CN107220987A - A kind of building roof Fast Edge Detection method based on principal component analysis - Google Patents
A kind of building roof Fast Edge Detection method based on principal component analysis Download PDFInfo
- Publication number
- CN107220987A CN107220987A CN201610161761.4A CN201610161761A CN107220987A CN 107220987 A CN107220987 A CN 107220987A CN 201610161761 A CN201610161761 A CN 201610161761A CN 107220987 A CN107220987 A CN 107220987A
- Authority
- CN
- China
- Prior art keywords
- roof
- point
- principal component
- building roof
- building
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10028—Range image; Depth image; 3D point clouds
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30181—Earth observation
- G06T2207/30184—Infrastructure
Abstract
Building roof Fast Edge Detection method based on principal component analysis, using two major axes orientations that the distribution of building roof point cloud is calculated based on principal component analytical method, by oriented projection maximum range value constraint detection roof edge point, it is characterized in that high automaticity and low algorithm complex.It can be widely applied to airborne LiDAR and low latitude unmanned plane tilt the building roof detection of image dense Stereo Matching point cloud.
Description
Technical field
The present invention be on one kind from dense three-dimensional discrete roof point cloud quick detection roof contour method, especially with regard to loading LiDAR and unmanned plane tilt Image Matching reconstruction point cloud roof contour detect apply.
Background technology
The roof contour detection method being commonly used in practical application has following several：The scanning line detecting method and cluster cluster convex closure detection algorithm axially traveled through completely according to horizontal and two coordinates of row.But often there are some important defects in these traditional methods：For example, the precision of scanning line detecting method is influenceed more serious by building and course angle, and limited precision；Convex closure detection algorithm considers that in the case of the constraint of the space search radius of neighbourhood building roof profile point can be accurately detected out, but amount of calculation is bigger than normal.
Building roof detection method based on principal component analysis solves two major axes orientations of single building roof using the method for projecting and building covariance matrix in the horizontal plane is put in cluster.According to projector distance of the point on major axes orientation, and then identify building edge contour point.
The content of the invention
The problem of present invention is in solution known technology provides a kind of extraction algorithm based on two major axes orientations of building roof.
It is of the invention to be to solve the technical scheme used the problem of in known technology：(1), the building roof Fast Edge Detection based on principal component analysis：Detect that the roof edge contour method of dense three-dimensional discrete point cloud data structure representation relates generally to three committed steps by principal component analytical method (Principal Component Analysis, abbreviation PCA)：First, for by the three-dimensional intensive cloud data of building roof after filtering and the processing of elevation threshold values, covariance matrix is built after floor projection.Secondly, by covariance matrix feature decomposition, two Main ways of building roof, the i.e. length of building roof and wide two direction of principal axis where calculating a cloud.Finally, by roof point set projection PxyPlanar range { (xmin, xmax)|(ymin, ymax) two main direction of principal axis on conjugate roof do roof contour slice analysis, then can quickly select the side marginal point of building roof four.
The present invention has the advantages and positive effects of：The present invention is the combination of computational geometry and data statistics knowledge, forms one kind and quickly recognizes building roof outline technology using principal component analytical method.For each class cluster cluster processing after clustering processing, therefrom fast automatic detecting roof approximate contours, recognize building monomer subject area.The method that the present invention is used reduces staff's participation, and algorithm computation complexity is low, and building roof detection during magnanimity City scenarios building singulation can be quickly handled in real time, data processing time cost and cost of human resources greatly reduces.
Brief description of the drawings
Below by the application field and committed step of one group of brief description of the drawings this algorithm.
Fig. 1 is the digital orthoimage of a complicated earth surface environment
Fig. 2 is one corresponding with Fig. 1 three-dimensional dense Stereo Matching point cloud
Fig. 3 is to solve for two principal directions of building roof point cloud in cluster
Fig. 4 is along the slice analysis in major axis principal direction
Fig. 5 is along the slice analysis in short axle principal direction
Fig. 6 is the roof edge of detection
Embodiment
The present invention is described in detail below.The technology is analyzed building roof point set with principal component analytical method, using projector distance principle, sets up marginal point identification model, is arrived the distance between main shaft by calculating point, is obtained building edge contour point.
The reality of the present invention mainly includes the following aspects：
(1), two major axes orientation detections of the building roof based on principal component analysis
First, for by the three-dimensional intensive cloud data of building roof after filtering and the processing of elevation threshold values, building covariance matrix in horizontal plane xoy, making PxyProjection coordinate of the roof point in xoy planes is represented, be can obtain
Secondly, feature decomposition ApV=λ v, calculate covariance matrix ApTwo characteristic value (λ1, λ2) and its corresponding characteristic vector (v1, v2).The two column vectors represent two Main ways of building roof point cloud distribution where being exactly, the i.e. length of building roof and wide two direction of principal axis.
