A ground point cloud density evaluation method based on an irregular triangle net model

Abstract

The application discloses a ground point cloud density evaluation method based on an irregular triangle net model, belongs to the technical field of point cloud data analysis, and can solve the problem that the existing evaluation method leads to a virtual high ground point cloud density due to the neglect of terrain undulation. The method comprises the following steps: S1, according to the ground point cloud data and image information of a target region, the target region is divided into a point cloud abnormal missing area and a point cloud evaluation area, and the ground point cloud density of the point cloud abnormal missing area is evaluated as unqualified; S2, according to the ground point cloud data of the point cloud evaluation area, an irregular triangle net model of the point cloud evaluation area is constructed, and a plurality of grid units are divided on the irregular triangle net model according to rectangular framing; S3, the terrain in the grid unit is determined, and according to the terrain in the grid unit and whether the local triangle net formed by the irregular triangle net model in the grid unit is qualified, the ground point cloud density corresponding to the grid unit is evaluated. The application is used for evaluating the ground point cloud density of a target region.

Description

Technical Field

[0001] The invention relates to a ground point cloud density evaluation method based on an irregular triangulated network model, and belongs to the technical field of point cloud data analysis. Background Art

[0002] In the field of geographic information, by using equipment such as lidar to emit laser pulses to the ground and receive echo signals, ground point cloud data can be obtained, geographic information data can be collected, and a data basis can be provided for terrain modeling.

[0003] Ground point cloud data is a discrete set of three-dimensional ground points. The distribution density of these points, or the density of the ground point cloud, directly impacts the accuracy of terrain modeling. High-density ground point cloud data not only allows for the detailed depiction of micro-topographic features such as gullies and steep slopes, but also helps reduce elevation interpolation errors during the modeling process, significantly improving the accuracy and fidelity of terrain modeling. Therefore, it is essential to assess the density of the ground point cloud before terrain modeling.

[0004] Existing methods for evaluating ground point cloud density are mostly based on the principle of two-dimensional plane projection. This involves projecting a three-dimensional ground point cloud onto a horizontal plane and calculating the ground point cloud density based on the number of ground points per unit projected area. However, in complex terrain areas such as steep slopes and valleys, the actual surface area of ​​the terrain is much larger than its horizontal projection area. Existing evaluation methods often ignore the terrain undulations, resulting in an inflated calculated ground point cloud density. This method fails to truly reflect the spatial distribution characteristics of ground point cloud data on complex terrain surfaces, compromising terrain modeling accuracy. Summary of the Invention

[0005] The present invention provides a ground point cloud density assessment method based on an irregular triangulated network model, which can solve the problem that the existing assessment method leads to an inflated ground point cloud density due to ignoring terrain undulations.

[0006] The present invention provides a ground point cloud density assessment method based on an irregular triangulated network model, the method comprising:

[0007] S1. Dividing the target area into a point cloud anomaly missing area and a point cloud evaluation area based on the ground point cloud data and image information of the target area, and evaluating the ground point cloud density of the point cloud anomaly missing area as unqualified;

[0008] S2. constructing an irregular triangulated network model of the point cloud evaluation area according to the ground point cloud data of the point cloud evaluation area, and dividing the irregular triangulated network model into a plurality of grid units according to rectangular sections;

[0009] S3. Determine the terrain within the grid cell, and evaluate whether the ground point cloud density corresponding to the grid cell is qualified based on the terrain within the grid cell and the local triangulated network formed by the irregular triangulated network model within the grid cell.

[0010] Optionally, S1 divides the target area into a point cloud anomaly missing area and a point cloud evaluation area according to the ground point cloud data and image information of the target area, specifically including:

[0011] Dividing the target area into a point cloud data missing area and a non-missing area according to the ground point cloud data and image information of the target area;

[0012] According to the data missing factor of the point cloud data missing area, the point cloud data missing area is divided into a reasonable point cloud missing area and an abnormal point cloud missing area; the non-missing area and the reasonable point cloud missing area are used as point cloud evaluation areas.

[0013] Optionally, the data missing factor includes reflection missing; dividing the point cloud data missing area into a reasonable point cloud missing area and an abnormal point cloud missing area specifically includes:

[0014] The missing area in the point cloud data missing area formed based on the reflection missing is divided into a point cloud reasonable missing area, and the remaining area in the point cloud data missing area is divided into a point cloud abnormal missing area.

