Coarse aggregate form acquisition and characterization method based on 3D point cloud

Abstract

The invention relates to the technical field of morphological characterization, in particular to a 3D point cloud-based coarse aggregate morphological acquisition and characterization method, which comprises the following steps of: acquiring a spatial point set of a 3D point cloud sampling fragment, dividing the sampling fragment according to grids, performing distance aggregation according to local geometric consistency, and identifying a boundary region. And analyzing spatial changes of the coarse aggregate boundary in different environments, recording point cloud density and morphological changes, analyzing an evolution path of coarse aggregate particles, and identifying an evolution trend to obtain a coarse aggregate evolution trend. According to the method, the spatial change amplitude and the volume change amplitude of the particles under the differentiated environment conditions can be more effectively analyzed through form recognition and performance prediction of the coarse aggregate, and through real-time tracking of the form change of the particles and combination of analysis of physical characteristics and surface characteristics, under the influence of different loading times and environment factors, the particle size can be accurately measured. The performance change and evolution trend of the coarse aggregate are accurately predicted, and optimization of material utilization and improvement of structural safety are ensured.

Description

Technical Field

[0001] The present invention relates to the technical field of morphological characterization, and in particular to a method for collecting and characterizing coarse aggregate morphology based on 3D point clouds. Background Art

[0002] The field of morphological characterization involves the precise analysis and evaluation of the shape, size, and surface characteristics of materials, particularly mineral aggregates. The core of this technical field is to obtain morphological data of materials through various measurement and analysis techniques in order to assess their suitability and performance in specific applications. This field covers the evaluation of morphological characteristics from macroscopic to microscopic levels, including the overall shape, angular details, and surface texture of aggregates. Technical methods can be direct physical measurements or performed through image analysis and computational models. The main purpose is to improve the efficiency of material application and the accuracy of performance predictions, especially in the application of construction and civil engineering materials.

[0003] Among them, the 3D point cloud coarse aggregate morphology acquisition and characterization method refers to the use of 3D scanning technology to capture the external morphology and details of coarse aggregate. The technical matters targeted by this patent subject include the use of 3D scanning equipment to perform high-precision spatial data acquisition on the surface of coarse aggregate, and the processing of the acquired point cloud data by software to form a three-dimensional model of the aggregate. This method allows for detailed analysis of the size, shape and surface texture of each aggregate particle, and can provide more comprehensive and accurate data compared to traditional methods based on projection or two-dimensional image analysis. This technology is mainly achieved through the acquisition of 3D point cloud data and subsequent data processing steps, focusing on accurate data capture and efficient data processing strategies.

[0004] Existing technologies for capturing and analyzing coarse aggregate morphology rely primarily on direct physical measurement or two-dimensional image analysis, which is limited in accurately capturing three-dimensional morphological features, particularly complex forms and details. Traditional methods are unable to effectively track and predict material performance changes under varying environmental conditions in real time, limiting their dynamic evaluation and optimized application. These deficiencies can lead to increased safety factors or material waste during project implementation due to inaccurate material property predictions. For example, in bridge construction, the heterogeneity of coarse aggregate can affect the overall quality and durability of concrete, impacting the long-term stability and safety of the structure. Summary of the Invention

[0005] In order to solve the problem of limitations in complex morphology and details in existing technologies, it is impossible to effectively track and predict the performance changes of materials under different environmental conditions in real time, which limits the dynamic evaluation and optimization application of materials. The shortcomings lead to an increase in the safety factor or material waste in the design process due to inaccurate prediction of material properties during project implementation. In bridge construction, the unevenness of coarse aggregate will affect the overall quality and durability of concrete, and affect the long-term stability and safety of the structure. The embodiment of the present invention provides a coarse aggregate morphology collection and characterization method based on 3D point cloud. The technical solution is as follows: On the one hand, a method for collecting and characterizing coarse aggregate morphology based on 3D point clouds is provided, which includes: S1: Obtain a set of spatial points in a 3D point cloud sampling segment, divide each sampling segment into a grid, calculate the distance difference between adjacent points in the grid, identify the point change amplitude per unit time, analyze the regional morphology, and generate the sampling density adjustment result; S2: Based on the point cloud data in the sampling density adjustment result, distance aggregation is performed according to local geometric consistency, the unit vector angle between adjacent points in the curvature gradient direction is calculated, and the angle change of three consecutive points is analyzed. If the angle changes suddenly, it is identified as a boundary identification area, and a coarse aggregate boundary identification result is generated; S3: Using the coarse aggregate boundary identification results, analyze the spatial changes of the coarse aggregate boundary under differentiated environmental conditions, analyze the volume change amplitude of each particle, and combine the spatial coordinate displacement data of the change area to generate the coarse aggregate performance stability evaluation results; S4: calling the coarse aggregate performance stability evaluation result, recording the point cloud density and morphological changes before and after adjustment in each area, analyzing the regional changes according to the magnitude of the change, combining the physical properties and surface characteristics of the particles, and tracking the morphological changes of the particles in real time to generate the coarse aggregate dynamic morphological adjustment result.

[0006] As a further solution of the present invention, the sampling density adjustment results include grid point spacing, point change rate, and regional density gradient; the coarse aggregate boundary identification results include boundary curvature difference, angle mutation position, and boundary continuity; the coarse aggregate performance stability evaluation results include particle displacement vector, particle spatial deformation, and environmental adaptability difference information; the coarse aggregate dynamic morphology adjustment results include regional density change rate, particle surface texture change, and particle dynamic stability.

[0007] As a further solution of the present invention, the step of obtaining the sampling density adjustment result is specifically as follows: S101: Obtain a set of spatial points in a 3D point cloud sampling segment, divide each sampling segment into cubic grid units along the coordinate axis direction, extract spatial point groups within the grid units, calculate the Euclidean distance set of each group of points in the three-dimensional coordinates, and perform statistics on the distance differences between two adjacent points in the distance set to generate a grid point distance difference set; S102: Based on the grid point distance difference set, sequentially compare the distance difference values ​​of the same grid unit in the differentiated time series sampling segments, analyze the offset interval between the distance difference value and the stability within the grid, and obtain the unit time point distance offset strength value; S103: Divide the grid unit into a density stable area and a density change area according to the offset intensity value of the unit time point, calculate the mean density of the sampling points based on the number and volume of spatial points in the density change area, and obtain a sampling density adjustment result.

