Point Cloud Data Processing System

Abstract

To provide a point cloud data processing system that facilitates maintenance of a slope surface.SOLUTION: A point cloud data processing system includes: means that performs first extraction processing (step S11) of extracting point cloud data in a rectangle B including a road geometric line A from point cloud data T acquired by a point cloud data acquisition device 10; means that performs removal processing (step S12) of removing point cloud data in a range C (point cloud data of a road) from the point cloud data within the rectangle B; means that performs projection processing (steps S13 and S14) of projecting the point cloud data after the removal processing to a two-dimensional plane horizontally crossing the road geometric line A; means that performs second extraction processing (step S15) of extracting point cloud data satisfying a predetermined condition as point cloud data of a slope surface possibility from the point cloud data after the projection processing; and means that performs third extraction processing (step S16) of extracting the point cloud data of the slope surface by a region expansion method having a start point G as an initial point from the extracted point cloud data of the slope surface possibility.SELECTED DRAWING: Figure 2

Description

[Technical Field]

[0001] The present invention relates to a technology for a point cloud data processing system that processes acquired point cloud data. [Background technology]

[0002] Technological innovations in cameras and lasers have led to a diversification of measurement methods for understanding the shape of feature surfaces. Methods for acquiring point cloud data can be broadly divided into two types: laser measurement and photogrammetry. Laser measurement methods include airborne LP (aircraft or UAV-mounted laser profiling systems) and MMS (Mobile Mapping Systems).

[0003] Patent Document 1 discloses an MMS technology that uses a vehicle-mounted laser scanner to acquire point cloud data representing the shapes of features along roads. By acquiring point cloud data in this way, the point cloud data can be used to help maintain and manage roads and the features around the roads.

[0004] One such feature requiring maintenance is a slope adjacent to a road. A slope is an artificially created incline adjacent to a road by cutting and filling soil. A type of slope that has been protected by spraying mortar or concrete to prevent weathering and erosion of the bedrock is called a shotcrete slope. Recently, natural disasters and aging have caused problems such as peeling, spalling, and collapse of slopes (especially shotcrete slopes). Therefore, there is a need to analyze and evaluate point cloud data of the slope to maintain and manage the slope.

[0005] However, the point cloud data acquired by MMS etc. includes not only point cloud data of slopes but also point cloud data of areas other than slopes. This makes it difficult to find point cloud data of slopes from the entire acquired point cloud data, which in turn makes it difficult to maintain and manage slopes using point cloud data. [Prior art documents] [Patent documents]

[0006] [Patent Document 1] Japanese Patent Application Laid-Open No. 2009-204615 Summary of the Invention [Problem to be solved by the invention]

[0007] The present invention has been made in consideration of the above-mentioned circumstances, and the problem that it aims to solve is to provide a point cloud data processing system that can facilitate the maintenance and management of slopes. [Means for solving the problem]

[0008] The problem to be solved by the present invention is as described above, and the means for solving this problem will now be described.

[0009] That is, claim 1 comprises a first extraction means that performs a first extraction process to extract point cloud data of a first region including a road alignment from point cloud data acquired by a predetermined method; a removal means that performs a removal process to remove point cloud data of the road from the point cloud data of the first region extracted by the first extraction means; a projection means that performs a projection process to project the point cloud data after the removal process has been performed by the removal means onto a two-dimensional plane that crosses the road alignment; a second extraction means that performs a second extraction process to extract point cloud data that satisfies a predetermined condition from the point cloud data after the projection process has been performed by the projection means as point cloud data of a slope candidate; and a third extraction means that performs a third extraction process to extract point cloud data of a slope from the point cloud data of the slope candidate extracted by the second extraction means by a region growing method that uses a point approximately at the center of the point cloud data of the slope candidate as a starting point. and a dividing means for dividing the point cloud data projected onto the two-dimensional plane into a plurality of point cloud data by labeling based on Euclidean distance, and the second extracting means extracts, as the predetermined condition, point cloud data of a slope candidate, from among the point cloud data divided by the dividing means, the point cloud data in which the angle of the normal vector of each point included in the point cloud data falls within a predetermined range. It is something.

[0011] Claim 2In the above, the second extraction means extracts, as the point cloud data of a slope candidate, the point cloud data in which the angle of the normal vector of the point cloud data is within a predetermined range, and the height of the point cloud data is equal to or greater than a certain value, as the point cloud data of a slope candidate.

[0012] Claim 3 In the method, when there are multiple point cloud data that satisfy the specified conditions, the second extraction means extracts, from the multiple point cloud data, the point cloud data that is closest to the road alignment as point cloud data of a slope candidate.

[0013] Claim 4 In the method, when the area of ​​the point cloud data extracted by the area growing method is smaller than a predetermined value, the third extraction means excludes the point cloud data within the area from the point cloud data of the slope.

[0014] Claim 5 The present invention includes a connecting means for dividing a three-dimensional space containing the point cloud data extracted by the third extraction means into a plurality of grids, and connecting adjacent grids among the grids in which point cloud data exists, thereby making the point cloud data in the connected grids into point cloud data of a single slope.