[0015] Optionally, constructing an irregular triangulated network model of the point cloud evaluation area according to the ground point cloud data of the point cloud evaluation area in S2 specifically includes:

[0016] According to the ground point cloud data of the point cloud evaluation area, an irregular triangulated network model of the point cloud evaluation area is constructed using the Delaunay triangulation method.

[0017] Optionally, constructing an irregular triangulated network model of the point cloud evaluation area using the Delaunay triangulation method specifically includes:

[0018] Using the Delaunay triangulation method, an initial irregular triangulated network of the point cloud evaluation area is constructed according to preset conditions; the preset conditions are that two vertices of the triangular mesh at the ridge are different elevation maxima points, and two vertices of the triangular mesh at the valley are different elevation minima points;

[0019] The initial irregular triangulated network is modified according to the reasonable missing area of ​​the point cloud to obtain an irregular triangulated network model of the point cloud evaluation area.

[0020] Optionally, S3 evaluates whether the ground point cloud density corresponding to the grid cell is qualified based on the terrain within the grid cell and the local triangulated network formed by the irregular triangulated network model within the grid cell, specifically including:

[0021] Determining the three-dimensional average point spacing, three-dimensional surface area, and horizontal projection area of ​​each type of terrain in the grid cell based on the local triangulated network formed by the irregular triangulated network model in the grid cell;

[0022] Whether the three-dimensional point cloud density of each terrain in the grid unit is qualified is determined based on the three-dimensional average point spacing, the three-dimensional surface area and the horizontal projection area, and when the three-dimensional point cloud density of all terrains is qualified, the ground point cloud density corresponding to the grid unit is evaluated as qualified.

[0023] Optionally, determining whether the three-dimensional point cloud density of each terrain in the grid unit is qualified according to the three-dimensional average point spacing, the three-dimensional surface area, and the horizontal projection area specifically includes:

[0024] Determining a discrimination point cloud density and a three-dimensional point cloud density of each terrain in the grid unit according to the three-dimensional average point spacing and the three-dimensional surface area respectively;

[0025] A point cloud density threshold for each terrain in the grid unit is determined based on the three-dimensional surface area and the horizontal projection area, and when the discrimination point cloud density and the three-dimensional point cloud density of the terrain are both greater than the point cloud density threshold, the three-dimensional point cloud density of the terrain is determined to be qualified.

[0026] Optionally, determining the discrimination point cloud density of each terrain type in the grid unit according to the three-dimensional average point spacing specifically includes:

[0027] The inverse of the cube of the three-dimensional average point spacing is determined as the discrimination point cloud density corresponding to the terrain in the grid unit.

[0028] Optionally, determining a point cloud density threshold for each type of terrain in the grid unit according to the three-dimensional surface area and the horizontal projection area specifically includes:

[0029] Determining a ratio of the three-dimensional surface area to the horizontal projection area as the terrain complexity of the corresponding terrain within the grid cell;

[0030] A point cloud density threshold corresponding to the terrain in the grid unit is determined according to the terrain complexity and a preset density standard.

[0031] Optionally, after S3, the method further includes:

[0032] Dividing a plurality of terrain areas on the irregular triangulated network model according to different terrains;

[0033] When the ground point cloud densities of all grid cells corresponding to the terrain area are evaluated to be qualified, the ground point cloud density of the terrain area is evaluated to be qualified.

[0034] The beneficial effects that the present invention can produce include:

[0035] The present invention divides the target area into a point cloud anomaly missing area and a point cloud evaluation area, and evaluates the ground point cloud density in the point cloud anomaly missing area as unqualified. The point cloud data anomaly missing area can be directly excluded from the subsequent refined evaluation, which is beneficial to improving the accuracy and efficiency of the evaluation.

[0036] The present invention constructs an irregular triangulated network model of the point cloud assessment area, divides it into several grid cells, and then evaluates the ground point cloud density corresponding to each grid cell based on the terrain within it. This not only enables a zonal assessment of the point cloud assessment area, improving the refinement of the assessment, but also fully considers the impact of terrain undulations on ground point cloud density during the assessment process, truly reflecting the spatial distribution characteristics of ground point cloud data on complex terrain surfaces, thereby improving the accuracy of ground point cloud density assessment.

