In this paper, we present a multi-objective segmentation approach for color images. Three objectives, overall deviation, edge value, and connectivity measure, are optimized simultaneously using a multi-objective evolutionary algorithm (MOEA). To demonstrate the effectiveness of the proposed approach, experiments are conducted on benchmark images. The results justify that the proposed approach is able to partition color images in a number of segments consistent with human visual perception. For quantitative evaluation, we extend the existing Probabilistic Rand Index (PRI) considering multi-objective segmentation. The outcomes show that the proposed approach can obtain non-dominated and near-optimal segment solutions satisfying several criteria simultaneously. It can also find the correct number of segments automatically.
CITATION STYLE
Ripon, K. S. N., Ali, L. E., Newaz, S., & Ma, J. (2017). A multi-objective evolutionary algorithm for color image segmentation. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 10682 LNAI, pp. 168–177). Springer Verlag. https://doi.org/10.1007/978-3-319-71928-3_17
Mendeley helps you to discover research relevant for your work.