[0015] Claim 6 The present invention is provided with a vegetation extraction means for extracting point cloud data located above the point cloud data of the slope as point cloud data of vegetation. [Effects of the Invention]

[0016] The present invention has the following effects.

[0017] In claim 1, by extracting only the point cloud data of the slope, it becomes easier to evaluate the slope, and in turn, it becomes easier to maintain and manage the slope. Furthermore, point cloud data of the slope candidate can be extracted with high accuracy.

[0019] Claim2 In this case, point cloud data of the slope candidate can be extracted with higher accuracy.

[0020] Claim 3 In this case, point cloud data of the slope candidate can be extracted with higher accuracy.

[0021] Claim 4 In this case, it is possible to prevent vegetation from being mistakenly identified as a slope.

[0022] Claim 5 In this case, the accuracy of slope extraction can be improved.

[0023] Claim 6 In this method, it is possible to calculate the slope area and vegetation rate even for slopes with thick vegetation. [Brief explanation of the drawings]

[0024] [Figure 1] 1 is a block diagram showing the configuration of a point cloud data processing system according to an embodiment of the present invention. [Figure 2] 10 is a flowchart showing a method for extracting point cloud data of a slope. [Figure 3] (a) Overview of range filtering using road alignment. (b) Schematic diagram showing the range of filtering. [Figure 4] A diagram showing an overview of removing ground points (point cloud data that constitutes the road surface, etc.). [Figure 5] A diagram showing an overview of the generation of cross-sectional survey lines. [Figure 6] An overview of cross-sectional view generation. [Figure 7] A diagram showing an overview of the extraction of slope candidates on a cross section. [Figure 8] A diagram showing an overview of slope extraction using region growing. [Figure 9] A diagram showing an overview of labeling. [Figure 10] A diagram showing an overview of vegetation extraction on slopes. [Figure 11]1 is a flowchart showing an extraction method for slope evaluation. [Figure 12] FIG. 10 is a diagram showing an overview of step S21. [Figure 13] FIG. 10 is a diagram showing an overview of step S22. [Figure 14] A diagram showing indicators for evaluating slopes. [Figure 15] A diagram showing an overview of classification by slope shape. [Figure 16] A diagram showing an overview of classification by slope shape. [Figure 17] FIG. 10 is an explanatory diagram showing a sloped pattern. [Figure 18] A diagram showing the relationship between deterioration of sprayed slopes and vegetation. [Figure 19] A diagram showing slope areas and vegetation areas. DETAILED DESCRIPTION OF THE INVENTION

[0025] A point cloud data processing system 20 according to one embodiment of the present invention processes point cloud data acquired by various methods. Specifically, the point cloud data processing system 20 analyzes the acquired point cloud data to extract slopes within a target area and determine the number, size, and shape of slopes to be managed, which can be used to formulate future inspection and repair plans. Furthermore, the point cloud data processing system 20 can evaluate the soundness of slopes (road slopes) from the point cloud data. Here, "soundness" indicates whether or not the function of a structure (slope) is impaired, and the degree of impairment. The point cloud data processing system 20 can also extract differences between disaster and normal times to compare point cloud data from two different periods to identify deformations, calculate the amount of collapse and the scale of damage, etc.

[0026] Furthermore, the point cloud data processing system 20 extracts point cloud data in the slope area from the acquired point cloud data in order to evaluate the soundness of the slope. In this embodiment, point cloud data in the slope area is extracted from the point cloud data acquired by the point cloud data acquisition device 10, and the soundness of the slope is evaluated based on the extracted point cloud data.

[0027] Before describing the point cloud data processing system 20, the point cloud data acquisition device 10 will be briefly described below.

[0028] The point cloud data acquisition device 10 acquires point cloud data. The point cloud data acquisition device 10 can acquire point cloud data by any method. In this embodiment, one method of the point cloud data acquisition device 10 is MMS. The point cloud data acquisition device 10 (MMS) acquires a road alignment A (see FIG. 3, etc.) by driving a vehicle along a road, and also acquires point cloud data representing the shape of features using a laser scanner mounted on the vehicle. The road alignment A can be acquired by any method, and may be acquired based on a sequence of points indicating the vehicle's travel path (GPS history) or may be acquired from map information. Alternatively, the road alignment A may be acquired based on the white lines (center lines) of the road acquired by a laser scanner, or may be manually specified. In this embodiment, the road alignment A is acquired based on a sequence of points indicating the vehicle's travel path (i.e., is configured by the sequence of points). Hereinafter, each point constituting the road alignment A will be referred to as a constituent point a. Other examples of point cloud data acquisition devices 10 include airborne laser profilers and handheld laser scanners, and when utilizing point cloud data as described below, point cloud data acquired by point cloud data acquisition devices using multiple methods may be combined and utilized.

[0029] In this way, slope maintenance can be performed by utilizing the point cloud data acquired by the point cloud data acquisition device 10. This will be specifically described below.