[0037] The present invention determines the discriminant point cloud density of each terrain type within a grid cell based on the three-dimensional average point spacing of that terrain, and determines the point cloud density threshold for that terrain based on its terrain complexity. The method then combines the discriminant point cloud density and the point cloud density threshold to comprehensively assess whether the terrain's three-dimensional point cloud density is qualified. If the three-dimensional point cloud density of all terrain types within that grid cell is qualified, the ground point cloud density corresponding to that grid cell is assessed as qualified. This allows the assessment process to take into account both the impact of terrain factors and the impact of point spacing on ground point cloud density, effectively improving assessment accuracy when point cloud distribution is uneven. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Figure 1 A flow chart of a method for evaluating ground point cloud density based on an irregular triangulated network model provided in an embodiment of the present invention;

[0039] Figure 2 A partial schematic diagram of an irregular triangulated network model provided by an embodiment of the present invention;

[0040] Figure 3 An overall schematic diagram of an irregular triangulated network model provided by an embodiment of the present invention;

[0041] Figure 4 The embodiment of the present invention provides Figure 3 A top view of

[0042] Figure 5A schematic diagram of square frames provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0043] The present invention is described in detail below with reference to the embodiments, but the present invention is not limited to these embodiments.

[0044] The embodiment of the present invention provides a ground point cloud density evaluation method based on an irregular triangulated network model, such as Figure 1 As shown, the method includes:

[0045] S1. Divide the target area into a point cloud anomaly missing area and a point cloud evaluation area based on the ground point cloud data and image information of the target area, and evaluate the ground point cloud density of the point cloud anomaly missing area as unqualified.

[0046] Generally, point cloud data collected by equipment such as lidar is a discrete set of three-dimensional points. This set typically consists of two types: ground points and non-ground points. Ground points primarily reflect terrain features, while non-ground points primarily reflect features of objects such as vegetation, buildings, and power lines. The collection of all ground points constitutes ground point cloud data. This embodiment of the present invention evaluates ground point cloud data for a target area.

[0047] In S1, the target area is divided into a point cloud anomaly missing area and a point cloud assessment area based on the ground point cloud data and image information of the target area, which may include:

[0048] According to the ground point cloud data and image information of the target area, the target area is divided into a point cloud data missing area and a non-missing area;

[0049] According to the data missing factor of the point cloud data missing area, the point cloud data missing area is divided into the point cloud reasonable missing area and the point cloud abnormal missing area; the non-missing area and the point cloud reasonable missing area are used as the point cloud evaluation area.

[0050] Specifically, data missing factors include reflection missing, occlusion missing and other missing factors. Among them, reflection missing refers to the situation where the ground object has a low reflectivity and the acquisition equipment cannot receive the effective signal reflected by the ground object, resulting in the missing of ground point cloud data at the location of the ground object, such as large areas of water, asphalt road surfaces, and coal accumulation areas. Occlusion missing refers to the situation where vegetation, buildings, etc. block the ground, resulting in the missing of ground point cloud data in the blocked area. Other missing factors refer to the missing of ground point cloud data due to reasons other than reflection missing and occlusion missing, such as the missing of ground point cloud data during the non-dry period of the surface.

[0051] The point cloud data missing area is divided into reasonable point cloud missing area and abnormal point cloud missing area, which can specifically include:

[0052] The missing area in the point cloud data missing area formed based on reflection missing is divided into the point cloud reasonable missing area, and the remaining area in the point cloud data missing area is divided into the point cloud abnormal missing area.

[0053] Specifically, image information includes topographic maps, image maps, and the like. This embodiment uses one or both of the existing topographic maps and image maps of the target area as a base map, overlaying and analyzing the base map with ground point cloud data. This allows identification of point cloud data missing and non-missing areas within the target area, and further identifies data missing factors for different areas within the point cloud data missing areas, thereby dividing the point cloud data missing areas into reasonable point cloud missing areas and abnormal point cloud missing areas. This embodiment prioritizes the use of large-scale, current topographic maps for quickly acquiring the extent of large water areas, roads, and the like. Image maps prioritize the latest, high-resolution images.

