CN115061150A - Building extraction method based on laser radar point cloud data pseudo-waveform feature processing - Google Patents
Building extraction method based on laser radar point cloud data pseudo-waveform feature processing Download PDFInfo
- Publication number
- CN115061150A CN115061150A CN202210388077.5A CN202210388077A CN115061150A CN 115061150 A CN115061150 A CN 115061150A CN 202210388077 A CN202210388077 A CN 202210388077A CN 115061150 A CN115061150 A CN 115061150A
- Authority
- CN
- China
- Prior art keywords
- point cloud
- building
- pseudo
- ground
- waveform
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000605 extraction Methods 0.000 title claims abstract description 21
- 238000012545 processing Methods 0.000 title claims abstract description 19
- 238000001914 filtration Methods 0.000 claims abstract description 21
- 238000005516 engineering process Methods 0.000 claims abstract description 4
- 238000012216 screening Methods 0.000 claims abstract description 3
- 238000004458 analytical method Methods 0.000 claims description 7
- 230000007797 corrosion Effects 0.000 claims description 7
- 238000005260 corrosion Methods 0.000 claims description 7
- 230000000877 morphologic effect Effects 0.000 claims description 5
- 230000003628 erosive effect Effects 0.000 claims description 3
- 238000000034 method Methods 0.000 abstract description 16
- 230000000694 effects Effects 0.000 abstract description 2
- 230000011218 segmentation Effects 0.000 description 3
- 238000011160 research Methods 0.000 description 2
- ATJFFYVFTNAWJD-UHFFFAOYSA-N Tin Chemical compound [Sn] ATJFFYVFTNAWJD-UHFFFAOYSA-N 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000007621 cluster analysis Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 230000002708 enhancing effect Effects 0.000 description 1
- 238000010191 image analysis Methods 0.000 description 1
- 238000003709 image segmentation Methods 0.000 description 1
- 238000002372 labelling Methods 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S17/00—Systems using the reflection or reradiation of electromagnetic waves other than radio waves, e.g. lidar systems
- G01S17/88—Lidar systems specially adapted for specific applications
- G01S17/89—Lidar systems specially adapted for specific applications for mapping or imaging
Landscapes
- Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Electromagnetism (AREA)
- General Physics & Mathematics (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Image Processing (AREA)
Abstract
The invention relates to a building extraction method based on laser radar point cloud data pseudo-waveform feature processing, and belongs to the technical field of remote sensing. Dividing collected laser radar point cloud data into ground points and non-ground points; then utilizing neighborhood interpolation technology to generate resolution d facing to ground point xy A digital terrain model; extracting ground height from the ground points, and subtracting the ground height from all the point cloud heights to obtain the ground height; setting a spatial vertical resolution d z According to spatial resolution d xy And d z Defining a voxel, and distributing the voxel into a corresponding voxel according to the x, y and z coordinates of each point to generate a pseudo-waveform feature; screening out building radar point cloud according to the pseudo-waveform characteristics, so as to obtain a building candidate area; and carrying out filtering processing to obtain the building. The method solves the problems of incomplete extraction (effect) and the like caused by the fact that buildings are partially shielded by ground objects such as trees and the like in the 2D image, and improves the integrity of the buildings extracted by the airborne laser radar.
Description
Technical Field
The invention relates to a building extraction method based on laser radar point cloud data pseudo-waveform feature processing, and belongs to the technical field of remote sensing.
Background
With increasing focus on earth observation, LiDAR technology has become the focus of research in recent years. Because an airborne laser radar system provides a powerful solution for rapid three-dimensional mapping of ground surfaces, the current main research is to extract information from laser radar point clouds. Where extracting the ground to generate a digital terrain model is a critical step in this process.
At present, methods for extracting buildings by airborne laser radars are mainly divided into extraction methods based on rasterization and extraction methods based on point cloud data space structures. The method for rasterizing the point cloud data is characterized in that on the basis of point cloud segmentation, a building point cloud set is rasterized into a binary image, cavities in the image are filled by utilizing the expansion corrosion principle of pixels, and contour line extraction is realized by utilizing image analysis means such as an image segmentation algorithm, a boundary line extraction algorithm and the like. The method processes the radar point cloud into the gray level image, the processing speed block and the method are simple and clear, however, a new error is introduced in the rasterization processing process, and an accurate building vector boundary cannot be obtained.