[0030] The point cloud data acquired by the point cloud data acquisition device 10 includes point cloud data of a slope.

[0031] Here, the point cloud data acquired by the point cloud data acquisition device 10 includes not only point cloud data necessary for slope maintenance but also point cloud data unnecessary for slope maintenance. Therefore, there is a problem that it is difficult to analyze and evaluate the point cloud data of the slope when it is just acquired by the point cloud data acquisition device 10.

[0032] Therefore, in this embodiment, the point cloud data processing system 20 extracts point cloud data of the slope from the point cloud data acquired by the point cloud data acquisition device 10. The point cloud data processing system 20 will be described below.

[0033] A point cloud data processing system 20 according to one embodiment of the present invention processes point cloud data acquired by the point cloud data acquisition device 10. Specifically, the point cloud data processing system 20 extracts point cloud data of a slope and evaluates the health of the slope based on the extracted point cloud data of the slope. The point cloud data processing system 20 can also extract point cloud data of vegetation as needed and evaluate the health of the slope based on the extracted point cloud data of vegetation. The point cloud data processing system 20 includes a data storage unit 21, an extraction processing unit 22, and a health evaluation unit 23.

[0034] 1 stores various data. The data storage unit 21 can store the point cloud data acquired by the point cloud data acquisition device 10.

[0035] 1 extracts point cloud data of a slope from the point cloud data acquired by the point cloud data acquisition device 10. The extraction processing unit 22 extracts the point cloud data of a slope by the method shown in the flowchart of FIG.

[0036] First, in step S11, the extraction processing unit 22 performs range filtering using the road alignment A. Hereinafter, a description will be given with reference to FIG.

[0037] FIG. 3 is a plan view showing a portion of the point cloud data T acquired by the point cloud data acquisition device 10. The extraction processing unit 22 superimposes the point cloud data T acquired by the point cloud data acquisition device 10 on the road alignment A (see FIG. 3(a)). Next, the extraction processing unit 22 generates a rectangle B having a buffer width along the road alignment A (see FIGS. 3(a) and 3(b)). Then, the extraction processing unit 22 extracts the point cloud data included in the rectangle B. In this way, by extracting only the point cloud data around the road alignment A, the extraction processing unit 22 removes point cloud data in a range unrelated to slope maintenance.

[0038] Next, in step S12, the extraction processing unit 22 removes ground points (point cloud data constituting the road surface, etc.), which will be described below with reference to FIG.

[0039] FIG. 4 is a partially enlarged plan view of FIG. 3. The extraction processing unit 22 uses a region growing method to remove ground points from the point cloud data extracted in step S11 (point cloud data within rectangle B). More specifically, the extraction processing unit 22 gradually expands the range starting from one of the multiple constituent points a of road alignment A, and considers the area up to the point just before the point changes from upward to sideways to be a road, and removes point cloud data within range C (the hatched area shown in FIG. 4) considered to be the road. The angle at which the point orientation changes from upward to sideways can be set to any angle. The orientation of each point is determined by the normal vector of a plane generated using multiple points near that point.

[0040] Next, in step S13, the extraction processing unit 22 generates a cross-sectional measurement line, which will be described below with reference to FIG.

[0041] FIG. 5 is a schematic plan view showing a road alignment A, etc. The extraction processing unit 22 sets a survey line generation position. The point cloud data acquisition device 10 (MMS) acquires constituent points a of the road alignment A (for example, a1 to a3 shown in FIG. 5) at regular time intervals. Therefore, the intervals between the constituent points a of the road alignment A (the interval between a1 and a2, the interval between a2 and a3) vary depending on the vehicle's traveling speed. For this reason, the extraction processing unit 22 sets the survey line generation position by interpolating points at regular intervals (for example, 20 cm). Then, the extraction processing unit 22 generates a transverse survey line D perpendicular to the road alignment A at the set survey line generation position.

[0042] Next, in step S14, the extraction processing unit 22 generates a cross-sectional view, which will be described below with reference to FIG.

[0043] 6 is a schematic plan view showing a road alignment A, etc. The extraction processing unit 22 extracts point cloud data within a rectangle E that has a certain buffer width around a cross-sectional measurement line D. The extraction processing unit 22 then projects the extracted point cloud data onto a cross section (two-dimensional plane) that crosses the road alignment A along the cross-sectional measurement line D.

[0044] Next, in step S15, the extraction processing unit 22 extracts slope candidates on the cross section view, as will be described below with reference to FIG.

[0045] FIG. 7 is a side view showing the point cloud data projected onto the cross section in step S14. The extraction processing unit 22 extracts point cloud data on the left and right sides of the road alignment A from the point cloud data projected onto the cross section in step S14. The extraction processing unit 22 then aggregates the point cloud data using labeling based on Euclidean distance to generate multiple labels (point sets). Specifically, any point included in a label satisfies the labeling condition based on Euclidean distance with at least one other point included in that label. The extraction processing unit 22 then extracts, as slope candidates, labels for which the angle of the normal vector of the entire set (each point) is within a certain range (30 to 80 degrees). Alternatively, the extraction processing unit 22 may extract, as slope candidates, labels for which the angle of the normal vector of the entire set is within a certain range (30 to 80 degrees) and the label height h1 is equal to or greater than a certain value.