[0008] As a further solution of the present invention, the steps for obtaining the coarse aggregate boundary recognition result are specifically as follows: S201: calling the point cloud data in the sampling density adjustment result, performing geometric consistency judgment on the local neighborhood of each point based on the spatial distance value and the difference in the three-dimensional normal vector direction between the point and the points in the neighborhood, merging the point sets with consistent features into an aggregation group, and analyzing the curvature gradient direction between adjacent points in the aggregation group to obtain the curvature gradient direction angle value; S202: According to the angle value of the curvature gradient direction, three consecutive points in each group are selected as a combination unit, and the numerical difference between the two angle values ​​is judged in turn. If the angle difference exceeds the boundary mutation judgment threshold, the real-time middle point is marked as the boundary identification point, and the spatial position coordinates and normal vector directions corresponding to the marked points are summarized to identify the boundary spatial feature distribution of the point cloud area and obtain the coarse aggregate boundary identification result.

[0009] As a further solution of the present invention, the steps for obtaining the coarse aggregate performance stability evaluation result are specifically as follows: S301: calling the boundary space feature distribution data in the coarse aggregate boundary recognition result, combining multiple groups of boundary point cloud samples under differentiated environmental conditions, referring to the spatial coordinates of the corresponding boundary points, performing a difference comparison on the position changes of the same particle under differentiated conditions, calculating the displacement distance of the particle boundary point in three-dimensional space, and obtaining the coarse aggregate boundary space displacement data; S302: performing volume reconstruction processing on the boundary point displacement in the coarse aggregate boundary spatial displacement data, extracting the boundary point set of the particle envelope in the three-dimensional coordinate system, constructing a spatial bounding box through the boundary point set, and performing difference calculation on the volume values ​​of the same particle under different conditions to obtain the particle volume change amplitude; S303: Based on the particle volume change amplitude and the spatial position of the particle center point, identifying an area where the volume change amplitude is greater than the particle stability fluctuation baseline value, extracting the three-dimensional coordinate displacement vector of the particle in the area under differentiated environmental conditions, evaluating the directional consistency of the displacement vector, and obtaining coordinate displacement data of the changed area; S304: Based on the coordinate displacement data of the change area, the displacement length value, change direction angle and volume change of each particle under differentiated environmental conditions are combined and calculated, and the particle stability performance is judged by the numerical change trend to obtain the coarse aggregate performance stability evaluation result.

[0010] As a further solution of the present invention, the formula for calculating the displacement distance of the particle boundary point in three-dimensional space is as follows: ; in, Representative particles The displacement distance in three-dimensional space, Representative particles In the Under different conditions coordinate, Representative particles In the The initial state under different conditions coordinate, Representative particles In the Under different conditions coordinate, Representative particles In the The initial state under different conditions coordinate, Representative particles In the Under different conditions coordinate, Representative particles In the The initial state under different conditions coordinate, represents the number of differentiation condition groups.

[0011] As a further solution of the present invention, the steps for obtaining the dynamic morphology adjustment result of the coarse aggregate are specifically as follows: S401: calling the spatial position identifiers of the particles in the coarse aggregate performance stability evaluation results, extracting the point cloud data before and after adjustment for the area where each particle is located, calculating the number of points per unit volume in the same coordinate system, analyzing the distance of regional morphological changes, and generating point cloud density and morphological change amount; S402: Based on the point cloud density and morphological change, matching and analyzing the change value of the region with the density value, hardness value, and surface roughness of the particle material parameters, identifying the density change trend, and obtaining a sensitive region identification mark group; S403: Based on the sensitive area identification mark group, the boundary coordinate change trajectory of the particles in the marked area in the continuous time frame is extracted, the coordinate change trend and velocity distribution value of the boundary point of each particle in the three-dimensional direction are tracked, the dynamic morphological state of the particles is classified and identified, and the dynamic morphological adjustment result of the coarse aggregate is obtained.

[0012] As a further solution of the present invention, the formula for analyzing regional morphological changes is as follows: ; in, Representative Region and The amount of morphological change in the region, represents the volume of the region, Represents the total number of point clouds involved in the calculation, Representative Region No. The coordinates of the particle, Representative Region No. The coordinates of the particle, Representatives in the The number of point clouds extracted at the particle position, is the adjustment index.

[0013] As a further solution of the present invention, the method further includes step S5: S5: using the coarse aggregate dynamic morphology adjustment results, analyzing the evolution path of coarse aggregate particles, identifying the evolution trend under differentiated environments, and linearly classifying the change trend with reference to loading time and environmental factors to obtain the coarse aggregate evolution trend; The coarse aggregate evolution trend includes evolution path pattern, environmental factor sensitivity, and loading time correlation.

[0014] As a further solution of the present invention, the steps for obtaining the evolution trend of the coarse aggregate are specifically as follows: S501: calling the boundary coordinate change trajectory of the particles in the coarse aggregate dynamic morphology adjustment result in the continuous time frame, connecting the paths in time sequence according to the particle identification, analyzing the morphology change direction of the particles through the continuity of the change direction and the fluctuation range of the speed, and generating a particle morphology evolution path set; S502: Based on the particle morphology evolution path set, for the particle path characteristics under differentiated environments, with reference to the loading time data and environmental factors of the corresponding time period, curve fitting processing is performed on the speed change trend of each type of path, and the paths are linearly classified through the correspondence between the three-dimensional coordinate change rate and the environmental response to obtain the coarse aggregate evolution trend.