Another extraction method based on the point cloud data space tends to construct a space Tin data structure more, and the point cloud is subjected to cluster analysis and detection by utilizing the spatial information among the triangulation networks. Extracting boundary points of a triangular net external convex hull by using a long-edge limiting method on the basis of the clustered point cloud triangular net, or judging and extracting boundary points outside the building by adopting a classical Alpha Shape algorithm to carry out set boundary judgment from a disordered point set. The methods establish the triangulation network by analyzing the point cloud space structure of the building, obtain a more accurate boundary contour of the building, but still have the following defects: firstly, the structure of the triangulation network is complex, and the quantity of the triangulation network is about 2 times that of the point cloud; when the point cloud data volume is large, the triangulation network construction speed is low, and the occupied space is large. Secondly, only the triangulation boundary longer than the threshold length can be extracted by using a long-edge limiting method; loss of detail can occur at building corners or concave edges.
Therefore, under the condition that the point cloud quantity of the building radar point cloud is very huge, particularly for a building which is shielded in a 2D image, the laser radar point cloud is adopted to penetrate trees which shield the building, so that the point cloud of the building under the tree is obtained, and more complete buildings are extracted.
Disclosure of Invention
The invention aims to provide a building extraction method based on laser radar point cloud data pseudo-waveform feature processing, which is used for solving the problems that a building in a 2D image is partially shielded and the like under the condition that ground objects such as trees and the like are shielded, and enhancing the integrity of the building in the building point cloud extraction process.
The technical scheme of the invention is as follows: a building extraction method based on laser radar point cloud data pseudo-waveform feature processing divides collected laser radar point cloud data into ground points and non-ground points, and then generates resolution d facing the ground points by using a neighborhood interpolation technology xy A digital terrain model; extracting ground height from the ground points, and subtracting the ground height from all the point cloud heights to obtain the ground height; setting a spatial vertical resolution d z According to spatial resolution d xy And d z Defining a voxel, and distributing the voxel into a corresponding voxel according to the x, y and z coordinates of each point to generate a pseudo-waveform feature; and screening the radar point cloud of the building according to the pseudo-waveform characteristics so as to obtain a candidate area of the building, and finally, carrying out filtering processing to obtain the building.
The method comprises the following specific steps:
step 1: inputting laser radar point cloud data, wherein the voxelized resolution ratio is d xy And d z ；
Step 2: dividing the point cloud data into ground points and non-ground points;
step 3: distributing the laser radar point cloud into corresponding pixel areas according to the x and y coordinates of the laser radar point cloud;
step 4: extracting ground height from the ground points;
step 5: subtracting the actual height from the heights of all the point clouds to obtain the ground height;
step 6: according to spatial resolution d xy And d z Defining x, y, z (height to ground) coordinates of a voxel and a laser radar point cloud, and distributing each point in the point cloud into a corresponding voxel;
step 7: summing the point cloud intensity values in each voxel to generate pseudo waveform data;
step 8: normalizing the pseudo waveform according to the number of point clouds in each voxel to finally obtain pseudo waveform characteristics;
step 9: extracting a building candidate region by utilizing the pseudo-waveform characteristics;
step 10: and filtering the extracted building point cloud to obtain the extracted building.
The Step9 comprises the following specific steps: dividing according to the intensity value of the pseudo-waveform characteristics, analyzing the point cloud intensity value distribution of the building, and obtaining through analysis: the intensity peak of a building area occurs only at a height of 4 meters above the ground and the intensity value is between 300 and 400, at 4 meters above, only the intensity of the cloud of the building point has a single peak. Therefore, according to the analysis result, radar point clouds meeting the following three points are extracted, and then the building area is obtained:
(1) radar point cloud with the height above 4 meters above the ground;
(2) the cloud intensity value of the pseudo-waveform point has a single peak value, and no peak value exists below 4 meters of the pseudo-waveform point;
(3) a radar point cloud having a pseudo-waveform point cloud intensity peak between 300 and 400.
