Title :
A New Vanishing Point Detection Algorithm Based on Hough Transform
Author :
Chen, Xuehui ; Jia, Ruiqing ; Ren, Hui ; Zhang, Yinbin
Author_Institution :
Sch. of Mech. Electron. & Inf. Eng., China Univ. of Min. & Technol., Beijing, China
Abstract :
Detecting vanishing points in digital image is important for 3D reconstruction of a scene using no calibrated cameras. A new vanishing point detection algorithm based on twice Hough Transform was presented in this paper. Hough Transform has been recognized as one more popular method for the detection of straight lines. Firstly, the straight lines that in the image are detected by first Hough Transform. Secondly, using a coordinate transform, points in a circle were translated to points in a line. Finally, for detection of the line, Hough Transform was used again. Thus, the positions of vanishing points can be calculated by corresponding to the line parameters in pole coordinate. Experimental results using real images of a scene demonstrate that the proposed algorithm is effective and can acquire the vanishing point.
Keywords :
Hough transforms; image reconstruction; object detection; 3D scene reconstruction; Hough transform; coordinate transform; digital image; vanishing point detection algorithm; Detection algorithms; Digital cameras; Digital images; Drives; H infinity control; Image reconstruction; Image sensors; Layout; Road vehicles; Vehicle driving; Hough Transform; Vanishing point; coordinate transform; line detection; point manual-extraction;
Conference_Titel :
Computational Science and Optimization (CSO), 2010 Third International Joint Conference on
Conference_Location :
Huangshan, Anhui
Print_ISBN :
978-1-4244-6812-6
Electronic_ISBN :
978-1-4244-6813-3
DOI :
10.1109/CSO.2010.163