DocumentCode :
2350255
Title :
Segmentation and parameterization of 2D lines based on Mean Shift Clustering
Author :
Berrío, Julie Stephany ; Paz, Lina María ; Bravo, Eduardo Caicedo
Author_Institution :
Electr. & Electron. Eng. Sch., Valle Univ., Cali, Colombia
fYear :
2012
fDate :
2-4 May 2012
Firstpage :
1
Lastpage :
6
Abstract :
This paper presents a robust algorithm for segmentation and characterization of lines detected by a laser sensor. We propose a strategy of Mean Shift Clustering which using the points of the laser scan performs a classification stage based on an ellipsoidal orientable window previous to the line segment parameterization. Each data set is processed by a RANSAC (Random Sample and Consensus) algorithm modified to detect spurious. This method reduces the amount of spurious and updates associated probability densities. The parameters of the detected segments are estimated by TLS (Total Least Squares) regression. The algorithm has been evaluated in indoor environments using as mobile platform the robot Pioneer 3DX equipped with a SICK laser.
Keywords :
geometry; least squares approximations; mobile robots; optical scanners; probability; random processes; regression analysis; robot vision; 2D line parameterization; 2D line segmentation; Pioneer 3DX robot; RANSAC; SICK laser; TLS regression; classification stage; ellipsoidal orientable window; indoor environments; laser scan; laser sensor; mean shift clustering; mobile platform; parameter estimation; probability densities; random sample and consensus algorithm; robust algorithm; spurious detection; total least squares regression; Clustering algorithms; Data models; Kernel; Lasers; Robot sensing systems; Vectors; laser; mean shift; segmentation;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Engineering Applications (WEA), 2012 Workshop on
Conference_Location :
Bogota
Print_ISBN :
978-1-4673-0871-7
Type :
conf
DOI :
10.1109/WEA.2012.6220085
Filename :
6220085
Link To Document :
بازگشت