Method for automatically extracting contour line of ground object in slice layering manner

Through the automatic extraction of land object contour lines using slice layered automatic extraction method, the existing problems of low surveying and mapping efficiency and unstable accuracy are solved, and efficient and accurate building edge extraction is achieved, replacing manual collection.

CN119068336APending Publication Date: 2024-12-03XIAMEN TUCHEN INFORMATION TECH CO LTD
View PDF 8 Cites 0 Cited by

Patent Information

Application Number
CN202411170829.6
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-26
Publication Date
2024-12-03

AI Technical Summary

Technical Problem

The existing methods of mapping large-area and large-scale topographic maps are inefficient and unstable in mapping accuracy, and manual processing affects data quality.

Method used

The automatic extraction of land object contour lines is adopted by the slice-layered method. By obtaining surface models, three-dimensional cutting layering, assigning point cloud elevation values, connecting them into vector lines and filtering irregular connections, the vector lines are optimized to extract the contour vector edges of the building. .

Benefits of technology

It improves the efficiency of building mapping and mapping accuracy, replaces manual manual collection, and can effectively deal with land boundaries of different heights and shapes.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119068336A_ABST
    Figure CN119068336A_ABST
Patent Text Reader

Abstract

The invention discloses a method for automatically extracting a contour line of a ground object in a slice layering manner. The method comprises the following steps: acquiring a surface model, and exporting ground elevation point classification data; carrying out three-dimensional cutting layering on the surface model based on ground elevation point classification data; based on a three-dimensional cutting and layering result, automatically endowing the point cloud elevation in each layer of data with an elevation value of a starting point in the layer to form dense point clouds with the same elevation value; connecting the dense point clouds of each layer into a vector line based on an algorithm, and filtering irregular connecting lines; and optimizing all the vector lines, and extracting contour vector side lines of the building and the structure. According to the method for automatically cutting and layering the point cloud and automatically extracting the outer contour vector line of the building and the structure, the mapping efficiency of the building and the structure can be improved, and the mapping precision is improved. And meanwhile, the scheme can replace the current manual acquisition and manufacturing mode.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention relates to the technical field of automatic extraction of building outline vector edges, and in particular to an automatic batch extraction method. Background Art

[0002] At present, the main method of mapping large-scale and large-scale topographic maps is to use professional technical means such as aerial photography, photogrammetry and remote sensing. Moreover, these mapping methods require a lot of manpower and material resources to draw topographic objects, have low work efficiency, and unstable mapping accuracy. However, the existing point cloud data obtained from the ground is of high quality, but there are human factors in the later data processing process that affect the final mapping quality. Summary of the invention

[0003] In view of this, the purpose of the present invention is to propose a method for automatically extracting the contour lines of objects in a slice-by-layer manner, which can greatly improve the surveying and mapping efficiency and mapping accuracy of buildings and structures.

[0004] According to one aspect of the present invention, a method for automatically extracting contour lines of objects by slicing and layering is provided, comprising:

[0005] Obtain the surface model and export ground elevation point classification data;

[0006] The surface model is cut and layered in three dimensions based on the ground elevation point classification data;

[0007] Based on the results of stereoscopic cutting and stratification, the elevation of the point cloud in each layer of data is automatically assigned to the elevation value of the starting point in the layer, forming a dense point cloud with the same elevation value;

[0008] Based on the algorithm, the dense point clouds of each layer are connected into vector lines, and irregular lines are filtered;

[0009] All vector lines are optimized to extract the vector edges of building outlines.

[0010] In the above technical solution, the method provided by the present invention for automatically cutting and layering point clouds and automatically extracting vector lines of the outer contour of buildings and structures can improve the surveying and mapping efficiency of buildings and structures and improve the mapping accuracy. At the same time, the solution can replace the current manual collection and production method.

[0011] In some embodiments, the surface model is a three-dimensional surface model, or a digital surface model.

[0012] In the above technical solution, there is no need to extract the building facade points or project the point cloud onto a two-dimensional plane. The mesh model is directly cut and layered in three dimensions. Because each real land feature has different sizes and shapes at different heights, the outer contour lines are different from the building boundary segments. This solves a variety of different land features (such as curved houses, Guangzhou Tower and rooftop buildings, street lights, electric poles, trees and other land features).

[0013] In some embodiments, the surface model is subjected to stereoscopic cutting and layering based on the ground elevation point classification data, specifically:

[0014] Take the ground point as the starting elevation, and divide all points above the ground point into sections at intervals of N meters (N is a non-zero natural number);

[0015] All points within N meters of each layer are divided into the same layer and named according to the middle value of each layer's elevation, forming spatial stereoscopic slice layered point cloud data for each height interval.