[0015] The beneficial effects brought about by the technical solution provided by the embodiment of the present invention include at least: By acquiring a collection of spatial points and dividing them into a grid, calculating the distance differences between points within the grid, and analyzing the unit vector angle along the curvature gradient through distance aggregation, more refined coarse aggregate morphology identification and performance prediction are provided. This refined management allows the subtle differences of each particle to be accurately captured, enabling more effective analysis of the spatial and volumetric changes of particles under differentiated environmental conditions. By tracking particle morphological changes in real time and combining analysis of physical properties and surface characteristics, it is possible to accurately predict the performance changes and evolution trends of coarse aggregate under different loading times and environmental factors. This greatly improves the application efficiency of coarse aggregate in construction and civil engineering and the accuracy of performance prediction, ensuring optimal material utilization and improved structural safety. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 It is a schematic diagram of the workflow of the present invention; DETAILED DESCRIPTION

[0017] The technical solution of the present invention is described below in conjunction with the accompanying drawings.

[0018] In the embodiments of the present invention, words such as "exemplarily" and "for example" are used to indicate examples, illustrations, or explanations. Any embodiment or design described as an "exemplary" in the present invention should not be interpreted as being preferred or advantageous over other embodiments or designs. Rather, the use of the word "exemplary" is intended to present concepts in a concrete manner. Furthermore, in the embodiments of the present invention, "and / or" can mean both or either of the two.

[0019] In order to make the technical problems, technical solutions and advantages to be solved by the present invention clearer, a detailed description will be given below with reference to the accompanying drawings and specific embodiments.

[0020] See also Figure 1 The embodiment of the present invention provides a method for collecting and characterizing the morphology of coarse aggregate based on 3D point clouds. The processing flow of the method may include the following steps: S1: Obtain a set of spatial points in a 3D point cloud sampling segment, divide each sampling segment into a grid, calculate the distance difference between adjacent points in the grid, identify the point change amplitude per unit time, analyze the regional morphology, and generate the sampling density adjustment result; S2: Based on the point cloud data in the sampling density adjustment result, distance aggregation is performed according to local geometric consistency, and the unit vector angle between adjacent points in the curvature gradient direction is calculated. The angle change of three consecutive points is analyzed. If the angle changes suddenly, it is identified as a boundary identification area, and the coarse aggregate boundary identification result is generated; S3: Using the coarse aggregate boundary identification results, analyze the spatial changes of the coarse aggregate boundary under differentiated environmental conditions, analyze the volume change amplitude of each particle, extract the spatial coordinate displacement data of the change area, predict the performance of the coarse aggregate based on the displacement data, and generate the coarse aggregate performance stability evaluation results; S4: Call the coarse aggregate performance stability assessment results, record the point cloud density and morphological changes before and after adjustment in each area, analyze the regional changes based on the magnitude of the change, combine the physical properties and surface characteristics of the particles, and track the morphological changes of the particles in real time to generate the dynamic morphological adjustment results of the coarse aggregate; S5: Using the results of dynamic morphology adjustment of coarse aggregate, the evolution path of coarse aggregate particles is analyzed, the evolution trend under different environments is identified, and the linear classification of the change trend is performed with reference to loading time and environmental factors to obtain the coarse aggregate evolution trend; The sampling density adjustment results include grid point spacing, point change rate, and regional density gradient; the coarse aggregate boundary identification results include boundary curvature difference, angle mutation position, and boundary continuity; the coarse aggregate performance stability assessment results include particle displacement vector, particle spatial deformation, and environmental adaptability difference information; the coarse aggregate dynamic morphology adjustment results include regional density change rate, particle surface texture change, and particle dynamic stability; the coarse aggregate evolution trend includes evolution path pattern, environmental factor sensitivity, and loading time correlation.

[0021] The specific steps for obtaining the sampling density adjustment results are as follows: S101: Obtain a set of spatial points in a 3D point cloud sampling segment, divide each sampling segment into cubic grid units along the coordinate axis direction, extract spatial point groups within the grid units, calculate the Euclidean distance set of each group of points in the three-dimensional coordinates, and perform statistics on the distance differences between two adjacent points in the distance set to generate a grid point distance difference set; Based on the premise of collecting 3D point cloud data with laser radar or structured light scanning equipment, a scene with a three-dimensional spatial structure is divided into multiple sampling segments. Each segment contains a frame of static or dynamic point cloud information. The actual scene can be warehouse shelves, urban street scenes, construction site structures, etc. The original point set of each frame of sampling segment contains hundreds of thousands of (x, y, z) coordinate information. On this basis, the spatial boundary is set and the grid system is constructed along the axial direction. The 30m×30m×5m space is divided into cubic units with a side length of 1m, thus obtaining 4500 spatial grid units. Spatial point screening is performed on each grid unit, that is, the original point set is split into multiple point groups by coordinate attribution judgment. Each point group Belonging to a unique grid unit, the distance set construction is performed on each point group, and the point pairs in the same unit are traversed and their spatial distances are recorded to form a complete distance set. The distance data in the set are further calculated to obtain a difference set. By setting the difference analysis rules between adjacent points, the range is divided by specific values ​​and statistics are performed. In the indoor point cloud, if the difference is less than 0.01 meters, it can be regarded as a dense state, and if the difference is greater than 0.1 meters, it is classified as a sparse state. Some differences between 0.01 and 0.1 meters can be further divided into levels at intervals of 0.02 meters, and the frequency of various differences is counted to evaluate the distribution characteristics and local change status of spatial points in the current grid, and generate a grid point distance difference set.

[0022] S102: Based on the grid point distance difference set, sequentially compare the distance difference values ​​of the same grid unit in the differentiated time series sampling segments, analyze the offset interval between the distance difference value and the stability within the grid, and obtain the unit time point distance offset intensity value; Compare multiple point cloud sampling segments at different time sequences. Each comparison is based on the same grid unit. Set a cubic grid unit at a fixed position and generate a difference set at each time point t1, t2, and t3. The data structure in the difference set is consistent for easy comparison. During the comparison process, extract the distance difference sequence within the grid from each time segment, and then compare them one by one according to the position of the sequence median. By setting a unified comparison benchmark, you can arrange them in chronological order and observe the trend of value changes. Analyze the amplitude and frequency of the difference value changes within the grid unit in each time slice. If a grid has a batch difference concentrated at 0.0 at t1, The deviation is less than 2 meters at t2, but most of the differences are concentrated above 0.1 meters at t2, indicating that the fluctuation of the distance between points has intensified, that is, the density instability zone has expanded. By counting the number corresponding to each difference range and recording its change direction, the trend of the offset interval is obtained in combination with the change frequency. Then, the time interval parameter is set, such as performing a difference value analysis every 1 second, calculating the maximum and minimum differences of each interval, and then calculating the average offset degree. This process is defined as the point distance offset intensity value per unit time. The judgment threshold can be set according to the initial scenario. For example, if the point distance offset intensity value exceeds 0.05 meters, it is judged that the offset is increasing, and if it is less than 0.01 meters, it is judged that the offset is stabilizing. The point distance offset intensity value per unit time is obtained.

