4 January 2006 A self-adaptive ant colony optimization approach for image segmentation
Author Affiliations +
Proceedings Volume 5985, International Conference on Space Information Technology; 59853F (2006) https://doi.org/10.1117/12.657993
Event: International Conference on Space information Technology, 2005, Wuhan, China
Abstract
Ant colony optimization, with good discretion, parallel, robustness and positive feedback, is well suited to image segmentation. But its search is random and has much computation for convergence. Constant evaporating coefficient leads to early convergence or stagnation. To improve it, the ideal of setting primary cluster center is proposed. Meanwhile, the algorithm is implemented in a small window so as to reduce its computation. The evaporating coefficient is also set to change with the number of ants which pass the allowable path in order to keep its good global convergence and stability. This method can segment an image accurately. Experimental results show it's an effective approach.
© (2006) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Jue Lu, "A self-adaptive ant colony optimization approach for image segmentation", Proc. SPIE 5985, International Conference on Space Information Technology, 59853F (4 January 2006); doi: 10.1117/12.657993; https://doi.org/10.1117/12.657993
PROCEEDINGS
6 PAGES


SHARE
Back to Top