analyze the consistency of their directions at the boundary of the partition, and combine the connectivity of the partition boundary to filter the partitions with connectivity and abnormal brightness gradient to form a suspected wear partition layer. S3: Based on the suspected wear partition layer, extract the gray-level distribution matrix and local structural features within the partition, analyze the gray-level uniformity and structural integrity, filter and match abnormal areas, and obtain a set of wear-linked partitions. S4: Based on the location number of the wear linkage zone set, analyze the road stress distribution trend corresponding to the zone, assess the degree of deviation from the original road smoothness curve, calibrate the wear level of the zone according to the deviation range, and output the road zone stress anomaly classification list.

2. The digital assessment method for road surface wear condition according to claim 1, characterized in that, The wear region candidate set includes brightness abrupt change partition numbers, texture fitting abnormal points, partition coordinate markers, and time series abnormality identifiers. The suspected wear partition layer includes brightness gradient abnormality partition identifiers, partition boundary connectivity units, partition boundary consistency blocks, and abnormal brightness distribution grids. The wear linkage partition set includes gray-scale abnormal continuous partitions, structurally incomplete areas, patch abrupt change overlapping areas, and linkage abnormality partition numbers. The pavement partition stress anomaly classification list includes risk level labels, partition response deviation values, local stress anomaly indicators, and smoothness offset levels.

3. The digital assessment method for road surface wear condition according to claim 1, characterized in that, The specific steps for obtaining the candidate set of wear regions are as follows: S111: Based on the detection partitioning of the highway pavement surface, extract the brightness distribution curve and surface texture trajectory of the partition, compare and analyze the two types of data in the same partition, and obtain the trend value of brightness and texture difference. S112: Based on the brightness and texture difference trend value, identify the fluctuation amplitude value and the deviation value of the texture trajectory in the brightness change curve, superimpose the two types of values ​​in time periods, extract the time interval where the fluctuation exceeds the reference value and the deviation value exceeds the set threshold, and generate a high-frequency abnormal interval time period set. S113: For the set of high-frequency abnormal interval time periods, match the corresponding partition number and spatial coordinate information, extract the partition location where the signal occurred, and generate a candidate set of wear regions.

4. The digital assessment method for road surface wear condition according to claim 3, characterized in that, The specific steps for obtaining the suspected wear-resistant partition layer are as follows: S211: Based on the wear region candidate set, identify the brightness gradient direction and texture fitting vector of the partition, extract the projection trajectory of the two at the boundary of the partition, analyze the distribution number and aggregation degree of the boundary point in the partition, and obtain the partition boundary consistency map. S212: Based on the boundary consistency map of the partition, filter the boundary areas with a higher degree of aggregation than the average level, compare the spatial boundary of the overall road structure map, identify the continuous boundary clusters belonging to the same partition, and obtain the brightness gradient anomaly zoning within the road partition. S213: Call the brightness gradient anomaly zone within the road surface zone, perform integrated analysis on the zone boundary consistency, brightness gradient dispersion, grayscale distribution uniformity and texture fitting delay, calculate the brightness grayscale difference value, identify the wear high sensitivity response coefficient, and perform zone matching based on the response blocks in the layer to form a suspected wear zone layer.

5. The digital assessment method for road surface wear condition according to claim 4, characterized in that, The specific steps for obtaining the wear-related partition set are as follows: S311: Based on the suspected wear partition layer, extract the gray-level distribution matrix and local structural features of the numbered partitions in the layer, align the data in the partitions with timestamps, identify the gray-level intensity fluctuation value and structural integrity offset, and obtain the local wear response feature set of the road surface. S312: Based on the local wear response feature set of the road surface, perform joint analysis on the gray-level uniformity and structural integrity within the partition, calculate the gray-level structural coupling feature value, filter the partition units of gray-level structural coupling degree in the layer, and establish a gray-level structural collaborative response spatial distribution map. S313: Call the grayscale structure collaborative response spatial distribution map, cluster the partitions in the coupled feature value layer that exceed the collaborative recognition benchmark, label the partition codes and coordinates corresponding to the continuous abnormal regions, and obtain the wear linkage partition set.

