Vision-based corridor line detection using k-means algorithm

Author

Lho, Hyung-Suk ; Kwon, Ji-Wook ; Chwa, Dongkyoung ; Hong, Suk-Kyo

Author_Institution

WONCHUN-DONG YEONGTONG-GU SUWON, Suwon

fYear

2007

fDate

17-20 Oct. 2007

Firstpage

1052

Lastpage

1056

Abstract

We propose a corridor lines extraction algorithm. The algorithm is based on a vision-system. It uses the patterns of slopes between two points, K-means, and least-square method. The proposed algorithm offers intersection lines between the wall and the floor. The proposed method consists of image preprocessing, noise filtering, and line extraction algorithm stage. The method has been tested on a large number of real corridor images. The experimental results are included to demonstrate the validity of the proposed algorithm.

Keywords

filtering theory; image denoising; least squares approximations; mobile robots; object detection; pattern clustering; robot vision; corridor lines extraction algorithm; image preprocessing; k-means clustering algorithm; least-square method; mobile robot; noise filtering; vision-based corridor line detection; Cameras; Clustering algorithms; Data mining; Detectors; Humans; Image edge detection; Image segmentation; Mobile robots; Partitioning algorithms; Robot vision systems; Image Processing; K-means; The Corridor Line Extraction; Vision-System;

fLanguage

English

Publisher

ieee

Conference_Titel

Control, Automation and Systems, 2007. ICCAS '07. International Conference on

Conference_Location

Seoul

Print_ISBN

978-89-950038-6-2

Electronic_ISBN

978-89-950038-6-2

Type

conf

DOI

10.1109/ICCAS.2007.4407053

Filename

4407053