[0046] 7, the extraction processing unit 22 extracts the label F as a slope candidate. If there are multiple slope candidates on each side of the road alignment A (if there are multiple slope candidates that satisfy the above conditions), the extraction processing unit 22 uses the label that is closest to the road alignment A on each side of the road alignment A.

[0047] Next, in step S16, the extraction processing unit 22 extracts the slope by region expansion, which will be described below with reference to FIG.

[0048] 8 is a plan view showing labels F and the like. The extraction processing unit 22 gradually widens the range starting from a point approximately in the center of the slope candidate (label F) extracted in step S15 as a starting point G, and considers the area up to the point just before the direction of the point (normal vector) changes as the slope, and extracts the point cloud data within this range H as slope point cloud data I. If the extraction result (area of ​​range H) is smaller than a certain value, the extraction processing unit 22 excludes it from the slope point cloud data I as noise. For example, if vegetation or the like has been selected as the starting point G, the extraction result (area of ​​range H) will be smaller than a certain value, and therefore it will not be extracted as a slope.

[0049] Next, in step S17, the extraction processing unit 22 performs labeling, which will be described below with reference to FIG.

[0050] 9 is a schematic diagram showing the labeling method. The extraction processing unit 22 divides the point cloud data of the slope extracted in step S16 using grids of a fixed size. More specifically, the extraction processing unit 22 first divides a three-dimensional space including the existence space of the label F into grids. Then, the extraction processing unit 22 checks the grids in the three-dimensional space that include point cloud data. Then, the extraction processing unit 22 searches for adjacent grids among the checked grids and connects the adjacent grids. In this way, the extraction processing unit 22 divides the point cloud data I of the slope into blocks.

[0051] Next, in step S18, the extraction processing unit 22 extracts vegetation on the slope, as will be described below with reference to FIG.

[0052] 10 is a side cross-sectional view showing the point cloud data after the processing of step S17. The extraction processing unit 22 extracts all point cloud data above the point cloud data of the slope extracted in step S17, thereby extracting point cloud data J of vegetation.

[0053] In this way, the point cloud data of the slope can be extracted from the point cloud data acquired by the point cloud data acquisition device 10. This makes it easier to evaluate the slope.

[0054] In the point cloud data processing system 20, first, only the point cloud data around the road alignment A is extracted (step S11), and then the point cloud data (ground points) that constitute the road surface are removed (step S12). By doing so, it is possible to easily grasp the range of the slope.

[0055] Furthermore, in the point cloud data processing system 20, point cloud data of potential slopes is extracted by generating a cross section (steps S13 to S15), and further point cloud data of slopes is extracted by the region growing method (step S16), making it possible to extract point cloud data of slopes over the entire area along the road alignment A. It is also possible to distinguish whether the slopes are located on the mountain side or the valley side when viewed from the road surface.

[0056] Furthermore, for slopes with a high rate of vegetation growth, the process of step S16 (slope extraction by area expansion) alone may result in a single slope being mistakenly recognized as multiple slopes. Therefore, in step S17, by determining continuity using a grid of a fixed size, even a slope with a high rate of vegetation growth can be identified as a single slope. Then, by performing the process of step S18, the slope area and vegetation rate can be calculated even for slopes with a high rate of vegetation growth.

[0057] The point cloud data of the slope and vegetation extracted in this manner is analyzed and evaluated by the soundness evaluation unit 23.

[0058] The soundness evaluation unit 23 shown in Fig. 1 evaluates the soundness of the point cloud data of the slope extracted by the point cloud data processing system 20. Here, as mentioned above, "soundness" indicates whether or not there is a problem with the function of the structure, or the degree of the problem that has occurred.

[0059] When a slope deteriorates, cracks may appear on the slope. In this case, cracks on the slope can be detected by analyzing the point cloud data of the slope together with other data such as image data.

[0060] However, the point cloud data acquired by the point cloud data acquisition device 10 (MMS) does not have enough accuracy to detect cracks on the slope, so it is not easy to detect cracks using only the point cloud data.

[0061] Therefore, the soundness evaluation unit 23 evaluates the soundness of the slope by the following method: First, the soundness evaluation unit 23 processes the point cloud data for evaluating the soundness of the slope by the method shown in the flowchart in FIG.

[0062] First, in step S21, the soundness evaluation unit 23 generates planes at regular intervals along the road alignment A. Hereinafter, a description will be given with reference to FIG.

[0063] 12 is a perspective view showing the point cloud data I etc. The road soundness evaluation unit 23 generates planes c1 to c5 that cross the road alignment A along the road alignment A at regular intervals.

[0064] Next, in step S22, the soundness evaluation unit 23 determines the top and bottom of the slope from the intersection of the plane and the slope.