[0054] This embodiment divides the target area into a point cloud anomaly missing area and a point cloud evaluation area, and evaluates the ground point cloud density in the point cloud anomaly missing area as unqualified. This allows the point cloud data anomaly missing area to be directly excluded from subsequent refined evaluations, which is beneficial to improving the accuracy and efficiency of the evaluation.

[0055] S2. Construct an irregular triangulated network model of the point cloud assessment area based on the ground point cloud data of the point cloud assessment area, and divide the irregular triangulated network model into multiple grid units according to rectangular partitioning.

[0056] In S2, an irregular triangulated network model of the point cloud assessment area is constructed based on the ground point cloud data of the point cloud assessment area, which may specifically include:

[0057] According to the ground point cloud data of the point cloud assessment area, the irregular triangulated network model of the point cloud assessment area is constructed using the Delaunay triangulation method.

[0058] The above-mentioned irregular triangulated network model of the point cloud assessment area constructed using the Delaunay triangulation method may specifically include:

[0059] The Delaunay triangulation method is used to construct the initial irregular triangulated network of the point cloud evaluation area according to the preset conditions; the initial irregular triangulated network is corrected according to the reasonable missing area of ​​the point cloud to obtain the irregular triangulated network model of the point cloud evaluation area.

[0060] Specifically, the preset conditions may include:

[0061] 1) Generate continuous triangular meshes with the shortest distance between adjacent points, and the edge formed by connecting two adjacent points can connect at most two triangular meshes, and maximize the minimum internal angle of the triangular mesh;

[0062] 2) The triangular mesh at the ridge connects the elevation maximum points, and the triangular mesh at the valley connects the elevation minimum points. That is, the two vertices of the triangular mesh at the ridge are different elevation maximum points, and the two vertices of the triangular mesh at the valley are different elevation minimum points.

[0063] The above correction of the initial irregular triangulated network based on the reasonable missing area of ​​the point cloud can specifically include:

[0064] Based on the boundaries of the reasonable missing point cloud area, the lines formed by the initial irregular triangulation within the reasonable missing point cloud area are deleted. This is because there is missing point cloud data in the reasonable missing point cloud area. The part of the initial irregular triangulation covering the reasonable missing point cloud area cannot represent the actual ground information and may interfere with subsequent evaluations. Therefore, it should be deleted. At the same time, repeated or intersecting triangular meshes in the initial irregular triangulation should be deleted. This will result in an irregular triangulation model of the point cloud evaluation area.

[0065] For example, a partial schematic diagram of the irregular triangulated network model in this embodiment is as follows: Figure 2 As shown, the irregular triangulated network model is composed of multiple adjacent triangular meshes. The vertices of the triangular meshes are discrete points in the ground point cloud data, and their coordinates can be determined based on the ground point cloud data. Figure 2 The serial number, vertex and adjacent triangle mesh information of each triangle mesh are shown in Table 1, and the vertex coordinates of the triangle mesh are shown in Table 2.

[0066] Table 1 Sequence number, vertex and adjacent triangle meshes of each triangle mesh

[0067]

[0068] Table 2 Vertex coordinates of triangle mesh

[0069]

[0070] The overall schematic diagram of the irregular triangulated network model of the target area generated by the above method in this embodiment is as follows Figure 3 and Figure 4 As shown.

[0071] Then, this embodiment divides the irregular triangulated network model into multiple grid units according to rectangular framing. The size of each grid unit can be 50cm×50cm or 40cm×50cm, etc. This embodiment takes a 50cm×50cm grid unit as an example for explanation, that is, this embodiment takes a square framing as an example for explanation. The schematic diagram of the square framing is shown as follows: Figure 5 As shown, Figure 5 The size of each grid unit is 50cm×50cm.

[0072] The embodiment can realize the partition evaluation of the point cloud evaluation area by constructing the irregular triangle mesh model of the point cloud evaluation area, dividing a plurality of grid cells on the irregular triangle mesh model, and evaluating the point cloud density of each grid cell, and is beneficial to improve the refinement of the evaluation.

