A effective immune multi-objective algorithm for SAR imagery segmentation
Lists
Only you see this
Abstract
A novel and effective immune multi-objective clustering algorithm (IMCA) is presented in this study. Two conflicting and complementary objectives, called compactness and connectedness of clusters, are employed as optimization targets. Besides, adaptive ranks clone, variable length chromosome crossover operation and k-nearest neighboring list based diversity holding strategies are featured by the algorithm. IMCA could automatically discover the right number of clusters with large probability. Seven complicated artificial data sets and two widely used synthetic aperture radar (SAR) imageries are used for test IMCA. Compared with FCM and VGA, IMCA has obtained good and encouraging clustering results. We believe that IMCA is an effective algorithm for solving these nine problems, which should deserve further research.
BibTeX
@inproceedings{yang2009effective,
author = {Dongdong Yang and Licheng Jiao and Maoguo Gong and Xiaoyun Si and Jin‐Ji Li and Jie Feng},
title = {{A effective immune multi-objective algorithm for SAR imagery segmentation}},
booktitle = {Proceedings of SPIE, the International Society for Optical Engineering/Proceedings of SPIE},
year = {2009},
doi = {10.1117/12.832363},
}Loading related content...