[0016] In the above technical solution, the elevation of the point cloud in each layer of data is automatically assigned to the elevation value of the starting point in the layer, forming a dense point cloud with the same elevation value, forming a more characteristic data set, which is convenient for the next step of automatically connecting into straight lines.

[0017] In some embodiments, dense point clouds of each layer are connected into vector lines based on an algorithm, and irregular lines are filtered. Specifically:

[0018] The parameter-finding method is used to automatically connect the point clouds in each layer of the data set into vector lines, and the irregular lines are filtered out;

[0019] The lines of the same building are automatically divided into an object layer, and the object height value is automatically calculated by multiplying the number of slices by the original slice interval unit, and stored in the attributes of the corresponding vector line.

[0020] In the above technical solution, the point clouds in each layer of data set are automatically connected into vector lines, and irregular lines are filtered out, leaving basically the edge lines of buildings. The lines of the same building are automatically divided into an object layer, and the object height value can be automatically calculated by multiplying the number of slices by the original slice interval unit and stored in the attributes of the corresponding vector line.

[0021] In some embodiments, all vector lines are optimized to extract the vector edges of the building outline, specifically:

[0022] A first vector region is formed based on the vector lines, and a second vector region is generated after removing the vector lines whose intervals are smaller than a threshold value;

[0023] Based on the second vector region, based on the upper, lower, left and right limit boundaries of the second vector region, square grid rasterization is used to obtain a first rasterized region, and the rasterized region coordinates are constructed with the center of the first rasterized region as the origin;

[0024] Based on dividing the first rasterized area into a plurality of sub-areas;

[0025] Based on the grid coordinates and boundaries of the vector lines in several sub-regions, the midpoint coordinate sequences of the sub-regions in the X direction or Y direction are calculated respectively; based on the midpoint coordinate sequences of the sub-regions, the midpoint coordinate sequences are fitted by the least square method to obtain the midlines of the sub-regions;

[0026] The center lines of the sub-areas are connected to form the boundary of the building, and the second rasterized area is generated by reorganizing the boundaries. The second rasterized area is converted to obtain a third vector area with vector edges.

[0027] In the above technical solution, the contour lines generated by the inner and outer surfaces are optimized to obtain the edge lines that are closest to the actual surface model of each height segment to form each contour line set of different height intervals of the ground object.

[0028] In some embodiments, optimizing all vector lines to extract the vector edges of the building outlines also includes:

[0029] Based on the grid coordinates and boundaries of the vector lines in the several sub-regions, the midpoint coordinate sequences of the sub-regions in the X direction or the Y direction are calculated respectively; based on the midpoint coordinate sequences of the sub-regions, the midlines of the sub-regions are fitted by the least square method;

[0030] For a number of adjacent, continuous and same-direction sub-regions, calculate the midline slopes of the sub-regions;

[0031] The average slope is calculated using the average weight value, and the formula is as follows:

[0032] Average slope = (w1*x1+w2*x2+...+w n *x n ) / (w1+w2+...+w n )

[0033] Where w1, w2, ..., w n is the weight value, and the segmentation interval of the sub-region in the midline direction is taken as the weight value of each sub-region; x1, x2, …, x n are the midline slopes of several sub-areas;

[0034] The average slope is used to replace the midline slopes of several sub-areas to generate a new midline;

[0035] The new center lines of each sub-area are connected to form the boundary of the building, and the second rasterized area is generated by reorganizing the boundaries. The second rasterized area is converted to obtain a third vector area with vector edges.

[0036] In the above technical solution, when individual vector lines do not fit the actual building structure, vector lines can be automatically merged to generate new vector lines.

[0037] In some embodiments, based on the result of stereoscopic cutting and stratification, the elevation of the point cloud in each layer of data is automatically assigned to the elevation value of the starting point in the layer, forming a dense point cloud with the same elevation value, and then it can also be:

[0038] The neural network algorithm is used to identify the objects outside the wall of the surface model after three-dimensional cutting and layering, and the vector edges of the building contours are extracted based on the identified objects outside the wall.

[0039] In the above technical solution, AI technology is used to automatically select a vector edge line that is closest to the building's outer contour. AI can select the edge line that is closest to the wall by automatically identifying important objects outside the wall, such as solid walls, glass windows, and air conditioner outdoor units.