[0073] S3, determine the terrain in the grid cell, and determine whether the ground point cloud density corresponding to the grid cell is qualified according to the terrain in the grid cell and the local triangle mesh formed by the irregular triangle mesh model in the grid cell.

[0074] In S3, whether the ground point cloud density corresponding to the grid cell is qualified according to the terrain in the grid cell and the local triangle mesh formed by the irregular triangle mesh model in the grid cell, specifically includes:

[0075] According to the local triangle mesh formed by the irregular triangle mesh model in the grid cell, determine the three-dimensional average point distance, three-dimensional surface area and horizontal projection area of each terrain in the grid cell.

[0076] Specifically, the terrain includes flat land, hilly land, mountainous land and high mountainous land, and different terrains are divided into different local areas in the grid cell. It is worth noting that in the same grid cell, the local areas divided by the same terrain may be concentrated to form an integral area, or may be scattered to form a plurality of scattered sub-local areas in the grid cell, although these sub-local areas are not connected together, but they all belong to the same terrain. For example, different positions in the same grid cell may have multiple flat lands, each flat land is a sub-local area of the flat land terrain, and the collection of all sub-local areas constitutes the local area of the flat land terrain.

[0077] In practical applications, the judgment criteria of different terrains can be determined according to the slope and height difference.

[0078] For example, the terrain judgment criteria can be:

[0079] 1) Flat land: slope < 2° or height difference < 40m;

[0080] 2) Hilly land: slope 2°-6° or height difference 40-300m;

[0081] 3) Mountainous land: slope 6°-25° or height difference 300-1000m;

[0082] 4) High mountainous land: slope ≥ 25° or height difference > 1000m.

[0083] In the above judgment criteria, the slope is prior to the height difference, that is, when the slope index is used for judgment, the slope index is used for judgment.

[0084] Specifically, the three-dimensional average point spacing of each terrain refers to the average value of the sum of the three-dimensional spatial distances of all adjacent points in the local area corresponding to the terrain in the grid unit. The calculation process is as follows: specify a point in the area corresponding to the terrain in the grid unit, then search for its neighboring points and calculate the three-dimensional spatial distance between the two, then calculate the three-dimensional spatial distance of all adjacent points point by point according to the topological relationship of the point cloud neighborhood, and finally sum up all the three-dimensional spatial distances and take the average value to obtain the three-dimensional average point spacing of each terrain.

[0085] Specifically, suppose that a local area formed by a certain terrain in a certain grid cell contains triangular mesh, the calculation process of the three-dimensional surface area of ​​the terrain is as follows:

[0086] 1) Calculate the side length of each triangle mesh using the following formula:

[0087] (1)

[0088] In formula (1), For the The first triangle mesh The length of the side, is a positive integer, and 1≤ ≤3; 、 and Respectively The first triangle mesh The first vertex of the edge coordinate, Coordinates and coordinate; 、 and Respectively The first triangle mesh The second vertex of the edge coordinate, Coordinates and coordinate; =1,2,… .

[0089] 2) Calculate the semi-perimeter of each triangle mesh using the following formula:

[0090] (2)

[0091] In formula (2), For the The semi-perimeter of the triangle mesh; 、 and Respectively The lengths of the first, second, and third sides of a triangle mesh; =1,2,… .

[0092] 3) Calculate the three-dimensional surface area of ​​each triangular mesh using the following formula:

[0093] (3)

[0094] In formula (3), For the The three-dimensional surface area of ​​a triangular mesh; For the The semi-perimeter of the triangle mesh; 、 and Respectively The lengths of the first, second, and third sides of a triangle mesh; =1,2,… .

[0095] 4) Calculate the three-dimensional surface area of ​​the terrain using the following formula:

[0096] (4)

[0097] In formula (4), is the three-dimensional surface area of ​​the terrain, that is, the sum of the three-dimensional surface areas of all triangular meshes contained in the local area formed by the terrain; For the The three-dimensional surface area of ​​a triangular mesh; The total number of triangle meshes contained in the local area formed for this terrain; =1,2,… .