[0065] 13 is a perspective view showing the point cloud data I etc. The soundness evaluation unit 23 determines the slope ablation points m1 to m5, which are the highest points of the slope, and the slope toes n1 to n5, which are the lowest points of the slope, from the intersections of the planes c1 to c5 and the point cloud data I of the slope.

[0066] Next, in step S23, the soundness evaluation unit 23 generates a cross section including vegetation.

[0067] Specifically, the soundness evaluation unit 23 specifies a buffer width for each of the planes c1 to c5 shown in FIG. 13, and generates a cross section including point cloud data of the slopes and point cloud data of the vegetation.

[0068] In this way, the soundness evaluation unit 23 identifies the point cloud data of the slope, the point cloud data of the vegetation, and the toe m and toe n of the slope in the point cloud data generated by the method shown in Fig. 2. Furthermore, the soundness evaluation unit 23 identifies, from the point cloud data J of vegetation, vegetation J1 growing thick on the sprayed surface of the slope and vegetation J2 growing thick outside the sprayed surface of the slope (for example, above the toe m) (see Fig. 19). The soundness evaluation unit 23 uses these data to evaluate the soundness of the slope. The soundness of the slope is evaluated based on predetermined indices.

[0069] The index for evaluating the soundness of a slope can be set arbitrarily, but in this embodiment, the index for evaluating the soundness of a slope includes four items: classification by slope shape, information on the relationship with the road, information on vegetation growing from the sprayed surface, and information on vegetation growing on the upper part of the slope (growing from above the toe m) (see FIG. 14). Each item includes multiple parameters.

[0070] [Classification by slope shape] When peeling, falling, or collapsing of a slope occurs, the shape of the slope changes. Therefore, the soundness evaluation unit 23 evaluates the soundness of a slope using the "classification by slope shape" shown in Figure 14 as an index.

[0071] As shown in Figure 14, the items for "Classification by slope shape" include the average slope, slope height, slope extension, slope area, slope direction, maximum slope slope, calculated average roughness of the slope vertical cross section, and overall slope shape classification.

[0072] The "average gradient" is the average value of the angle θ (see Figure 15) between the horizontal plane and the straight line (transverse reference line L) connecting the toe m and toe n of the slope, and is calculated by calculating the angle θ for each transverse reference line L and outputting the average value.

[0073] "Slope height" is the height of the point cloud data of the slope. The slope height is calculated by outputting the difference h2 in the elevation values ​​of the toe m and toe n (see Figure 15).

[0074] The "slope extension" is the length of the slope extended along the road alignment A. The slope extension is calculated by calculating the sum of the distance P1 of the toe n of the slope between the crossing reference lines and the sum of the distance P2 of the toe m of the slope between the crossing reference lines (see Figure 16), and outputting the average of the sum of the distances P1 and P2.

[0075] "Slope area" is the surface area of ​​the point cloud data of the slope. To calculate the area of ​​the slope, first generate points at regular intervals on the crossing reference line L and obtain the neighboring point cloud. Then, generate a TIN (Delaunay triangulation) based on those points and output the total surface area to calculate the area.

[0076] The "direction of the slope" is the normal vector of the slope. The direction of the slope is calculated for each crossing reference line L as a single vector perpendicular to the crossing reference line L. Alternatively, the direction of the slope may be calculated for each TIN.

[0077] The "maximum slope gradient" is the maximum value of the angle θ (see FIG. 15) between the crossing reference line L and the horizontal plane, and is calculated by calculating the angle θ for each crossing reference line L and outputting the maximum value.

[0078] The "arithmetic mean roughness of a vertical cross section of a slope" indicates the degree of unevenness of the slope. The arithmetic mean roughness of a vertical cross section of a slope is calculated by calculating a plane using two adjacent cross-sectional reference lines L and outputting the root mean square (RMS) of the straight-line distance between that plane and each point in the point cloud data of the slope. Alternatively, the arithmetic mean roughness of a vertical cross section of a slope may be calculated for each TIN.

[0079] The "overall shape classification of a slope" is a classification of the overall shape of a slope. To calculate the overall shape classification of a slope, surfaces are generated at regular intervals for the elevation value of the cross-sectional reference line L, and the contour lines (the points that make up the contour lines) are extracted. Then, depending on the interval and extent of the contour lines, it is determined which of the nine slope types shown in Figure 17 described in the literature (Suzuki Ryusuke, Introduction to Topographic Map Reading for Construction Engineers, Vol. 1, Fundamentals of Map Reading, Kokin Shoin, p. 122, 1997) it falls into.

[0080] [Information related to roads] Furthermore, when the slope peels off, falls off, or collapses, soil and sand accumulates on the shoulder, sidewalk, or road on the slope side, causing changes in the width of the shoulder, sidewalk, and road. For this reason, the soundness evaluation unit 23 evaluates the soundness of the slope using the "information related to the road" as an index.

[0081] The item "Information relating to the road" includes the width of the shoulder or sidewalk on the slope side and the width of the road.

