DocumentCode :
2164032
Title :
A local approach for fast line detection
Author :
Lefevre, S. ; Dixon, C. ; Jeusse, C. ; Vincent, N.
Author_Institution :
Lab. d´´Informatique, Univ. de Tours, France
Volume :
2
fYear :
2002
fDate :
2002
Firstpage :
1109
Abstract :
In this paper we are concerned with line detection for scene modelling. So we focus on horizontal and vertical lines and we propose a fast local approach to detect lines in binary images. The approach described here can easily be extended to lines characterized by other directions. Pixels are analysed using an accumulator on a block-based basis in order to obtain possible line segments for each block. As the results obtained rely only on a local analysis, their coherence can be further improved using a postprocessing step. The proposed method has been compared to classical approaches in order to show its efficiency.
Keywords :
edge detection; image sequences; video signal processing; accumulator; binary images; block-based basis; efficiency; fast local approach; horizontal lines; line detection; pixel analysis; postprocessing step; scene modelling; vertical lines; video sequence understanding; Convolution; Humans; Image analysis; Image edge detection; Image processing; Image segmentation; Layout; Video sequences; Voting;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Digital Signal Processing, 2002. DSP 2002. 2002 14th International Conference on
Print_ISBN :
0-7803-7503-3
Type :
conf
DOI :
10.1109/ICDSP.2002.1028286
Filename :
1028286
Link To Document :
بازگشت