Abstract
A novel image segmentation method based on modified fuzzy c-means (FCM) is proposed in this paper. By using the neighborhood pixels as spatial information, a spatial constraint penalty term is added in the objective function of standard FCM to improve the robustness. Meanwhile, the neighbor pixels information is used selectively to reduce computing time and iterations. Experiments on real images segmentation proved the availability of the improvement. © 2010 Springer-Verlag Berlin Heidelberg.
Author supplied keywords
Cite
CITATION STYLE
Li, K., Cao, Z., Cao, L., & Liu, M. (2010). An improved FCM algorithm for image segmentation. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 6401 LNAI, pp. 551–556). https://doi.org/10.1007/978-3-642-16248-0_75
Register to see more suggestions
Mendeley helps you to discover research relevant for your work.