[0040] According to another aspect of the present invention, a device for automatically extracting contour lines of objects in a slice-by-layer manner is provided, based on the above method; comprising:

[0041] The acquisition module is used to obtain the surface model and export the ground elevation point classification data;

[0042] The segmentation module is used to perform stereoscopic cutting and layering of the surface model based on ground elevation point classification data;

[0043] The generation module is used to automatically assign the elevation of the point cloud in each layer of data to the elevation value of the starting point in the layer based on the result of the three-dimensional cutting and stratification, so as to form a dense point cloud with the same elevation value;

[0044] The connection module is used to connect the dense point clouds of each layer into vector lines based on the algorithm and filter irregular lines;

[0045] Optimization module: optimize all vector lines and extract the vector edges of building outlines.

[0046] In the above technical scheme, in order to better use the above method, the present application proposes a device for automatically extracting the contour lines of objects in a sliced ​​and layered manner. Each module corresponds to each step of the above method. The specific principles have been described above and will not be repeated here.

[0047] According to another aspect of the present invention, there is provided a device for automatically extracting contour lines of objects in a slice-by-layer manner, comprising:

[0048] at least one processor and a memory communicatively coupled to the at least one processor;

[0049] The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can perform the above method.

[0050] In the above technical solution, in order to better run and process the method, the above method is stored in a memory, and the stored method is executed by a processor. It should be noted that the principle and effect of each step have been described above and will not be described in detail here.

[0051] According to another aspect of the present invention, a computer-readable storage medium is provided, storing a computer program, wherein the computer program implements the above method when executed by a processor.

[0052] In the above technical solution, in order to better run and use the method, the above method is stored in a computer-readable storage medium, and the above method is implemented using a processor. It should be noted that the principle and effect of each step have been described above and will not be described in detail here. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0054] Figure 1 It is a flow chart of an embodiment of a method for automatically extracting contour lines of land features by slicing and layering according to the present invention;

[0055] Figure 2 It is a schematic diagram of a surface model of an embodiment of a method for automatically extracting contour lines of ground objects in a slice-by-layer manner according to the present invention;

[0056] Figure 3 It is a schematic diagram of stereoscopic sliced ​​layered point cloud data of an embodiment of a method for automatically extracting contour lines of ground objects in sliced ​​layered manner according to the present invention;

[0057] Figure 4 It is a schematic diagram of dense point cloud data with the same elevation value according to an embodiment of a method for automatically extracting contour lines of objects in a sliced ​​and layered manner according to the present invention;

[0058] Figure 5 It is a schematic diagram of automatically connecting point clouds into vector lines according to an embodiment of a method for automatically extracting contour lines of objects in a sliced ​​and layered manner according to the present invention;

[0059] Figure 6 It is a schematic diagram of the vector edge line of the building outline of an embodiment of a method for automatically extracting the contour line of a ground object in a slice-by-layer manner according to the present invention;

[0060] Figure 7 It is a structural schematic diagram of an embodiment of a device for automatically extracting contour lines of objects in a sliced ​​and layered manner according to the present invention. DETAILED DESCRIPTION

[0061] The present invention will be further described in detail below in conjunction with the accompanying drawings and examples. It is particularly noted that the following examples are only used to illustrate the present invention, but are not intended to limit the scope of the present invention. Similarly, the following examples are only partial embodiments of the present invention rather than all embodiments, and all other embodiments obtained by those of ordinary skill in the art without creative work are within the scope of protection of the present invention.

[0062] The present invention provides a method for automatically extracting the contour lines of land objects by slicing and layering, which can greatly improve the surveying and mapping efficiency and mapping accuracy of buildings and structures.

[0063] Embodiment 1

[0064] See also Figure 1 , a method for automatically extracting the contour line of a ground object by slicing and layering, comprising:

[0065] S1. Obtain the surface model and derive ground elevation point classification data;

[0066] In this embodiment, the surface model is a three-dimensional surface model, or a digital surface model. Figure 2 , through unmanned aerial vehicles equipped with optical cameras or laser point cloud equipment to take oblique aerial photos of the ground, automatically generate surface three-dimensional model (mesh) data or digital surface model (DSM) data after aerial triangulation, and export dense point cloud (las) data. Point cloud data can also be obtained through airborne and satellite-borne array three-dimensional synthetic aperture radar (SAR) data. After automatic filtering of the point cloud data, most of the ground elevation point classification data can be obtained.