Finally, by roof point set projection PxyPlanar range { (xmin, xmax)|(ymin, ymax) two main direction of principal axis on conjugate roof do roof contour slice analysis, then can quickly select the side marginal point of building roof four.
(2), based on the maximum detected edge points of distance between beeline and dot in section
Straight line ax+by+c=0 segmentation planes space in plane is two directed subset spaces, makes the point set in any oriented subspace be combined intoTo point setDo " point is projected on straight line " and calculate subpoint of each point on straight line ax+by+c=0, it is final to obtain projection point setAccording to projection point setAbscissa or ordinate sequence, and take 4 unit dot densities it is done section segmentation.In each section, according to equations point to the distance between straight line：
It is marginal point apart from straight line solstics in each section.
Claims (1)
1. a kind of building roof Fast Edge Detection method based on principal component analysis, its core is：It includes roof data space and is distributed the solution of two principal direction and roof slice analysis detection roof edge axially；
(1) two major axes orientations of building roof are solved based on principal component analysis
Detect that the roof edge contour method of dense three-dimensional discrete point cloud data structure representation relates generally to three committed steps by principal component analytical method (Principal Component Analysis, abbreviation PCA)：
First, for by the three-dimensional intensive cloud data of building roof after filtering and the processing of elevation threshold values, building covariance matrix in horizontal plane xoy, making PxyProjection coordinate of the roof point in xoy planes is represented, be can obtain
Secondly, feature decomposition ApV=λ v, calculate covariance matrix ApTwo characteristic value (λ1, λ2) and its corresponding characteristic vector (v1, v2).The two column vectors represent two Main ways of building roof point cloud distribution where being exactly, the i.e. length of building roof and wide two direction of principal axis.
(2) roof edge point is detected by roof main shaft
Straight line ax+by+c=0 segmentation planes space in plane is two directed subset spaces, makes the point set in any oriented subspace be combined intoTo point setDo " point is projected on straight line " and calculate subpoint of each point on straight line ax+by+c=0, it is final to obtain projection point setAccording to projection point setAbscissa or ordinate sequence, and take a unit dot density subpoint is done its do section segmentation.In the mapping relations of subpoint, the original roof of division to different sections, according to equations point to the directed distance between straight line：
Directed distance diIt is positive and negative identify a little be located at current main shaft certain side.By in each section, being detected to maximum distance in the same direction, identifying marginal point.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610161761.4A CN107220987A (en) | 2016-03-22 | 2016-03-22 | A kind of building roof Fast Edge Detection method based on principal component analysis |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610161761.4A CN107220987A (en) | 2016-03-22 | 2016-03-22 | A kind of building roof Fast Edge Detection method based on principal component analysis |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107220987A true CN107220987A (en) | 2017-09-29 |
Family
ID=59928350
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610161761.4A Pending CN107220987A (en) | 2016-03-22 | 2016-03-22 | A kind of building roof Fast Edge Detection method based on principal component analysis |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107220987A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112801022A (en) * | 2021-02-09 | 2021-05-14 | 青岛慧拓智能机器有限公司 | Method for rapidly detecting and updating road boundary of unmanned mine card operation area |
| CN113989310A (en) * | 2021-10-22 | 2022-01-28 | 广州市城市规划勘测设计研究院 | Method, device and equipment for estimating building volume data and storage medium |
| CN115908424A (en) * | 2023-02-14 | 2023-04-04 | 广东建准检测技术有限公司 | Building health detection method, system and medium based on three-dimensional laser scanning |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2008032551A (en) * | 2006-07-28 | 2008-02-14 | Okamura Printing Industries Co Ltd | Method for calculating roof area by using gauge plate |
| CN102411778A (en) * | 2011-07-28 | 2012-04-11 | 武汉大学 | Automatic registration method of airborne laser point cloud and aerial image |
| CN102915558A (en) * | 2011-08-01 | 2013-02-06 | 李慧盈 | Method for quickly extracting building three-dimensional outline information in onboard LiDAR (light detection and ranging) data |
| US20140198978A1 (en) * | 2013-01-11 | 2014-07-17 | National Central University | Method for searching a roof facet and constructing a building roof structure line |
| CN104036544A (en) * | 2014-06-25 | 2014-09-10 | 西安煤航信息产业有限公司 | Building roof reconstruction method based on airborne LiDAR data |
| CN104751479A (en) * | 2015-04-20 | 2015-07-01 | 中测新图(北京)遥感技术有限责任公司 | Building extraction method and device based on TIN data |
-
2016
- 2016-03-22 CN CN201610161761.4A patent/CN107220987A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2008032551A (en) * | 2006-07-28 | 2008-02-14 | Okamura Printing Industries Co Ltd | Method for calculating roof area by using gauge plate |
| CN102411778A (en) * | 2011-07-28 | 2012-04-11 | 武汉大学 | Automatic registration method of airborne laser point cloud and aerial image |
| CN102915558A (en) * | 2011-08-01 | 2013-02-06 | 李慧盈 | Method for quickly extracting building three-dimensional outline information in onboard LiDAR (light detection and ranging) data |
| US20140198978A1 (en) * | 2013-01-11 | 2014-07-17 | National Central University | Method for searching a roof facet and constructing a building roof structure line |
| CN104036544A (en) * | 2014-06-25 | 2014-09-10 | 西安煤航信息产业有限公司 | Building roof reconstruction method based on airborne LiDAR data |
| CN104751479A (en) * | 2015-04-20 | 2015-07-01 | 中测新图(北京)遥感技术有限责任公司 | Building extraction method and device based on TIN data |
Non-Patent Citations (1)
| Title |
|---|
| 秦家鑫 等: "一种建筑物点云轮廓线的自动提取方法", 《遥感信息》 * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112801022A (en) * | 2021-02-09 | 2021-05-14 | 青岛慧拓智能机器有限公司 | Method for rapidly detecting and updating road boundary of unmanned mine card operation area |
| CN113989310A (en) * | 2021-10-22 | 2022-01-28 | 广州市城市规划勘测设计研究院 | Method, device and equipment for estimating building volume data and storage medium |
| CN115908424A (en) * | 2023-02-14 | 2023-04-04 | 广东建准检测技术有限公司 | Building health detection method, system and medium based on three-dimensional laser scanning |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Fan et al. | Pothole detection based on disparity transformation and road surface modeling | |
| JP6681729B2 (en) | Method for determining 3D pose of object and 3D location of landmark point of object, and system for determining 3D pose of object and 3D location of landmark of object | |
| US8744168B2 (en) | Target analysis apparatus, method and computer-readable medium | |
| EP2249311B1 (en) | Systems and methods for extracting planar features, matching the planar features, and estimating motion from the planar features | |
| CN111915677A (en) | Ship pose estimation method based on three-dimensional point cloud characteristics | |
| EP2887315B1 (en) | Camera calibration device, method for implementing calibration, program and camera for movable body | |
| Cheng et al. | Building boundary extraction from high resolution imagery and lidar data | |
| WO2015096507A1 (en) | Method for recognizing and locating building using constraint of mountain contour region | |
| EP4086846A1 (en) | Automatic detection of a calibration standard in unstructured lidar point clouds | |
| KR20110020718A (en) | Target analysis apparatus and method of the same | |
| CN113628263A (en) | Point cloud registration method based on local curvature and neighbor characteristics thereof | |
| CN111915517A (en) | Global positioning method for RGB-D camera in indoor illumination adverse environment | |
| CN107220987A (en) | A kind of building roof Fast Edge Detection method based on principal component analysis | |
| CN114549549B (en) | Dynamic target modeling tracking method based on instance segmentation in dynamic environment | |
| Sun et al. | Oriented point sampling for plane detection in unorganized point clouds | |
| CN112037282B (en) | Aircraft attitude estimation method and system based on key points and skeleton | |
| Kochi et al. | 3D modeling of architecture by edge-matching and integrating the point clouds of laser scanner and those of digital camera | |
| CN108665470B (en) | Interactive contour extraction method | |
| CN115909099A (en) | Side slope dangerous rock identification and monitoring method based on unmanned aerial vehicle inspection | |
| Tan et al. | Automatic Registration Method of Multi-Source Point Clouds Based on Building Facades Matching in Urban Scenes | |
| Kovacs et al. | Edge detection in discretized range images | |
| Zhang et al. | Feature regions segmentation based RGB-D visual odometry in dynamic environment | |
| Ibelaiden et al. | Scene description from depth images for visually positioning | |
| Ganbold et al. | Coarse-to-fine evolutionary method for fast horizon detection in maritime images | |
| CN109636844B (en) | Complex desktop point cloud segmentation method based on 3D bilateral symmetry |
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 | ||
| WD01 | Invention patent application deemed withdrawn after publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20170929 |