[0098] Specifically, the calculation process of the horizontal projection area of ​​the terrain is as follows:

[0099] 1) Calculate the horizontal projection area of ​​each triangular mesh using the following formula:

[0100] (5)

[0101] In formula (5), For the The horizontal projection area of ​​a triangular mesh; and Respectively The first vertex of the triangle mesh Coordinates and coordinate; and Respectively The second vertex of the triangle mesh Coordinates and coordinate; and Respectively The third vertex of the triangle mesh Coordinates and coordinate; =1,2,… .

[0102] 2) Calculate the horizontal projection area of ​​the terrain using the following formula:

[0103] (6)

[0104] In formula (6), is the horizontal projection area of ​​the terrain, that is, the sum of the horizontal projection areas of all triangular meshes contained in the local area formed by the terrain; For the The horizontal projection area of ​​a triangular mesh; The total number of triangle meshes contained in the local area formed for this terrain; =1,2,… .

[0105] Furthermore, whether the three-dimensional point cloud density of each terrain in the grid unit is qualified is determined based on the three-dimensional average point spacing, three-dimensional surface area and horizontal projection area of ​​each terrain. When the three-dimensional point cloud density of all terrains in a certain grid unit is qualified, the ground point cloud density corresponding to the grid unit is evaluated as qualified.

[0106] The above determination of whether the 3D point cloud density of each terrain in the grid cell is qualified based on the 3D average point spacing, 3D surface area and horizontal projection area of ​​each terrain may specifically include:

[0107] Determine the discriminant point cloud density and 3D point cloud density of each terrain in the grid cell according to the 3D average point spacing and 3D surface area respectively;

[0108] The point cloud density threshold of each terrain in the grid unit is determined according to the three-dimensional surface area and the horizontal projection area. When the discrimination point cloud density and the three-dimensional point cloud density of the terrain are both greater than the point cloud density threshold, the three-dimensional point cloud density of the terrain is determined to be qualified.

[0109] The above determination of the point cloud density of each terrain in the grid cell based on the three-dimensional average point spacing may specifically include:

[0110] The inverse of the cube of the three-dimensional average point spacing is determined as the discriminant point cloud density corresponding to the terrain in the grid cell.

[0111] Specifically, if there are multiple terrains in the grid cell, the point cloud density of the first terrain is calculated according to the following formula: The formula for calculating the point cloud density of the first terrain is as follows:

[0112] (7)

[0113] In formula (7), is the point cloud density of the first terrain; is the three-dimensional average point distance of the first terrain. The formula for calculating the point cloud density of the first terrain is as follows:

[0114] The formula for calculating the point cloud density of the first terrain is as follows:

[0115] (8)

[0116] In formula (8), is the point cloud density of the first terrain; is the number of vertices of all triangular meshes contained in the first sub-local area of the first terrain; is the sum of the number of vertices of all triangular meshes contained in all sub-local areas of the first terrain; is the three-dimensional surface area of the first sub-local area of the first terrain; is the sum of the three-dimensional surface areas of all sub-local areas of the first terrain. The above determination of the point cloud density threshold of each terrain in the grid cell according to the three-dimensional surface area and the horizontal projection area can specifically include: determining the ratio of the three-dimensional surface area and the horizontal projection area of each terrain as the terrain complexity of the corresponding terrain in the grid cell;

[0117] determining the point cloud density threshold of the corresponding terrain in the grid cell according to the terrain complexity and the preset density standard.

[0118] Specifically, the formula for calculating the terrain complexity of the first terrain is as follows:

[0119]

[0120] In formula (9), is the terrain complexity of the first terrain;

[0121] (9)

[0122] In formula (9), is the terrain complexity of the first terrain; is the three-dimensional average point distance of the first terrain. ​​​​​​​​​Type of terrain The three-dimensional surface area of ​​the local region of the sub- No. The sum of the three-dimensional surface areas of all sub-local regions of the terrain; For the Type of terrain The horizontal projection area of ​​each sub-local area; No. The sum of the horizontal projection areas of all sub-local areas of a terrain.

[0123] Specifically, no. The calculation formula for the point cloud density threshold of the terrain is:

[0124] (10)

[0125] In formula (10), For the The point cloud density threshold of the terrain; It is the standard value of the 2D plane point cloud density, which can be pre-set according to relevant specifications or project requirements; For the The terrain complexity of the terrain.