[0023] S103: Divide the grid cells into a density stable region and a density changing region based on the offset intensity value of the unit time point, calculate the mean density of the sampling points based on the number and volume of spatial points in the density changing region, and obtain a sampling density adjustment result; The grid cells are classified into density stable areas and density changing areas. In actual operation, the intensity thresholds for the stable and changing areas are set. The point offset intensity less than 0.01 meters is classified as the density stable area, and greater than or equal to 0.01 meters is classified as the density changing area. The spatial point group in the density changing area is counted to obtain the number of points and the volume of the grid cell. The mean density of the sampling points is calculated by dividing the number of points by the volume. If 85 points are detected in a density changing area and the cell volume is 1 cubic meter, the average density is 85 points / m³. By comparing the data of multiple density changing cells, the density distribution can be formed. Trend chart is generated, and sampling density can be adjusted accordingly. The adjustment can be performed within a custom density range. For example, the target density range is set to 50-100 points / m³. If the density of a unit is lower than the lower limit, it is recorded as a sparse area. If it is higher than the upper limit, it is recorded as an over-dense area. The density adjustment result is used to further analyze whether additional sampling or denoising is required. If the current device supports the density adaptation mechanism, the sampling frequency or sampling position can be adjusted according to the result. The density change trajectory of the same area in different time slices is recorded to provide basic data support for subsequent dynamic point cloud modeling or environmental change detection, and obtain the sampling density adjustment result.

[0024] The specific steps for obtaining the coarse aggregate boundary identification results are as follows: S201: Calling the point cloud data in the sampling density adjustment result, performing geometric consistency judgment on the local neighborhood of each point based on the spatial distance value and the difference in the 3D normal vector direction between the point and the points in the neighborhood, merging the point sets with consistent features into an aggregation group, and analyzing the curvature gradient direction between adjacent points in the aggregation group to obtain the curvature gradient direction angle value; Call the point cloud data marked as the density change area in the sampling density adjustment result, and construct the neighborhood of the spatial points according to a certain neighborhood radius. Set the neighborhood radius to 0.1 meters. For each center point, search for the neighboring points within the radius, build a local neighborhood point set, count the spatial distance between the points in the neighborhood and the center point, and combine the normal vector direction calculated in advance for each point to further analyze the difference in the normal vector angle between the center point and each point in the neighborhood. If the difference is within a certain range and is set to less than 10 degrees, it is considered that the point is consistent with the center point direction and meets the geometric consistency condition. The points that meet the conditions are added to the same aggregation group in turn to form a local geometric consistency point set. Each aggregation A cluster represents a local area with the same surface features or curvature trends. In the cluster, continuous point pairs are selected, and the difference between the main curvature directions of each pair of points is calculated according to the spatial distribution order to determine whether adjacent points change along the same curvature direction. The angle value of the curvature gradient direction is obtained by the angle difference between the curvature directions derived from the normal vector in three-dimensional space. The cluster is set to contain 12 points, forming a total of 11 pairs of adjacent points. By comparing the direction changes between each point, angle change data support can be provided for subsequent boundary recognition. In order to improve actual operability, the angle function in the PCL library or the vector direction difference analysis method can be used to realize angle measurement to obtain the curvature gradient direction angle value.

[0025] S202: Based on the angle value of the curvature gradient direction, three consecutive points in each group are selected as a combination unit, and the numerical difference between the two angle values ​​is determined in sequence. If the angle difference exceeds the boundary mutation judgment threshold, the real-time middle point is marked as the boundary recognition point, and the spatial position coordinates and normal vector directions corresponding to the marked points are summarized to identify the boundary spatial feature distribution of the point cloud area and obtain the coarse aggregate boundary recognition result; Perform three-point combination operations on the midpoints of each aggregation group, select each group of three points as a group of combination units, and judge the numerical difference between the front and rear angle values. In this judgment process, analyze the angle changes corresponding to the center points of each combination unit in sequence. Set the front angle of a combination unit to 15 degrees and the rear angle to 30 degrees. The difference between the two is 15 degrees. If the difference exceeds the set threshold, that is, the boundary mutation condition is triggered, the current center point is marked as the boundary identification point. In order to improve the accuracy, the mutation judgment threshold can be set to three levels: 10 degrees, 20 degrees, and 30 degrees. Among them, more than 30 degrees is defined as a significant mutation area, and 10 to 30 degrees is a medium mutation zone, and less than 10 degrees is a non-mutation zone. If combined with the coarse aggregate detection needs, the appropriate grading standard can be set according to the actual particle size, and the spatial position coordinates of the marked points and their corresponding normal vector direction information are selected and summarized to form a boundary point set. The sparseness or density of the boundary points can also be used to identify whether it is the boundary area of ​​coarse aggregate. At the edge of the coarse aggregate surface, the normal vector of the point changes quickly and the angle changes greatly, so it is easier to form concentrated boundary point marks. By uniformly projecting or visualizing the boundary points, it provides a basis for subsequent boundary encapsulation, polygon reconstruction or voxel filtering, and obtains the coarse aggregate boundary recognition result.