[0067] S2, performing three-dimensional cutting and layering of the surface model based on ground elevation point classification data;

[0068] In this embodiment, S2, the surface model is cut and layered based on the ground elevation point classification data, specifically:

[0069] S21, taking the ground point as the starting elevation, and dividing and stratifying all points above the ground point by dividing the interval into sections at every N (N is a non-zero natural number) meters of elevation;

[0070] S22. All points within N meters of each layer are divided into the same layer, and named according to the middle value of the elevation of each layer, so as to form spatial stereoscopic slice layered point cloud data of each height interval segment.

[0071] For example, see Figure 3 , the ground point is used as the starting elevation, and all points above the ground point are cut and layered at intervals of 1 meter (the parameter can be customized according to different survey areas). For example, when the ground starting point is 10.50 meters, 11.50 meters is taken as the cutting middle value, and all points from 11.25 meters to 11.75 meters (the extracted interval value can also be customized according to different survey areas) are divided into the same layer. The layer name is automatically named according to the cutting middle value to form spatial stereo slice layered point cloud data of each height interval.

[0072] S3, based on the result of stereoscopic cutting and stratification, the elevation of the point cloud in each layer of data is automatically assigned to the elevation value of the starting point in the layer, forming a dense point cloud with the same elevation value;

[0073] In this example, see Figure 4 , using the program to automatically process, the point cloud elevation in each layer of data is automatically assigned the elevation value of the starting point in the layer, forming a dense point cloud with the same elevation value, forming a more characteristic data set, which is convenient for the next step of automatically connecting into a straight line.

[0074] S4, connect the dense point clouds of each layer into vector lines based on the algorithm, and filter irregular lines;

[0075] In this embodiment, S4, based on an algorithm, connects the dense point clouds of each layer into vector lines, and filters irregular lines, specifically:

[0076] S41, using the parameter-finding method, automatically connect the point clouds in each layer of the data set into vector lines, and filter out irregular lines;

[0077] S42. Lines of the same building are automatically divided into an object layer, and the object height value is automatically calculated according to the number of slices multiplied by the original slice interval unit, and stored in the attribute of the corresponding vector line.

[0078] For example, see Figure 5 , set up an algorithm program to obtain the point cloud position (xy coordinates), normal direction, curvature and other parameters, automatically connect the point clouds in each layer of the data set into vector lines, and filter out irregular lines, leaving basically the edge lines of buildings. And automatically divide the lines of the same building into an object layer, and automatically calculate the object height value by multiplying the number of slices by the original slice interval unit, and store it in the attributes of the corresponding vector line.

[0079] S5. Optimize all vector lines and extract the vector edges of building outlines.

[0080] In this example, see Figure 6 , S5, optimize all vector lines and extract the vector edge lines of the building outline, specifically:

[0081] S511, forming a first vector region based on the vector lines, removing the vector lines whose vector line intervals are less than a threshold, and then generating a second vector region;

[0082] S512, based on the second vector region and the upper, lower, left and right limit boundaries of the second vector region, a first rasterized area is obtained by using a square grid rasterization, and the coordinates of the rasterized area are constructed with the center of the first rasterized area as the origin;

[0083] S513, dividing the first rasterized area into a plurality of sub-areas;

[0084] S514, based on the grid coordinates and boundaries of the vector lines in the plurality of sub-regions, respectively calculate the midpoint coordinate sequences of the sub-regions in the X direction or the Y direction; based on the midpoint coordinate sequences of the sub-regions, use the least squares method to fit the midpoint coordinate sequences to obtain the midlines of the sub-regions;

[0085] S515, connecting the center lines of the sub-areas to form the boundary of the building, generating a second rasterized area by reorganizing the boundaries, and obtaining a third vector area with vector edges from the second rasterized area.

[0086] In this embodiment, S5, optimizing all vector lines to extract vector edges of building outlines, also includes:

[0087] S521, based on the grid coordinates and boundaries of the vector lines in the plurality of sub-regions, respectively calculate the midpoint coordinate sequences of the sub-regions in the X direction or the Y direction; based on the midpoint coordinate sequences of the sub-regions, use the least squares method to fit the midlines of the sub-regions;

[0088] S522, for a plurality of adjacent, continuous and same-direction sub-regions, calculating the midline slopes of the plurality of sub-regions;

[0089] S523. Calculate the average slope using the average weight value, using the following formula:

[0090] Average slope = (w1*x1+w2*x2+...+w n *x n ) / (w1+w2+...+w n )

[0091] Where w1, w2, ..., w n is the weight value, and the segmentation interval of the sub-region in the midline direction is taken as the weight value of each sub-region; x1, x2, …, xn are the midline slopes of several sub-areas;

[0092] S524, using the average slope to replace the midline slopes of the sub-regions to generate a new midline;

[0093] S525, connecting the new center lines of the sub-areas to form the boundary of the building, generating a second rasterized area by reorganizing the boundaries, and obtaining a third vector area with vector edges from the second rasterized area.

