DocumentCode :
2307086
Title :
Segmentation and outlier removal in 3D line identification based on fuzzy clustering
Author :
Nguyen, Thach B. ; Sukhan, Lee
Author_Institution :
Intell. Syst. Res. Center, Sungkyunkwan Univ., Suwon, South Korea
fYear :
2010
fDate :
18-23 July 2010
Firstpage :
1
Lastpage :
8
Abstract :
In this paper, we present a novel method based on clustering for identifying 3D line from point clouds, called “self-organizing fuzzy k-means algorithm”. The algorithm automatically finds the optimal number of cluster and self organizes the clusters based on inter/intra-cluster distances and cluster´s performance evaluation. The self-organizing fuzzy k-means is applied in 3D line identification from point clouds. We use the point clouds provided by Stereo camera and 2D images. The 3D point clouds of each line is clustered by clustering algorithm, then we perform eigen-analysis on clusters and estimate the final 3D lines; the 3D lines can be cut off into several segments. In addition, to increase the accuracy of detection, the error evaluation is invoked to analyze the error of the 3D candidate lines. Our algorithm was evaluated on the real test scenes, which content noisy point clouds, and shows the high performance and robust results.
Keywords :
cameras; edge detection; eigenvalues and eigenfunctions; error analysis; fuzzy set theory; image segmentation; object recognition; pattern clustering; solid modelling; stereo image processing; 2D images; 3D candidate line identification; 3D point clouds; cluster performance evaluation; eigenanalysis; error analysis; error evaluation; fuzzy clustering; inter-intracluster distances; self-organizing fuzzy k-means algorithm; stereo camera; Algorithm design and analysis; Cameras; Clouds; Clustering algorithms; Feature extraction; Shape; Three dimensional displays;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Fuzzy Systems (FUZZ), 2010 IEEE International Conference on
Conference_Location :
Barcelona
ISSN :
1098-7584
Print_ISBN :
978-1-4244-6919-2
Type :
conf
DOI :
10.1109/FUZZY.2010.5584309
Filename :
5584309
Link To Document :
بازگشت