[0026] The specific steps for obtaining the coarse aggregate performance stability evaluation results are as follows: S301: calling boundary spatial feature distribution data from the coarse aggregate boundary recognition result, combining multiple sets of boundary point cloud samples under differentiated environmental conditions, referring to the spatial coordinates of the corresponding boundary points, performing a difference comparison on the position changes of the same particle under differentiated conditions, calculating the displacement distance of the particle boundary point in three-dimensional space, and obtaining the coarse aggregate boundary spatial displacement data; The formula for calculating the displacement distance of the particle boundary point in three-dimensional space is as follows: ; in, Representative particles The displacement distance in three-dimensional space, Representative particles In the Under different conditions coordinate, Representative particles In the The initial state under different conditions coordinate, Representative particles In the Under different conditions coordinate, Representative particles In the The initial state under different conditions coordinate, Representative particles In the Under different conditions coordinate, Representative particles In the The initial state under different conditions coordinate, represents the number of differentiation condition groups; Parameter meaning and calculation process: particle exist 、 、 Coordinates in the direction ( , , ) Collect data under different environmental conditions through sensors or laser radar and other equipment; The coordinates of the initial state ( , , ) is obtained by recording the position of the particle before the environmental conditions change; For each differentiation condition, the distance that the particle coordinates change in the three-dimensional direction is calculated, and the square root of the sum of the coordinate differences is used to represent the total distance that the particle position changes; During the calculation process, the particles under different environmental conditions are used. The coordinates of , , ) and the initial coordinates ( , , ) for comparison; Setting up particles Under three different environmental conditions ( ) are: Under environmental conditions Next, particles The coordinates are m, m, m; Under environmental conditions Next, particles The initial coordinates are m, m, m; In environmental conditions Next, particles The coordinates are m, m, m; In environmental conditions Next, particles The initial coordinates are m, m, m; In environmental conditions Next, particles The coordinates are m, m, m; In environmental conditions Next, particles The initial coordinates are m, m, m, Substitute into the formula for calculation: ; Calculate item by item: Item 1: ; Item 2: ; Item 3: ; calculate: ; The calculation results show that the particles Under three different environmental conditions, the spatial displacement under three differentiated environmental conditions is 1.6431m. This value indicates the overall displacement of the particle in space relative to its initial position. This displacement is of great significance for analyzing the behavior of particles under different environmental conditions and can reflect the changes of particles under different experimental conditions. S302: Performing volume reconstruction processing on the boundary point displacement in the spatial displacement data of the coarse aggregate boundary, extracting the boundary point set of the particle envelope in the three-dimensional coordinate system, constructing a spatial bounding box based on the boundary point set, and performing difference calculation on the volume values ​​of the same particle under different conditions to obtain the particle volume change amplitude; Envelope reconstruction is performed on the coordinate change value of each boundary point. The reconstruction process can construct the envelope boundary of the particle based on the point set, set the boundary point set obtained by three-dimensional scanning to establish extreme value constraints within the coordinate range, and select the boundary point subset that surrounds the entire particle. The bounding box structure is generated based on the boundary points. The bounding box structure can be set as a three-dimensional minimum bounding cube or polygon combination. After constructing the bounding box for the same particle under multiple environmental conditions, the volume change values ​​of each bounding box are compared. The initial volume is set to 0.125m³, and the volume after the environmental effect is 0.132m³. , the difference is 0.007m³, which can be used as the volume change amplitude of the particle in the corresponding environment. In actual engineering, a reference value for the volume change interval can be set. For example, in the coarse aggregate compression test, if the change amplitude is less than 0.005m³, it is considered to be in the stable range; if it is between 0.005 and 0.01m³, it is considered to be a slight change; if it exceeds 0.01m³, it is defined as a significant change. Particle samples are classified and summarized according to this standard, and the effect trend of different environments on the volume change of the same particle is recorded at the same time, providing a distribution basis for further identifying weak stability areas and obtaining the particle volume change amplitude.

[0027] S303: Based on the particle volume change amplitude and the spatial position of the particle center point, identify the area where the volume change amplitude is greater than the particle stability fluctuation baseline value, extract the three-dimensional coordinate displacement vector of the particle in the area under the differentiated environmental conditions, evaluate the directional consistency of the displacement vector, and obtain the coordinate displacement data of the change area; Based on the spatial coordinates of the particle center point, areas with fluctuations exceeding the set fluctuation baseline are identified. When setting the baseline value, the maximum allowable volume change of particles in actual testing can be referenced. 0.008 m³ is set as the baseline value for coarse aggregate stability fluctuation. If the volume change of a particle is 0.012 m³, it should be identified as a stability anomaly area. The three-dimensional coordinate displacement vectors of the particles in this area under various environmental conditions are further extracted, and a vector direction consistency analysis mechanism is established. In specific operations, the displacement direction change trends of the particles in multiple environments are compared in sequence. If the displacement direction of the particles in the three environments remains roughly northwest, it can be determined to be consistent in direction. If the displacement direction in the three environments is toward southeast, northwest, and due north, respectively, it indicates low directional consistency. The sum or mean of the vector angle deviations can be defined as the directional consistency evaluation indicator. A judgment threshold is set. For example, a mean deviation exceeding 45 degrees is considered poor consistency, and a deviation below 20 degrees is considered strong consistency. This judgment result will serve as an important basis for determining whether the particles are stable in the spatial structure. After marking the abnormal fluctuation areas, they can be used for performance grading and obtain coordinate displacement data for the changing areas.

[0028] S304: Based on the coordinate displacement data of the change area, the displacement length value, change direction angle and volume change of each particle under differentiated environmental conditions are combined and calculated. The stability performance of the particles is judged by the numerical change trend to obtain the coarse aggregate performance stability assessment result; The displacement length value, change direction angle and volume change are combined and processed to form a data set describing the spatial behavior of particles in differentiated environments. The three indicators of total displacement length, main change direction angle and volume change amplitude are calculated for each particle respectively. A joint judgment model of the three indicators is established. In the indoor experimental scene, a particle displacement length of 0.12 meters, a direction angle change range of 35 degrees, and a volume change of 0.009 m³ are obtained. Combined with the preset stability performance grade grading standard, a four-level model can be set: Level I is completely stable (all indicators are below the set lower limit), Level II is slightly unstable (any indicator slightly exceeds the set value), Level III is moderately unstable (two indicators exceed the threshold), and Level IV is highly unstable (all three indicators significantly exceed the standard). By judging the level of the particle, its characteristics are recorded and incorporated into the performance prediction model. The output structure includes five types of information: particle number, location, environmental conditions, evaluation level, offset vector, etc., to generate the coarse aggregate performance prediction results.

