Title :
An Algorithm for Extracting Geometric Features from Point Cloud
Author :
Pang, Xu-Fang ; Pang, Ming-Yong
Author_Institution :
Dept. of Educ. Technol., Nanjing Normal Univ., Nanjing, China
Abstract :
We proposed a robust algorithm for valley-ridge extraction from point set. Our algorithm separately flag potential valley points and ridge points according to principle curvature of every point, then enhance the valley-ridge points by projecting them onto their local nearest potential valleys or ridges. Using an optimized principal covariance analysis approach, we smooth the projected points, finally valley-ridge lines are obtained after resolving gaps and relaxing the results.
Keywords :
covariance analysis; feature extraction; surface fitting; geometric feature extraction; principal covariance analysis; ridge points; valley points; valley-ridge extraction algorithm; Clouds; Data mining; Feature extraction; Least squares approximation; Least squares methods; Multilevel systems; Robustness; Shape; Surface fitting; Surface reconstruction; MLS surface fitting; ridges; valley-ridge extraction; valleys;
Conference_Titel :
Information Management, Innovation Management and Industrial Engineering, 2009 International Conference on
Conference_Location :
Xi´an
Print_ISBN :
978-0-7695-3876-1
DOI :
10.1109/ICIII.2009.481