[0094] In this embodiment, S5, based on the result of the three-dimensional cutting and stratification, the elevation of the point cloud in each layer of data is automatically assigned to the elevation value of the starting point in the layer, forming a dense point cloud with the same elevation value, and then it can also be:

[0095] The neural network algorithm is used to identify the objects outside the wall of the surface model after three-dimensional cutting and layering, and the vector edge lines of the building outline are extracted based on the identified objects outside the wall. AI technology is used to automatically select a vector edge line that is closest to the building's outer contour. AI can select the edge line closest to the wall as long as it automatically identifies important objects outside the wall such as solid walls, glass windows, and air conditioner outdoor units.

[0096] Embodiment 2

[0097] See also Figure 7 , a device for automatically extracting the contour line of a land object by slicing and layering, based on the above method; comprising:

[0098] The acquisition module is used to obtain the surface model and export the ground elevation point classification data;

[0099] The segmentation module is used to perform stereoscopic cutting and layering of the surface model based on ground elevation point classification data;

[0100] The generation module is used to automatically assign the elevation of the point cloud in each layer of data to the elevation value of the starting point in the layer based on the result of the three-dimensional cutting and stratification, so as to form a dense point cloud with the same elevation value;

[0101] The connection module is used to connect the dense point clouds of each layer into vector lines based on the algorithm and filter irregular lines;

[0102] Optimization module: optimize all vector lines and extract the vector edges of building outlines.

[0103] In the above technical scheme, in order to better use the method described in one of the embodiments, the present application proposes a device for automatically extracting the contour lines of objects in a sliced ​​and layered manner. Each module corresponds to each step of the above method. The specific principles have been described above and will not be repeated here.

[0104] Embodiment 3

[0105] A device for automatically extracting contour lines of objects by slicing and layering, comprising:

[0106] at least one processor and a memory communicatively coupled to the at least one processor;

[0107] The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the method described in one of the embodiments.

[0108] In the above technical solution, in order to better run and process the method described in one of the embodiments, the above method is stored in a memory, and a processor is used to execute the stored method. It should be noted that the principle and effect of each step have been described above and will not be described in detail here.

[0109] Embodiment 4

[0110] A computer-readable storage medium stores a computer program, which implements the method described in one of the embodiments when the computer program is executed by a processor.

[0111] In the above technical solution, in order to better run and use the method described in one of the embodiments, the above method is stored in a computer-readable storage medium, and the above method is implemented using a processor. It should be noted that the principle and effect of each step have been described above and will not be described in detail here.

[0112] The above descriptions are only some embodiments of the present invention, and are not intended to limit the protection scope of the present invention. Any equivalent device or equivalent process transformation made using the contents of the present invention specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A method for automatically extracting the contour lines of objects by slicing and layering. It is characterized in that include: Obtain the surface model and export ground elevation point classification data; The surface model is cut and layered in three dimensions based on the ground elevation point classification data; Based on the results of stereoscopic cutting and stratification, the elevation of the point cloud in each layer of data is automatically assigned to the elevation value of the starting point in the layer, forming a dense point cloud with the same elevation value; Based on the algorithm, the dense point clouds of each layer are connected into vector lines, and irregular lines are filtered; All vector lines are optimized to extract the vector edges of building outlines.

2. A method for automatically extracting contour lines of objects by slicing and layering as claimed in claim 1, It is characterized in that The surface model is a three-dimensional surface model, or a digital surface model.

3. A method for automatically extracting contour lines of objects by slicing and layering as claimed in claim 1, It is characterized in that The surface model is cut and layered based on the ground elevation point classification data. Specifically: Take the ground point as the starting elevation, and divide all points above the ground point into sections at a rate of N (N is a non-zero natural number) meters. All points within N meters of each layer are divided into the same layer and named according to the middle value of each layer elevation to form spatial stereo slice layered point cloud data for each height interval.

4. A method for automatically extracting contour lines of objects by slicing and layering as claimed in claim 1, It is characterized in that Based on the algorithm, the dense point clouds of each layer are connected into vector lines, and irregular lines are filtered. Specifically: The parameter-finding method is used to automatically connect the point clouds in each layer of the data set into vector lines, and the irregular lines are filtered out; The lines of the same building are automatically divided into an object layer, and the object height value is automatically calculated by multiplying the number of slices by the original slice interval unit, and stored in the attributes of the corresponding vector line.