[0029] The specific steps for obtaining the dynamic morphology adjustment results of coarse aggregate are as follows: S401: Recalling the spatial position identifiers of particles in the coarse aggregate performance stability evaluation results, extracting point cloud data before and after adjustment for the area where each particle is located, calculating the number of points per unit volume in the same coordinate system, analyzing the distance of regional morphological changes, and generating point cloud density and morphological change amount; The formula for analyzing regional morphological changes is as follows: ; in, Representative Region and The amount of morphological change in the region, represents the volume of the region, Represents the total number of point clouds involved in the calculation, Representative Region No. The coordinates of the particle, Representative Region No. The coordinates of the particle, Representatives in the The number of point clouds extracted at the particle position, is the adjustment index; Detailed explanation of the formula and the process of formula calculation and derivation: Representative Region and The morphological change of the region is analyzed by two regions ( and ) between the point cloud data, the point cloud data contains the particle coordinate information in space, the goal is to use the data to evaluate the morphological differences between regions, Parameter meaning and formula calculation derivation process: Represents the regional volume. The regional volume is calculated based on the distribution of point cloud data in three-dimensional space. It can be obtained by measuring the spatial boundaries within the region. Set the region and region The volume of (This value is calculated based on the size of the actual experimental area and the distribution range of the point cloud); Represents the total number of point clouds involved in the calculation. The point cloud dataset consists of multiple particles. It is the total number of particles included in the analysis process, set to collect in each area Coordinate data of each particle; Representative Region No. The coordinates of the particle, similarly, Representative Region No. The coordinates of the particles are obtained through point cloud scanning technology and can be measured using lidar or spatial data acquisition equipment and set at the first particle position. , in the region and region The coordinates in are and ; Representatives in the The number of point clouds extracted at the particle position. The parameter determines the sampling density of the point cloud data at each particle position. In actual calculation, it is set at each particle position, with an average of Point cloud sample data; It is the adjustment index of morphological change. This index is used to adjust the degree of influence of point cloud density on morphological change. The value is set according to the experimental design and the spatial characteristics of the region. The value range is arrive , in this example, it is set to ,This value is used when the sensitivity of regional morphological changes is high; For the Particles, the coordinate difference is: ; That is, the first particle is in the region and region The difference between them is 12; Calculate the point cloud weight for each particle position: Set at the first particle position, , so the weights are: ; For all From 1 to , using the above calculation process, find the square of the difference between each particle and add them up after applying the point cloud number weight: ; The results show that the region and region The morphological change between them is 0.03792, which characterizes the difference in point cloud density and morphology between the two regions. This value reflects the distribution difference and density change degree of particles in space.

[0030] S402: Based on the point cloud density and morphological change, the change value of the region is matched and analyzed with the density value, hardness value and surface roughness of the particle material parameters to identify the density change trend and obtain a sensitive area identification mark group; Based on the acquired point cloud density and morphological change data, particle material parameter information is further introduced for matching analysis. Material parameters include three core indicators: density, hardness, and surface roughness. The density of coarse aggregate is set to 2.6 g / cm³, the hardness to level 6, and the roughness to 0.9 μm. By setting parameter intervals and matching them with regional density change values, a basis for determining density trends is formed. If the point cloud density in a certain area increases significantly, and its original material parameters show a high-density, low-hardness type, it is determined to be prone to deformation under external forces or environmental influences and further marked as a density-sensitive area. Areas that meet the mutation characteristics and produce abnormal parameter matching are included in the sensitive area identification marker group. The marker group data structure includes fields such as region index, point cloud density change value, material parameter number, and change trend description, providing a positioning basis for subsequent dynamic behavior analysis. During operation, change threshold judgment rules can be set. For example, if the density change exceeds 20 points / m³, the morphological deviation exceeds 0.15 meters, and the material hardness is less than level 5, the area is automatically added to the marker group. The marked area simultaneously enters the continuous frame boundary tracking process to obtain the sensitive area identification marker group.

[0031] S403: Based on the sensitive area identification marker group, the boundary coordinate change trajectory of the particles in the marked area in the continuous time frame is extracted, the coordinate change trend and velocity distribution value of the boundary point of each particle in the three-dimensional direction are tracked, the dynamic morphological state of the particles is classified and identified, and the dynamic morphological adjustment result of the coarse aggregate is obtained; For each region in the sensitive region identification marker group, the trajectory of particle boundary coordinate changes in continuous time frames was extracted. The sampling frequency of the time frames used was kept stable, with a setting of 10 frames per second. By comparing the position changes of the boundary points of the marked particles in each frame, the coordinate changes in the three-dimensional direction were recorded, and the movement trend of the particle boundary points in the time dimension was calculated. The displacement per unit time of each point was also recorded to obtain its velocity distribution value. During the analysis process, classification was performed based on the mean and standard deviation of the velocity changes. For example, particles with significant velocity fluctuations are in a loose state, while those with stable velocity but continuous direction changes are in a slow rotation or deformation state. Further combined with the morphological offset value, a dynamic morphological state discrimination model was established. The average movement distance of the boundary points of a particle in 10 frames was set to 0.05 meters, the velocity changes were concentrated between 0.03 and 0.07 meters per second, and the directional change angle distribution was between 30 and 50 degrees. The dynamic morphology of the particle was determined to be "highly active deformation type". Conversely, if the position change was small, the velocity was stable, and there was no significant directional drift, it was classified as "stable type". This was used for subsequent model prediction and engineering evaluation to obtain the dynamic morphological adjustment results of coarse aggregate.