The Step10 comprises the following specific steps:
step10.1: performing morphological corrosion filtering on the extracted building area;
step10.2: and performing expansion filtering on the basis of corrosion filtering to obtain the extracted building.
The method utilizes the point cloud data of the laser radar to extract the pseudo-waveform data, utilizes the extracted pseudo-waveform characteristics to analyze and extract the point cloud of the building meeting the conditions, and further obtains the area of the building. And then, morphological corrosion filtering processing is carried out on the extracted building area, so that noise points can be effectively removed. And processing through expansion filtering to finally obtain a complete building area.
The invention has the beneficial effects that: the method is suitable for the non-tag radar point cloud data, semantic segmentation is carried out on the building radar point cloud in a non-supervision mode, the radar point cloud can be used for effectively extracting the radar point cloud of the shielded building area, and the building integrity in the building point cloud extraction process is enhanced.
Drawings
FIG. 1 is a general flow chart of the present invention;
FIG. 2 is a cloud height profile of the present invention;
FIG. 3 is a pseudo waveform signature statistical chart of the present invention;
FIG. 4 is a graph of experimental results of a filtering operation of the present invention;
FIG. 5 is a final experimental result of the present invention.
Detailed Description
The invention is further described with reference to the following drawings and detailed description.
As shown in fig. 1, a building extraction method based on pseudo-waveform feature processing of laser radar point cloud data specifically includes the following steps:
step 1: inputting some laser radar point cloud data as basic data, wherein the voxelized resolution ratios are d xy And d z ；
Step 2: dividing the laser radar point cloud data into ground points and non-ground points;
step 3: distributing the laser radar point cloud into corresponding pixel areas according to the x and y coordinates of the laser radar point cloud;
step 4: extracting ground height from the ground points;
step 5: subtracting the actual height from the heights of all the point clouds to obtain the ground height; the point cloud height statistics are shown in fig. 2.
Step 6: according to spatial resolution d xy And d z Defining x, y, z (height to ground) coordinates of a voxel and a laser radar point cloud, and distributing each point in the point cloud into a corresponding voxel;
step 7: summing the point cloud intensity values in each voxel to generate pseudo waveform data;
step 8: and normalizing the pseudo waveform according to the number of the point clouds in each voxel to finally obtain the pseudo waveform characteristics, as shown in fig. 3.
Step 9: extracting a building candidate region by utilizing the pseudo-waveform characteristics;
step 10: and filtering the extracted building point cloud to obtain the extracted building.
The Step9 comprises the following specific steps: dividing according to the intensity value of the pseudo-waveform characteristics, analyzing the point cloud intensity value distribution of the building, and obtaining through analysis: the intensity peak of a building area occurs only at a height of 4 meters above the ground and the intensity value is between 300 and 400, at 4 meters above, only the intensity of the cloud of the building point has a single peak. Therefore, according to the analysis result, radar point clouds meeting the following three points are extracted, and then the building area is obtained:
(1) radar point cloud with the height above 4 meters above the ground;
(2) the cloud intensity value of the pseudo-waveform points has a single peak value, and no peak value exists below 4 meters of the pseudo-waveform;
(3) a radar point cloud having a pseudo-waveform point cloud intensity peak between 300 and 400.
The Step10 comprises the following specific steps:
step10.1: performing morphological erosion filtering on the extracted building region, as shown in fig. 4, wherein (a) is a building region obtained by performing threshold segmentation (removing pixels corresponding to cloud data having a distance of 3 meters (inclusive) or less from the ground) by using data having a pseudo-waveform feature and having a distance of 3 meters (inclusive) or less from the ground; (b) performing morphological corrosion filtering on the building area extracted from the upper surface; (c) expansion filtering is performed on the basis of erosion filtering.
Step10.2: and performing expansion filtering on the basis of corrosion filtering to obtain the extracted building. As shown in fig. 5, (a) is an intensity map of the extracted building superimposed on the corresponding point cloud data; (b) the extracted building is superimposed on the corresponding RGB image.