[0082] The "width of the road shoulder or sidewalk on the slope side" indicates the width of the road shoulder or sidewalk adjacent to the toe n of the slope. To calculate the width of the road shoulder or sidewalk on the slope side, the road shoulder or sidewalk is extracted from the point cloud data using the area data (point cloud data extracted in step S11). Then, the width is calculated from the point cloud data of the road shoulder or sidewalk near the toe n of the slope.

[0083] "Road width" indicates the width of the roadway portion adjacent to the toe n of the slope. To calculate the road width, the roadway portion is extracted from the point cloud data using the area data (point cloud data extracted in step S11). Then, the width is calculated from the point cloud data of the roadway portion near the toe n of the slope.

[0084] [Information about vegetation growing on the sprayed surface] Figure 18 shows the relationship between the deterioration of a sprayed slope and vegetation. When cracks appear on the surface of the slope (the surface of the sprayed mortar), vegetation grows from these cracks. The cracks then expand as the vegetation grows. As the vegetation grows further, landslides occur in the ground behind it. When strong winds shake the vegetation, the cracks and landslides expand even further. When surface water and groundwater flow into the ground behind it, the entire slope becomes unstable. This can lead to the collapse of the slope and fallen trees, which can cause road traffic to become impeded.

[0085] In this way, as the vegetation grows, the deterioration of the sprayed slope may also progress. For this reason, the soundness evaluation unit 23 evaluates the slope using "information on the vegetation growing from the sprayed slope" as an index.

[0086] The item "Information on vegetation growing on the sprayed surface" includes the vegetation coverage rate (growth rate) and the height of the vegetation relative to the slope.

[0087] "Vegetation coverage rate (ratio of luxuriance)" indicates the rate at which vegetation covers a slope. To calculate the vegetation coverage rate, first calculate the area of ​​the point cloud data for the slope (slope area Q1) and the area of ​​the point cloud data for vegetation J1 growing from the sprayed surface (vegetation area Q2) (see Figure 19). Vegetation area Q2 is calculated by projecting the constituent points of vegetation J1 onto a TIN plane and extracting the outlines of the projected constituent points. The vegetation coverage rate (the surface area of ​​the vegetation area relative to the surface area of ​​the slope area) is calculated from the ratio of the surface area of ​​slope area Q1 to the surface area of ​​vegetation area Q2.

[0088] The "height of vegetation relative to the slope" indicates the height of the vegetation relative to the slope in the direction perpendicular to the slope. To calculate the height of the vegetation relative to the slope, a plane is created using four points: the toe m and toe n of the adjacent crossing reference line L. Then, the point cloud data of the vegetation J1 is projected onto the plane, and the projection distance is calculated. The minimum and maximum, or average, of the projection distances of each point in the point cloud data of the vegetation is then calculated as the "height of the vegetation relative to the slope."

[0089] [Information about vegetation on the top of the slope] Furthermore, if there is vegetation on the upper part of the slope (thriving above the toe m), the vegetation may be shaken by strong winds, putting strain on the slope and leading to its deterioration. Furthermore, the growth of vegetation on the upper part of the slope increases the risk of trees falling. For this reason, the soundness evaluation unit 23 evaluates the slope using "information on vegetation on the upper part of the slope" as an index.

[0090] The "Information on vegetation at the top of the slope" item includes the height of the vegetation at the top of the slope and the length that the vegetation extends beyond the top of the slope.

[0091] "Height of vegetation on the slope toe" indicates the elevation value of the point of vegetation J2 at the top of the slope, located near the slope toe m (see Figure 19).

[0092] The "length of vegetation protruding from the slope abutment" indicates the horizontal length of vegetation J2 protruding from the slope abutment m (see Figure 19). The length of vegetation protruding from the slope abutment is calculated by outputting the length from the slope abutment m to the tip of the vegetation (the length protruding from the slope).

[0093] In this way, the soundness of the slope can be properly evaluated by analyzing and evaluating the point cloud data of the slope and vegetation based on the indices shown in Figure 14. More specifically, a numerical range that is judged to be sound (normal) is set for each index (parameter), and the soundness of the slope can be evaluated based on whether the calculated numerical value falls within that numerical range. This makes it possible to evaluate the soundness of the slope without actually visiting the site.

[0094] Furthermore, even if it is not possible to detect the occurrence of cracks on a slope, it is possible to estimate the presence or absence of cracks by evaluating the rate of vegetation growth, etc. Furthermore, if the rate of vegetation growth is above a certain percentage, it is highly likely that the slope is a natural slope rather than a slope. Therefore, based on the calculated rate of vegetation growth, it is possible to identify whether the extracted point cloud data represents a road structure such as a slope, or a natural object such as a natural slope.

[0095] Although the evaluation may be performed using any one of the parameters of the indexes shown in FIG. 14, the accuracy of the evaluation of the health level is improved by performing the evaluation using a combination of a plurality of parameters.