[0032] The specific steps for obtaining the evolution trend of coarse aggregate are as follows: S501: calling the boundary coordinate change trajectory of the particles in the coarse aggregate dynamic morphology adjustment result in the continuous time frame, connecting the paths in time sequence according to the particle identification, analyzing the morphology change flow of the particles through the continuity of the change direction and the fluctuation range of the speed, and generating a particle morphology evolution path set; The boundary coordinate change trajectory data of each particle in the continuous time frame in the dynamic morphology adjustment result of the coarse aggregate is called, and the data of each frame is sorted and arranged in time sequence according to the unique identifier of the particle to construct a complete set of particle time series boundary points. Then, based on the spatial position change of the particle in each frame, the boundary points under adjacent time frames are connected in sequence to generate continuous path segments. The direction vector of each path segment is calculated, and the degree of continuity of the change direction is judged by the change angle between the directions of continuous segments. If the angle between the directions of adjacent segments is less than 15 degrees, it is considered to be strong continuity, and if it is greater than 45 degrees, it is judged to be a mutation. Then, combined with the direction vectors of each segment, the direction vectors of each segment are connected in time sequence to generate continuous path segments. The velocity fluctuations within the corresponding time interval are evaluated as a whole, and the velocity range and directional continuity score of each particle are recorded. If the velocity fluctuation of a particle is between 0.02 and 0.06 m / s and the directional angle is less than 20 degrees, it can be determined as a stable evolution path. Conversely, if the velocity fluctuation exceeds 0.1 m / s and the direction changes frequently, it is classified as an unstable path type. The path information of all particles is unified. The path set contains key fields such as particle number, path coordinate sequence, average velocity value, and direction change statistics. It serves as a structured input data source for the motion trend of multi-particle groups and generates a particle morphology evolution path set.

[0033] S502: Based on the particle morphology evolution path set, for the particle path characteristics under differentiated environments, with reference to the loading time data and environmental factors of the corresponding period, curve fitting is performed on the velocity change trend of each type of path. The paths are linearly classified based on the correspondence between the three-dimensional coordinate change rate and the environmental response to obtain the coarse aggregate evolution trend; The path data under different environmental conditions were selected for feature comparison and analysis, and the speed change trend of each section in the path was extracted. The speed change trend was then correlated with the recorded loading time data and environmental parameters for modeling. Under the conditions of 60% humidity and 500kPa loading pressure, the particle path showed a trend of continuous speed increase. However, when the humidity rose to 80% at the same loading time, the speed change began to fluctuate and increase, indicating that the environment has a correlation with the evolution of the particle path. The path was segmented into time windows using a time period division method. The speed change value was recorded in each segment, and a trend line was established to analyze its change direction. The continuity of the speed value and the change The amplitude of the change is combined with the change point of the environmental variables to determine the corresponding relationship between the two. The curve fitting process can adopt a piecewise linear fitting strategy to fit the path data of the same particle and the corresponding three-dimensional coordinate change rate. It is set in the interval of 10 to 20 seconds of the loading time. The coordinate change value of the particle in the Z-axis direction increases from 1.02 meters to 1.12 meters. The fitting result shows that it is linearly positively correlated with the loading rate, which is classified as a pressure-responsive path. By analyzing the correspondence between the path change rate and each environmental response index, linear classification is performed, and the paths are grouped to form an evolution trend set, which provides a data basis for structural stability modeling and prediction and obtains the evolution trend of coarse aggregate.

[0034] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any modifications or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in the present invention should be included in the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be based on the scope of protection of the claims.

Claims

1. A method for collecting and characterizing coarse aggregate morphology based on 3D point clouds, characterized in that: The following steps are involved: S1: Obtain a set of spatial points in a 3D point cloud sampling segment, divide each sampling segment into a grid, calculate the distance difference between adjacent points in the grid, identify the point change amplitude per unit time, analyze the regional morphology, and generate the sampling density adjustment result; S2: Based on the point cloud data in the sampling density adjustment result, distance aggregation is performed according to local geometric consistency, the unit vector angle between adjacent points in the curvature gradient direction is calculated, and the angle change of three consecutive points is analyzed. If the angle changes suddenly, it is identified as a boundary identification area, and a coarse aggregate boundary identification result is generated; S3: Using the coarse aggregate boundary identification results, analyze the spatial changes of the coarse aggregate boundary under differentiated environmental conditions, analyze the volume change amplitude of each particle, and combine the spatial coordinate displacement data of the change area to generate the coarse aggregate performance stability evaluation results; S4: calling the coarse aggregate performance stability evaluation result, recording the point cloud density and morphological changes before and after adjustment in each area, analyzing the regional changes according to the magnitude of the change, combining the physical properties and surface characteristics of the particles, and tracking the morphological changes of the particles in real time to generate the coarse aggregate dynamic morphological adjustment result.

2. The method for collecting and characterizing coarse aggregate morphology based on 3D point cloud according to claim 1, characterized in that: The sampling density adjustment results include grid point spacing, point change rate, and regional density gradient; the coarse aggregate boundary identification results include boundary curvature difference, angle mutation position, and boundary continuity; the coarse aggregate performance stability assessment results include particle displacement vector, particle spatial deformation, and environmental adaptability difference information; the coarse aggregate dynamic morphology adjustment results include regional density change rate, particle surface texture change, and particle dynamic stability.

3. The method for collecting and characterizing coarse aggregate morphology based on 3D point cloud according to claim 1, characterized in that: The steps for obtaining the sampling density adjustment result are specifically as follows: S101: Obtain a set of spatial points in a 3D point cloud sampling segment, divide each sampling segment into cubic grid units along the coordinate axis direction, extract spatial point groups within the grid units, calculate the Euclidean distance set of each group of points in the three-dimensional coordinates, and perform statistics on the distance differences between two adjacent points in the distance set to generate a grid point distance difference set; S102: Based on the grid point distance difference set, sequentially compare the distance difference values ​​of the same grid unit in the differentiated time series sampling segments, analyze the offset interval between the distance difference value and the stability within the grid, and obtain the unit time point distance offset strength value; S103: Divide the grid unit into a density stable area and a density change area according to the offset intensity value of the unit time point, calculate the mean density of the sampling points based on the number and volume of spatial points in the density change area, and obtain a sampling density adjustment result.