6. The digital assessment method for road surface wear condition according to claim 5, characterized in that, The specific steps for obtaining the pavement zone stress anomaly classification list are as follows: S411: Based on the position number of the wear linkage partition set, extract the partition stress distribution curve under the specified number, perform time uniform processing, identify the stress change per unit time, and obtain the partition stress abnormal change rate set. S412: Based on the set of abnormal stress change rates in the partitions, identify the stress distribution curve of the original flatness stage, compare the current stress change sequence with the reference curve, identify the stress deviation level of the partitions, extract and mark the partitions whose deviation level exceeds the warning upper limit, and obtain the set of partitions with sudden increase in deviation. S413: Based on the aforementioned deviation increase partition set, bind the deviation level value of each partition to the location number in the pavement structure spatial map, sort them according to risk level, and output a list of pavement partition stress anomaly classifications.

7. The digital assessment method for road surface wear condition according to claim 1, characterized in that, The method also includes step S5: S5: Call the pavement zone stress anomaly classification list, identify the corresponding number of the zone in the highway function map, retrieve the repair response unit list, compare the response level with the pavement protection priority sequence, filter the zone number that needs to adjust the response coverage, and output the pavement repair linkage adjustment data table. The road surface repair linkage adjustment data table includes the adjustment target zone number, response level adjustment parameters, protection priority comparison items, and linkage response trigger type.

8. The digital assessment method for road surface wear condition according to claim 7, characterized in that, The specific steps for obtaining the road surface repair linkage adjustment data table are as follows: S511: Call the pavement zone stress anomaly classification list, extract the zone number in the highway function map, map the zone risk level value with the area coordinate boundary, identify the zone information corresponding to the pavement protection level, and generate a pavement zone risk distribution map. S512: Based on the road zone risk distribution map, extract the repair response unit number and response level, match the zone risk level with the repair response level, identify the unit number with insufficient response coverage, and obtain the road zone response risk disconnect list. S513: Based on the pavement zone response risk disconnect list, extract the key zone numbers that need to improve the response coverage according to the level number in the pavement protection priority sequence, output the adjustment control parameters linked with the original repair unit in sequence, and output the pavement repair linkage adjustment data table.

9. A digital assessment system for road surface wear condition, characterized in that, The system is used to implement the digital assessment method for road surface wear status according to any one of claims 1-8, and the system includes: The brightness monitoring module is based on the detection zones divided on the road surface. It compares the brightness change amplitude and texture trajectory fluctuation within the same time period, filters out the synchronous abnormal intervals of the two, extracts the zone number and spatial coordinates, summarizes the abnormal time period and zone number, and generates a candidate set of wear areas. Based on the wear region candidate set, the partition localization module identifies the consistency between the brightness gradient direction and the texture fitting vector, marks the partition boundary number, matches the overall road surface structure map, extracts the partition number range of the brightness gradient abnormal area, and establishes a suspected wear partition layer. Based on the suspected wear partition layer, the wear linkage module retrieves the gray-level distribution matrix and the continuous data sequence of local structural features in the region, judges the connectivity and structural integrity of gray-level anomaly boundaries, marks the partition numbers that meet the linkage threshold of both, and outputs the wear linkage partition set. The stress early warning module analyzes the stress distribution trend and the degree of deviation of the original smoothness curve of the corresponding partition based on the partition number of the wear linkage partition set, extracts the partition number of the deviation trend, completes the level identification according to the risk classification standard, and generates a list of abnormal stress classification of road partitions. The repair and optimization module, based on the pavement zone stress anomaly classification list, finds the corresponding position number of the risk level zone in the highway function map, retrieves the current repair response unit configuration list, compares the pavement protection priority with the current response level, filters the zone numbers that need to be updated, and outputs the pavement repair linkage adjustment data table.

Images