[0096] As described above, the point cloud data processing system 20 according to this embodiment: a first extraction means (extraction processing unit 22) that performs a first extraction process (step S11 shown in FIG. 2) to extract point cloud data within a rectangle B (first region) including a road alignment A from point cloud data T acquired by a point cloud data acquisition device 10 (a predetermined method); removal means (extraction processing unit 22) for performing removal processing (step S12 shown in FIG. 2) for removing point cloud data of range C (point cloud data of roads) from the point cloud data within the rectangle B extracted by the first extraction means; a projection means (extraction processing unit 22) that performs a projection process (steps S13 and S14 shown in FIG. 2) of projecting the point cloud data after the removal process by the removal means onto a two-dimensional plane that crosses the road alignment A; a second extraction means (extraction processing unit 22) that performs a second extraction process (step S15 shown in FIG. 2) to extract point cloud data that satisfies a predetermined condition as point cloud data of a slope candidate from the point cloud data after the projection process has been performed by the projection means; a third extraction means (extraction processing unit 22) that performs a third extraction process (step S16 shown in FIG. 2 ) to extract point cloud data of the slope candidate from the point cloud data of the slope candidate extracted by the second extraction means by a region growing method starting from a starting point G (a point approximately at the center of the point cloud data of the slope candidate); It is equipped with the following.

[0097] With this configuration, by extracting only the point cloud data of the slope, it becomes easier to evaluate the slope, which in turn makes it easier to maintain and manage the slope.

[0098] Furthermore, the point cloud data processing system 20 according to this embodiment: A dividing means (extraction processing unit 22) is provided for dividing the point cloud data projected onto the two-dimensional plane into a plurality of point cloud data; The second extraction means As the predetermined condition, among the point cloud data divided by the dividing means, the point cloud data in which the angle of the normal vector of the point cloud data is within a predetermined range (for example, 30 to 80 degrees) is extracted as point cloud data of a slope candidate (step S15 shown in Figure 2).

[0099] By configuring in this way, point cloud data of the slope candidate can be extracted with high accuracy.

[0100] Further, the second extraction means As the predetermined condition, among the point cloud data whose normal vector angle is within a predetermined range, the point cloud data whose height is equal to or greater than a certain value is extracted as point cloud data of a slope candidate (step S15 shown in FIG. 2).

[0101] By configuring in this way, point cloud data of the slope candidate can be extracted with higher accuracy.

[0102] Further, the second extraction means If there are multiple point cloud data that satisfy the predetermined conditions, the point cloud data that is closest to the road alignment A is extracted as point cloud data of the slope candidate (step S15 shown in Figure 2).

[0103] By configuring in this way, point cloud data of the slope candidate can be extracted with higher accuracy.

[0104] Moreover, the third extraction means If the area of ​​the point cloud data extracted by the area growing method is smaller than a predetermined value, the point cloud data within that area is excluded from the point cloud data of the slope (step S16 shown in FIG. 2).

[0105] By configuring in this way, it is possible to prevent vegetation from being mistakenly identified as a slope.

[0106] Furthermore, the point cloud data processing system 20 according to this embodiment: The method is provided with a connecting means (extraction processing unit 22) that divides the three-dimensional space containing the point cloud data extracted by the third extraction means into a plurality of grids, and connects adjacent grids among the grids in which point cloud data exists, thereby forming the point cloud data in the connected grids into point cloud data of one slope (step S17 shown in FIG. 2).

[0107] This configuration improves the accuracy of slope extraction. Specifically, even if a slope has a high degree of vegetation growth, it can be identified as a single slope without being mistaken for multiple slopes.

[0108] Furthermore, the point cloud data processing system 20 according to this embodiment: The system includes a vegetation extraction means (extraction processing unit 22) that extracts point cloud data located above the point cloud data of the slope as point cloud data of vegetation (step S18 shown in FIG. 2).

[0109] By configuring in this way, it is possible to calculate the slope area and growth rate even for slopes with thick vegetation.

[0110] Furthermore, the point cloud data processing system 20 according to this embodiment: extraction means (extraction processing unit 22) for performing extraction processing to extract point cloud data including a slope area from the point cloud data T acquired by the point cloud data acquisition device 10 (a predetermined method); a soundness evaluation unit 23 (evaluation means) that evaluates the soundness of the slope based on predetermined indices related to the slope by utilizing the point cloud data extracted by the extraction means; It is equipped with the following.

[0111] By configuring in this way, the soundness of the slope can be appropriately evaluated.

[0112] In addition, the indicators include: It includes information about the shape of the slope.

[0113] By configuring in this way, the soundness of the slope can be evaluated more appropriately.

[0114] The information about the shape of the slope includes: At least one of the average gradient of the slope, the height of the slope, the extension length of the slope, the surface area of ​​the slope, and the maximum gradient of the slope is included.

[0115] By configuring in this way, the soundness of the slope can be evaluated more appropriately.

[0116] In addition, the indicators include: This includes information regarding the relationship between the slope and the road.

[0117] By configuring in this way, the soundness of the slope can be evaluated more appropriately.

[0118] In addition, the information regarding the relationship between the slope and the road includes: At least one of the width of the shoulder or sidewalk on the slope side, or the width of the road is included.