[0126] This embodiment uses the complexity of the terrain Standard value for the density of two-dimensional plane point cloud Adjust the point cloud density threshold It can objectively reflect the impact of terrain factors on point cloud density, thereby combining the complexity of the terrain with the point cloud density threshold to improve the assessment accuracy of ground point cloud density.

[0127] In this embodiment, when and When The 3D point cloud density of the terrain is determined to be qualified; when the 3D point cloud density of all terrains in a certain grid cell is qualified, the ground point cloud density corresponding to the grid cell is evaluated as qualified.

[0128] This embodiment determines the discriminant point cloud density of each terrain type within the grid cell based on the three-dimensional average point spacing of the terrain, and determines the point cloud density threshold of the terrain based on the terrain complexity of the terrain. Then, the discriminant point cloud density and the point cloud density threshold are combined to comprehensively evaluate whether the three-dimensional point cloud density of the terrain is qualified. In this way, the evaluation process can take into account both the impact of terrain factors and the impact of point spacing factors on the ground point cloud density. When faced with the situation where the point cloud is locally concentrated, the deviation between the calculated ground point cloud density during the evaluation and the actual situation can be effectively reduced, thereby improving the accuracy of the evaluation.

[0129] In comparison, if only terrain factors are considered and the point spacing factors are ignored, when the point cloud is distributed in a localized concentration, that is, only certain sub-local areas have qualified actual point cloud densities due to dense point clouds, and other sub-local areas have unqualified actual point cloud densities due to sparse point clouds, if only terrain factors are considered and the point spacing factors are ignored, then the calculated ground point cloud density actually reflects the average point cloud density within the terrain, which may be qualified. This masks the information that certain sub-local areas are unqualified and cannot truly reflect the actual point cloud density in different sub-local areas of the same terrain, resulting in large errors in the evaluation results. Therefore, this embodiment considers both terrain factors and point spacing factors during the evaluation process, which can effectively improve the evaluation accuracy when the point cloud is unevenly distributed.

[0130] Furthermore, for grid cells whose ground point cloud density assessment is unqualified, this embodiment can carry out supplementary measurement of ground point cloud data according to the requirements of the engineering project.

[0131] After S3, the method may further include:

[0132] Divide various terrain areas on the irregular triangulated network model according to different terrains;

[0133] When the ground point cloud density of all grid cells corresponding to a certain terrain area is evaluated as qualified, the ground point cloud density of the terrain area is evaluated as qualified.

[0134] By dividing the terrain areas on the irregular triangulated network model and evaluating each terrain area, the terrain in the point cloud evaluation area can be classified from a macro perspective, and whether the overall point cloud density of each terrain in the point cloud evaluation area is qualified can be determined from a macro perspective, so that decision makers can understand the macro collection effect of the point cloud data of each terrain.

[0135] This embodiment divides the target area into a point cloud anomaly missing area and a point cloud evaluation area, and evaluates the ground point cloud density in the point cloud anomaly missing area as unqualified. The point cloud data anomaly missing area can be directly excluded from the subsequent refined evaluation, which is beneficial to improving the accuracy and efficiency of the evaluation.

[0136] This embodiment constructs an irregular triangulated network model of the point cloud assessment area and divides it into several grid cells. The density of the ground point cloud corresponding to each grid cell is then evaluated based on the terrain within that grid cell. This not only enables a zonal assessment of the point cloud assessment area, improving the refinement of the assessment, but also fully considers the impact of terrain undulation on ground point cloud density during the assessment process. This allows for a true reflection of the spatial distribution characteristics of ground point cloud data on complex terrain surfaces, thereby improving the accuracy of ground point cloud density assessments.

[0137] This embodiment determines the discriminant point cloud density of each terrain type within a grid cell based on the average 3D point spacing of each type of terrain within the grid cell, and determines the point cloud density threshold for that terrain based on its terrain complexity. The discriminant point cloud density and the point cloud density threshold are then combined to comprehensively assess whether the terrain's 3D point cloud density is qualified. If the 3D point cloud density of all terrain types within the grid cell is qualified, the ground point cloud density corresponding to that grid cell is assessed as qualified. This allows the assessment process to consider both the impact of terrain factors and the impact of point spacing on ground point cloud density, effectively improving assessment accuracy when point cloud distribution is uneven.