Compared with the traditional method, the method does not depend on the radar point cloud data which is already labeled. The unsupervised mode is adopted for extraction and processing, so that the labor and financial resources consumed by labeling and other processing of the point cloud data can be saved. Meanwhile, aiming at the problems that the extraction is incomplete (effect) and the like caused by the fact that the building is partially shielded by ground objects such as trees and the like in the current 2D image, the invention improves the integrity of the building extracted by the airborne laser radar.
While the present invention has been described in detail with reference to the embodiments shown in the drawings, the present invention is not limited to the embodiments, and various changes can be made without departing from the spirit and scope of the present invention.
Claims (4)
1. A building extraction method based on laser radar point cloud data pseudo-waveform feature processing is characterized by comprising the following steps: dividing the collected laser radar point cloud data into ground points and non-ground points, and then generating the resolution d by using the neighborhood interpolation technology for the ground points xy A digital terrain model; extracting ground height from the ground points, and subtracting the ground height from all the point cloud heights to obtain the ground height; setting a spatial vertical resolution d z According to spatial resolution d xy And d z Defining a voxel, and distributing the voxel into a corresponding voxel according to the x, y and z coordinates of each point to generate a pseudo-waveform feature; and screening the radar point cloud of the building according to the pseudo-waveform characteristics to obtain a candidate area of the building, and finally filtering to obtain the building.
2. The building extraction method based on the pseudo-waveform feature processing of the laser radar point cloud data according to claim 1, which is characterized by comprising the following specific steps of:
step 1: inputting laser radar point cloud data, wherein the voxelized resolution ratio is d xy And d z ；
Step 2: dividing the point cloud data into ground points and non-ground points;
step 3: distributing the laser radar point cloud into a corresponding pixel area according to the x and y coordinates of the laser radar point cloud;
step 4: extracting ground height from the ground points;
step 5: subtracting the ground height from the heights of all the point clouds to obtain the ground height z of the laser radar point cloud;
step 6: according to spatial resolution d xy And d z Defining x, y and z coordinates of a voxel and a laser radar point cloud, and distributing each point in the point cloud into a corresponding voxel;
step 7: summing the point cloud intensity values in each voxel to generate pseudo waveform data;
step 8: normalizing the pseudo waveform data according to the number of point clouds in each voxel to finally obtain pseudo waveform characteristics;
step 9: extracting a building candidate region by utilizing the pseudo-waveform characteristics;
step 10: and filtering the extracted candidate building area to obtain the extracted building.
3. The building extraction method based on the pseudo-waveform feature processing of the lidar point cloud data according to claim 1, wherein Step9 comprises the following steps: dividing the obtained laser radar building point cloud according to the intensity value of the pseudo-waveform characteristic, analyzing the distribution of the intensity value of the building point cloud according to the divided intensity value, and obtaining the following result through analysis: the intensity peak value of the building area only appears at the position with the height of more than 4 meters to the ground, the intensity value is between 300 and 400, and more than 4 meters, only the intensity of the cloud of the building points appears a single peak value, so according to the analysis content, according to the analysis result, the radar point cloud satisfying the following three points is extracted, and the building area is further obtained:
(1) radar point cloud with the height above 4 meters above the ground;
(2) the cloud intensity value of the pseudo-waveform point has a single peak value, and no peak value exists below 4 meters of the pseudo-waveform point;
(3) a radar point cloud having a pseudo-waveform point cloud intensity peak between 300 and 400.