4. The method for collecting and characterizing coarse aggregate morphology based on 3D point cloud according to claim 3, characterized in that: The steps for obtaining the coarse aggregate boundary recognition result are specifically as follows: S201: calling the point cloud data in the sampling density adjustment result, performing geometric consistency judgment on the local neighborhood of each point based on the spatial distance value and the difference in the three-dimensional normal vector direction between the point and the points in the neighborhood, merging the point sets with consistent features into an aggregation group, and analyzing the curvature gradient direction between adjacent points in the aggregation group to obtain the curvature gradient direction angle value; S202: According to the angle value of the curvature gradient direction, three consecutive points in each group are selected as a combination unit, and the numerical difference between the two angle values ​​is judged in turn. If the angle difference exceeds the boundary mutation judgment threshold, the real-time middle point is marked as the boundary identification point, and the spatial position coordinates and normal vector directions corresponding to the marked points are summarized to identify the boundary spatial feature distribution of the point cloud area and obtain the coarse aggregate boundary identification result.

5. The method for collecting and characterizing coarse aggregate morphology based on 3D point cloud according to claim 4, characterized in that: The steps for obtaining the coarse aggregate performance stability evaluation results are specifically as follows: S301: calling the boundary space feature distribution data in the coarse aggregate boundary recognition result, combining multiple groups of boundary point cloud samples under differentiated environmental conditions, referring to the spatial coordinates of the corresponding boundary points, performing a difference comparison on the position changes of the same particle under differentiated conditions, calculating the displacement distance of the particle boundary point in three-dimensional space, and obtaining the coarse aggregate boundary space displacement data; S302: performing volume reconstruction processing on the boundary point displacement in the coarse aggregate boundary spatial displacement data, extracting the boundary point set of the particle envelope in the three-dimensional coordinate system, constructing a spatial bounding box through the boundary point set, and performing difference calculation on the volume values ​​of the same particle under different conditions to obtain the particle volume change amplitude; S303: Based on the particle volume change amplitude and the spatial position of the particle center point, identifying an area where the volume change amplitude is greater than the particle stability fluctuation baseline value, extracting the three-dimensional coordinate displacement vector of the particle in the area under differentiated environmental conditions, evaluating the directional consistency of the displacement vector, and obtaining coordinate displacement data of the changed area; S304: Based on the coordinate displacement data of the change area, the displacement length value, change direction angle and volume change of each particle under differentiated environmental conditions are combined and calculated, and the particle stability performance is judged by the numerical change trend to obtain the coarse aggregate performance stability evaluation result.

6. The method for collecting and characterizing coarse aggregate morphology based on 3D point cloud according to claim 5, characterized in that: The formula for calculating the displacement distance of the particle boundary point in three-dimensional space is as follows: ； in, Representative particles The displacement distance in three-dimensional space, Representative particles In the Under different conditions coordinate, Representative particles In the The initial state under different conditions coordinate, Representative particles In the Under different conditions coordinate, Representative particles In the The initial state under different conditions coordinate, Representative particles In the Under different conditions coordinate, Representative particles In the The initial state under different conditions coordinate, represents the number of differentiation condition groups.

7. The method for collecting and characterizing coarse aggregate morphology based on 3D point cloud according to claim 6, characterized in that: The steps for obtaining the dynamic morphology adjustment result of the coarse aggregate are specifically as follows: S401: calling the spatial position identifiers of the particles in the coarse aggregate performance stability evaluation results, extracting the point cloud data before and after adjustment for the area where each particle is located, calculating the number of points per unit volume in the same coordinate system, analyzing the distance of regional morphological changes, and generating point cloud density and morphological change amount; S402: Based on the point cloud density and morphological change, matching and analyzing the change value of the region with the density value, hardness value, and surface roughness of the particle material parameters, identifying the density change trend, and obtaining a sensitive region identification mark group; S403: Based on the sensitive area identification mark group, the boundary coordinate change trajectory of the particles in the marked area in the continuous time frame is extracted, the coordinate change trend and velocity distribution value of the boundary point of each particle in the three-dimensional direction are tracked, the dynamic morphological state of the particles is classified and identified, and the dynamic morphological adjustment result of the coarse aggregate is obtained.

8. The method for collecting and characterizing coarse aggregate morphology based on 3D point cloud according to claim 7, characterized in that: The formula for analyzing regional morphological changes is as follows: ； in, Representative Region and The amount of morphological change in the region, represents the volume of the region, Represents the total number of point clouds involved in the calculation, Representative Region No. The coordinates of the particle, Representative Region No. The coordinates of the particle, Representatives in the The number of point clouds extracted at the particle position, is the adjustment index.

9. The method for collecting and characterizing coarse aggregate morphology based on 3D point cloud according to claim 1, characterized in that: The method further comprises step S5: S5: using the coarse aggregate dynamic morphology adjustment results, analyzing the evolution path of coarse aggregate particles, identifying the evolution trend under differentiated environments, and linearly classifying the change trend with reference to loading time and environmental factors to obtain the coarse aggregate evolution trend; The coarse aggregate evolution trend includes evolution path pattern, environmental factor sensitivity, and loading time correlation.

10. The method for collecting and characterizing coarse aggregate morphology based on 3D point cloud according to claim 9, characterized in that: The steps for obtaining the evolution trend of coarse aggregate are as follows: S501: calling the boundary coordinate change trajectory of the particles in the coarse aggregate dynamic morphology adjustment result in the continuous time frame, connecting the paths in time sequence according to the particle identification, analyzing the morphology change direction of the particles through the continuity of the change direction and the fluctuation range of the speed, and generating a particle morphology evolution path set; S502: Based on the particle morphology evolution path set, for the particle path characteristics under differentiated environments, with reference to the loading time data and environmental factors of the corresponding time period, curve fitting processing is performed on the speed change trend of each type of path, and the paths are linearly classified through the correspondence between the three-dimensional coordinate change rate and the environmental response to obtain the coarse aggregate evolution trend.

Images