5. The method for automatically extracting the contour lines of objects by slicing and layering as claimed in claim 1, It is characterized in that Optimize all vector lines and extract the vector edges of building outlines. Specifically: A first vector region is formed based on the vector lines, and a second vector region is generated after removing the vector lines whose intervals are smaller than a threshold value; Based on the second vector region, based on the upper, lower, left and right limit boundaries of the second vector region, square grid rasterization is used to obtain a first rasterized region, and the rasterized region coordinates are constructed with the center of the first rasterized region as the origin; Based on dividing the first rasterized area into a plurality of sub-areas; Based on the grid coordinates and boundaries of the vector lines in several sub-regions, the midpoint coordinate sequences of the sub-regions in the X direction or Y direction are calculated respectively; based on the midpoint coordinate sequences of the sub-regions, the midpoint coordinate sequences are fitted by the least square method to obtain the midlines of the sub-regions; The center lines of the sub-areas are connected to form the boundary of the building, and the second rasterized area is generated by reorganizing the boundaries. The second rasterized area is converted to obtain a third vector area with vector edges.

6. A method for automatically extracting contour lines of objects by slicing and layering as claimed in claim 5, It is characterized in that Optimize all vector lines and extract the vector edges of building outlines, including: Based on the grid coordinates and boundaries of the vector lines in the several sub-regions, the midpoint coordinate sequences of the sub-regions in the X direction or the Y direction are calculated respectively; based on the midpoint coordinate sequences of the sub-regions, the midlines of the sub-regions are fitted by the least square method; For a number of adjacent, continuous and same-direction sub-regions, calculate the midline slopes of the sub-regions; The average slope is calculated using the average weight value, and the formula is as follows: Average slope = (w 1 *x 1 +w 2 *x 2 +...+w n *x n ) / (w 1 +w 2 +...+w n ) In the formula, w 1 、w 2 ,…,w n is the weight value, and the segmentation distance of the sub-region in the midline direction is taken as the weight value of each sub-region; 1 ,,x 2 , …, x n are the midline slopes of several sub-areas; The average slope is used to replace the midline slopes of several sub-areas to generate a new midline; The new center lines of each sub-area are connected to form the boundary of the building, and the second rasterized area is generated by reorganizing the boundaries. The second rasterized area is converted to obtain a third vector area with vector edges.

7. The method for automatically extracting the contour lines of objects by slicing and layering as claimed in claim 1, It is characterized in that Based on the results of stereoscopic cutting and stratification, the point cloud elevation in each layer of data is automatically assigned to the starting point elevation value in the layer, forming a dense point cloud with the same elevation value, and then it can also be: The neural network algorithm is used to identify the objects outside the wall of the surface model after three-dimensional cutting and layering, and the vector edges of the building contours are extracted based on the identified objects outside the wall.

8. A device for automatically extracting the contour lines of objects by slicing and layering. It is characterized in that A method according to any one of claims 1 to 7; comprising: The acquisition module is used to obtain the surface model and export the ground elevation point classification data; The segmentation module is used to perform stereoscopic cutting and layering of the surface model based on ground elevation point classification data; The generation module is used to automatically assign the elevation of the point cloud in each layer of data to the elevation value of the starting point in the layer based on the result of the three-dimensional cutting and stratification, so as to form a dense point cloud with the same elevation value; The connection module is used to connect the dense point clouds of each layer into vector lines based on the algorithm and filter irregular lines; Optimization module: optimize all vector lines and extract the vector edges of building outlines.

9. A device for automatically extracting the contour lines of objects by slicing and layering. It is characterized in that include: at least one processor and a memory communicatively coupled to the at least one processor; The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the method as claimed in any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, It is characterized in that When the computer program is executed by a processor, the method according to any one of claims 1 to 7 is implemented.

Citation Information

Patent Citations

  • Method and system for remote-sensing image and laser point cloud registration based on road features

    CN104123730A

  • Automatic segmentation method for point cloud of facade of large scene city building

    CN105844629A

  • Multi-floor indoor structured three-dimensional modeling method and system based on laser point cloud

    CN110189412A

  • Grid-based contour line segment feature extraction algorithm

    CN110335282A

  • Topographic three-dimensional model and topographic map construction method and system based on laser point cloud, and storage medium

    CN113034689A