[0138] The above descriptions are merely a few embodiments of the present application and do not constitute any form of limitation to the present application. Although the present application discloses the preferred embodiments as above, they are not intended to limit the present application. Any technical personnel familiar with the present profession, without departing from the scope of the technical solution of the present application, using the technical content disclosed above to make slight changes or modifications are equivalent to equivalent implementation cases and fall within the scope of the technical solution.

Claims

1. A ground point cloud density assessment method based on an irregular triangulated network model, characterized in that: The method comprises: S1. Divide the target area into a point cloud data missing area and a non-missing area based on the ground point cloud data and image information of the target area; divide the point cloud data missing area into a reasonable point cloud missing area and an abnormal point cloud missing area based on the data missing factor of the point cloud data missing area; use the non-missing area and the reasonable point cloud missing area as point cloud evaluation areas, and evaluate the ground point cloud density of the abnormal point cloud missing area as unqualified; S2. Based on the ground point cloud data of the point cloud assessment area, construct an initial irregular triangulated network of the point cloud assessment area using the Delaunay triangulation method and in accordance with preset conditions; the preset conditions are that two vertices of the triangular mesh at the ridge are different elevation maxima points, and two vertices of the triangular mesh at the valley are different elevation minima points; Correcting the initial irregular triangulated network according to the reasonable missing area of ​​the point cloud to obtain an irregular triangulated network model of the point cloud evaluation area, and dividing the irregular triangulated network model into a plurality of grid units according to rectangular framing; S3. Determine the terrain within the grid cell, and determine the three-dimensional average point spacing, three-dimensional surface area, and horizontal projection area of ​​each type of terrain within the grid cell based on the local triangulated network formed by the irregular triangulated network model within the grid cell; Whether the three-dimensional point cloud density of each terrain in the grid unit is qualified is determined based on the three-dimensional average point spacing, the three-dimensional surface area and the horizontal projection area, and when the three-dimensional point cloud density of all terrains is qualified, the ground point cloud density corresponding to the grid unit is evaluated as qualified.

2. The method according to claim 1, characterized in that The data missing factor includes reflection missing; the point cloud data missing area is divided into a reasonable point cloud missing area and an abnormal point cloud missing area, specifically including: The missing area in the point cloud data missing area formed based on the reflection missing is divided into a point cloud reasonable missing area, and the remaining area in the point cloud data missing area is divided into a point cloud abnormal missing area.

3. The method according to claim 1, characterized in that Determining whether the three-dimensional point cloud density of each terrain in the grid unit is qualified according to the three-dimensional average point spacing, the three-dimensional surface area, and the horizontal projection area, specifically includes: Determining a discrimination point cloud density and a three-dimensional point cloud density of each terrain in the grid unit according to the three-dimensional average point spacing and the three-dimensional surface area respectively; A point cloud density threshold for each terrain in the grid unit is determined based on the three-dimensional surface area and the horizontal projection area, and when the discrimination point cloud density and the three-dimensional point cloud density of the terrain are both greater than the point cloud density threshold, the three-dimensional point cloud density of the terrain is determined to be qualified.

4. The method according to claim 3, characterized in that Determining the discrimination point cloud density of each terrain type in the grid unit according to the three-dimensional average point spacing specifically includes: The inverse of the cube of the three-dimensional average point spacing is determined as the discrimination point cloud density corresponding to the terrain in the grid unit.

5. The method according to claim 3, characterized in that Determining a point cloud density threshold for each type of terrain within the grid unit according to the three-dimensional surface area and the horizontal projection area specifically includes: Determining a ratio of the three-dimensional surface area to the horizontal projection area as the terrain complexity of the corresponding terrain within the grid cell; A point cloud density threshold corresponding to the terrain in the grid unit is determined according to the terrain complexity and a preset density standard.

6. The method according to claim 3, characterized in that After S3, the method further includes: Dividing a plurality of terrain areas on the irregular triangulated network model according to different terrains; When the ground point cloud densities of all grid cells corresponding to the terrain area are evaluated to be qualified, the ground point cloud density of the terrain area is evaluated to be qualified.

Images