4. The building extraction method based on the pseudo-waveform feature processing of the lidar point cloud data according to claim 1, wherein Step10 comprises the following steps:
step10.1: performing morphological erosion filtering on the extracted building area;
step10.2: and performing expansion filtering on the basis of corrosion filtering to obtain the extracted building.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210388077.5A CN115061150A (en) | 2022-04-14 | 2022-04-14 | Building extraction method based on laser radar point cloud data pseudo-waveform feature processing |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210388077.5A CN115061150A (en) | 2022-04-14 | 2022-04-14 | Building extraction method based on laser radar point cloud data pseudo-waveform feature processing |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115061150A true CN115061150A (en) | 2022-09-16 |
Family
ID=83197462
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210388077.5A Pending CN115061150A (en) | 2022-04-14 | 2022-04-14 | Building extraction method based on laser radar point cloud data pseudo-waveform feature processing |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115061150A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115375902A (en) * | 2022-10-26 | 2022-11-22 | 昆明理工大学 | Multi-spectral laser radar point cloud data-based over-point segmentation method |
| CN117953384A (en) * | 2024-03-27 | 2024-04-30 | 昆明理工大学 | Cross-scene multispectral laser radar point cloud building extraction and vectorization method |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113920420A (en) * | 2020-07-07 | 2022-01-11 | 香港理工大学深圳研究院 | Building extraction method and device, terminal equipment and readable storage medium |
-
2022
- 2022-04-14 CN CN202210388077.5A patent/CN115061150A/en active Pending
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113920420A (en) * | 2020-07-07 | 2022-01-11 | 香港理工大学深圳研究院 | Building extraction method and device, terminal equipment and readable storage medium |
Non-Patent Citations (1)
| Title |
|---|
| JINHA JUNG等: ""A framework for land cover classification using discrete return LiDAR data:adopting pseudo-waveform and hierarchical segmentation"", 《IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING》, vol. 7, no. 2, 28 February 2014 (2014-02-28), pages 491 - 502, XP011538989, DOI: 10.1109/JSTARS.2013.2292032 * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115375902A (en) * | 2022-10-26 | 2022-11-22 | 昆明理工大学 | Multi-spectral laser radar point cloud data-based over-point segmentation method |
| CN117953384A (en) * | 2024-03-27 | 2024-04-30 | 昆明理工大学 | Cross-scene multispectral laser radar point cloud building extraction and vectorization method |
| CN117953384B (en) * | 2024-03-27 | 2024-06-07 | 昆明理工大学 | Cross-scene multispectral laser radar point cloud building extraction and vectorization method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104049245B (en) | Urban building change detection method based on LiDAR point cloud spatial difference analysis | |
| CN115061150A (en) | Building extraction method based on laser radar point cloud data pseudo-waveform feature processing | |
| US9330435B2 (en) | Bare earth finding and feature extraction for 3D point clouds | |
| Li et al. | Adaptive building edge detection by combining LiDAR data and aerial images | |
| CN108399424B (en) | Point cloud classification method, intelligent terminal and storage medium | |
| CN110008207B (en) | Airborne L iDAR point cloud data vulnerability rapid detection method based on density histogram | |
| CN114612488A (en) | Building-integrated information extraction method, computer device, and storage medium | |
| CN111681212B (en) | Three-dimensional target detection method based on laser radar point cloud data | |
| CN113570621B (en) | Tree information extraction method and device based on high-precision point cloud and image | |
| CN114926602B (en) | Building singleization method and system based on three-dimensional point cloud | |
| CN112669333A (en) | Single tree information extraction method | |
| CN116704333B (en) | Single tree detection method based on laser point cloud data | |
| CN113066004A (en) | Point cloud data processing method and device | |
| Rashidi et al. | Ground filtering LiDAR data based on multi-scale analysis of height difference threshold | |
| Hao et al. | A graph-based progressive morphological filtering (GPMF) method for generating canopy height models using ALS data | |
| Li et al. | Feature extraction and modeling of urban building from vehicle-borne laser scanning data | |
| CN111091071A (en) | Underground target detection method and system based on ground penetrating radar hyperbolic wave fitting | |
| CN107564024B (en) | SAR image aggregation region extraction method based on single-side aggregation line segment | |
| CN116958808A (en) | Forest parameter estimation method based on real-time target detection network | |
| Omidalizarandi et al. | Segmentation and classification of point clouds from dense aerial image matching | |
| CN114743008B (en) | Single plant vegetation point cloud data segmentation method and device and computer equipment | |
| CN116309284A (en) | Slope top/bottom line extraction system and method | |
| CN113838199B (en) | Three-dimensional terrain generation method | |
| CN115880322A (en) | Point cloud road landmark line extraction method based on gradient complexity | |
| CN111932574B (en) | Building vertical point cloud extraction system and method based on multi-level semantic features |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |