Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Extraction method and system of three-dimensional line segments of scattered point cloud

An extraction method and extraction system technology, which is applied in the field of 3D line segment extraction from scattered point clouds, can solve problems such as the number of projected images, the resolution and the difficulty in determining the viewpoint

Inactive Publication Date: 2018-09-07
SHENZHEN JIMUYIDA TECH CO LTD
View PDF13 Cites 17 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

The extraction method based on two-dimensional image is fully dependent on the image generation

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Extraction method and system of three-dimensional line segments of scattered point cloud
  • Extraction method and system of three-dimensional line segments of scattered point cloud
  • Extraction method and system of three-dimensional line segments of scattered point cloud

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0060] A method 100 for extracting three-dimensional line segments from scattered point clouds, such as figure 1 shown, including:

[0061] Step 110, after constructing a k-d tree for the input scattered point cloud, use the nearest K points around each point in the k-d tree to fit the plane curvature of the point.

[0062] Step 120, divide the scattered point cloud into multiple three-dimensional planes through region growing and region merging according to the plane curvature of each point.

[0063] Step 130. For each 3D plane, project all the points in it onto the preset plane to form a 2D image, extract the 2D line segment of the 2D image, and back-project the 2D line segment onto the 3D plane to obtain 3D line segment.

[0064] Step 140: Obtain new 3D line segments of the scattered point cloud by performing abnormal line segment removal and line segment merging on the 3D line segments corresponding to each 3D plane.

[0065] In the first step of this embodiment, the po...

Embodiment 2

[0067] On the basis of Example 1, as figure 2 As shown, step 120 includes:

[0068] Step 121, sort the points in the scattered point cloud according to the order of the plane curvature of each point from small to large, and start from the point corresponding to the minimum curvature, sequentially determine the coplanar points of each point, and obtain multiple Point cloud regions, where each point cloud region consists of coplanar points.

[0069] Step 122, fitting a normal vector to multiple point cloud regions one by one and determining a fitting plane, setting a label for each point in the fitting plane, and determining the adjacent plane of each fitting plane through the label, and The fitted plane is merged with its corresponding adjacent planes to obtain a plurality of three-dimensional planes.

[0070] It should be noted that step 121 is region growing. Specifically, sort each point in the point cloud data according to the curvature of each point, and first select t...

Embodiment 3

[0076] On the basis of Embodiment 2, step 110 includes:

[0077] Step 111, constructing a k-dtree corresponding to the scattered point cloud.

[0078] Step 112, for each point p in the scattered point cloud i , let the set of its K nearest neighbors in the k-d tree be Construct the covariance matrix Σ:

[0079]

[0080] In the formula, Σ represents a 3×3 covariance matrix, and K is the number of midpoints, yes Average of midpoints.

[0081] Step 113, solve for the eigenvalue of the covariance matrix Σ, and use the minimum eigenvalue as the point p i plane curvature of

[0082] Specifically, the eigenvalue equation is first formed: λV=ΣV. Singular value decomposition (SVD decomposition) of this equation yields three eigenvalues ​​and corresponding three eigenvectors, also known as principal components (PCs). Arrange these three eigenvalues ​​from largest to smallest: λ 2 >λ 1 >λ 0 , and the corresponding three eigenvectors are denoted as v 2 , v 1 and v ...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention relates to an extraction method and system of three-dimensional line segments of a scattered point cloud. The three-dimensional line segments of the large-scale scattered point cloud aredetected and extracted. The method mainly comprises the three steps of: first step, dividing the point cloud into three-dimensional planes through region growth and region merging; second step, realizing two-dimensional projection for each three-dimensional plane, extracting contours on two-dimensional images, obtaining two-dimensional line segments by least square fitting, and then reversely projecting the two-dimensional line segments onto the three-dimensional planes to obtain corresponding three-dimensional line segments; and third step, eliminating outliers through post-processing, and merging adjacent three-dimensional line segments. Being different from methods which are of traditional algorithms and are to extract three-dimensional edge points first and then fit three-dimensionalline segments, the method of the invention realizes extraction of the three-dimensional line segments on the basis of point cloud segmentation and two-dimensional projection, is simple and highly efficient, and is suitable for use in line segment extraction of scattered point clouds in different scenes.

Description

technical field [0001] The invention relates to the technical field of point cloud processing, in particular to a method and system for extracting three-dimensional line segments of scattered point clouds. Background technique [0002] A 3D point cloud is a collection of discrete points with 3D spatial coordinate information. Compared with two-dimensional image data, point cloud data has advantages in dimension, and the coordinate information of three-dimensional points provides a more intuitive spatial description than two-dimensional pixel information. Therefore, point cloud data can better describe the geometry and topology of the real world. With the continuous upgrading of laser scanning technology, 3D laser scanners can quickly and easily obtain high-precision point cloud data of the target object, and the corresponding 3D model can be obtained by using point cloud data modeling. This technology has penetrated into smart cities Construction, machinery manufacturing, ...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
IPC IPC(8): G06T7/187G06T7/11
CPCG06T2207/10028G06T2207/20068G06T7/11G06T7/187
Inventor 姚剑涂静敏鲁小虎谢仁平吴俊霖许哲源
Owner SHENZHEN JIMUYIDA TECH CO LTD
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com