[0119] By configuring in this way, the soundness of the slope can be evaluated more appropriately.

[0120] Furthermore, the point cloud data extracted by the extraction means (extraction processing unit 22) includes point cloud data of vegetation, The indicators include: This includes information about vegetation J1 growing thickly on the sprayed surface of the slope.

[0121] By configuring in this way, the soundness of the slope can be evaluated more appropriately.

[0122] In addition, the information about the vegetation J1 growing on the sprayed surface of the slope includes the following: At least one of the growth rate of the vegetation J1 and the height of the vegetation J1 relative to the slope is included.

[0123] By configuring in this way, the soundness of the slope can be evaluated more appropriately.

[0124] Furthermore, the point cloud data extracted by the extraction means (extraction processing unit 22) includes point cloud data of vegetation, The indicators include: This includes information about vegetation J2 growing on areas other than the sprayed surface of the slope.

[0125] By configuring in this way, the soundness of the slope can be evaluated more appropriately.

[0126] In addition, the information on vegetation J2 growing on areas other than the sprayed surface of the slope includes the following: At least one of the height of the vegetation J2 at the slope shoulder or the protruding length of the vegetation J2 from the slope shoulder m is included.

[0127] By configuring in this way, the soundness of the slope can be evaluated more appropriately.

[0128] Although the embodiment of the present invention has been described above, the present invention is not limited to the above configuration, and various modifications are possible within the scope of the invention described in the claims.

[0129] For example, in this embodiment, the point cloud data acquisition device 10 is an MMS (i.e., the point cloud data processing system 20 processes point cloud data acquired by the MMS), but the method of acquiring point cloud data is not limited to this and can be any method. The point cloud data acquisition device 10 can be, for example, a road surface condition measurement system (LCMS), an airborne LP (aircraft or UAV-mounted laser profiling system), a handheld laser scanner, a wearable laser scanner, a terrestrial laser scanner, or a photo-based mobile mapping system. In particular, in the case of an airborne LP, slopes can be extracted with even greater accuracy by using ground data and original data.

[0130] In this embodiment, the soundness of the slope is evaluated based on whether the output values ​​of the parameters shown in FIG. 14 are within a predetermined range, but the method of evaluating the soundness is not limited to this. For example, the soundness of the slope may be evaluated by comparing past output values ​​with current output values ​​(values ​​obtained over a period of time). If the current output values ​​(especially the output values ​​for "classification by slope shape") have changed compared to past output values, it is possible that the soundness of the slope may have deteriorated. [Explanation of symbols]

[0131] 20 Point Cloud Data Processing System 22 Extraction processing section 23 Soundness Evaluation Department

Claims

1. a first extraction means for performing a first extraction process to extract point cloud data of a first region including a road alignment from point cloud data acquired by a predetermined method; a removal unit that performs a removal process to remove the point cloud data of the road from the point cloud data of the first area extracted by the first extraction unit; a projection means for performing a projection process to project the point cloud data after the removal process has been performed by the removal means onto a two-dimensional plane that crosses the road alignment; a second extraction means for performing a second extraction process to extract point cloud data that satisfies a predetermined condition as point cloud data of a slope candidate from the point cloud data after the projection process has been performed by the projection means; a third extraction means for performing a third extraction process to extract point cloud data of the slope candidate from the point cloud data of the slope candidate extracted by the second extraction means by a region growing method using a point approximately at the center of the point cloud data of the slope candidate as a starting point; Equipped with a dividing means for dividing the point cloud data projected onto the two-dimensional plane into a plurality of point cloud data by labeling based on Euclidean distance, The second extraction means extracting, as the predetermined condition, point cloud data of a slope candidate, from among the point cloud data divided by the dividing means, point cloud data in which angles of normal vectors of each point included in the point cloud data are within a predetermined range; Point cloud data processing system.

2. The second extraction means extracting, as the predetermined condition, point cloud data of which the angle of the normal vector of the point cloud data falls within a predetermined range, point cloud data of which the height is equal to or greater than a certain value, as point cloud data of a slope candidate; The point cloud data processing system according to claim 1 .

3. The second extraction means If there are a plurality of pieces of point cloud data that satisfy the predetermined condition, the point cloud data that is closest to the road alignment is extracted as point cloud data of a slope candidate.

3. The point cloud data processing system according to claim 1.

4. The third extraction means If the area of ​​the point cloud data extracted by the area growing method is smaller than a predetermined value, the point cloud data within the area is excluded from the point cloud data of the slope. The point cloud data processing system according to any one of claims 1 to 3.

5. A connecting means is provided for dividing a three-dimensional space containing the point cloud data extracted by the third extraction means into a plurality of grids, and connecting adjacent grids among the grids in which point cloud data exists, thereby forming the point cloud data in the connected grids into point cloud data of a single slope. The point cloud data processing system according to any one of claims 1 to 4.

6. A vegetation extraction means is provided for extracting point cloud data located above the point cloud data of the slope as point cloud data of vegetation. The point cloud data processing system according to any one of claims 1 to 5.

Images