3 April 2010 A new corner detection method based on differential code
Author Affiliations +
Proceedings Volume 7651, International Conference on Space Information Technology 2009; 765125 (2010); doi: 10.1117/12.855497
Event: International Conference on Space Information Technology 2009, 2009, Beijing, China
Abstract
Detecting dominant points is an important subject in pattern recognition and computer vision. The paper sums up the traditional corner detection methods based on Freeman codes and presents a new way to detect corner. Firstly, the points are classified using the differential code and the exact corners and uncertain corners are got. Then the curvature of the uncertain corners is calculated, and the points whose curvature is greater than the threshold are candidate corner. At last the right corners based on the Moving Line method is got. Experimental results show that the proposed method is superior to the conventional methods. Besides, the proposed method can be carried out more quickly.
© (2010) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Shi-hu Wu, Xu-chu Yu, Min Xu, "A new corner detection method based on differential code", Proc. SPIE 7651, International Conference on Space Information Technology 2009, 765125 (3 April 2010); doi: 10.1117/12.855497; https://doi.org/10.1117/12.855497
PROCEEDINGS
6 PAGES


SHARE
KEYWORDS
Corner detection

Machine vision

Computer vision technology

Detection and tracking algorithms

Pattern recognition

Lithium

Aerospace engineering

Back to Top