A rough set incorporated fuzzy C-means (FCM) algorithm for color image segmentation is introduced. It aims construction of more appropriate clusters in the domain. Dominant peaks in hue (H), saturation (S) and intensity (I) histograms are captured from the input image and all possible combinations of them are taken as initial set of points for processing. Reduction theory of rough set is applied for refinement to the set. The centers thus obtained represent overall pixel colors and hence generate improved clusters when given as input to FCM algorithm. Experiments on several images exhibit effectiveness of the proposed approach. © Springer-Verlag Berlin Heidelberg 2010.
CITATION STYLE
Mandal, B., & Bhattacharyya, B. (2010). Rough Set Integrated Fuzzy C-Means Algorithm for Color Image Segmentation. In Communications in Computer and Information Science (Vol. 101, pp. 339–343). https://doi.org/10.1007/978-3-642-15766-0_51
Mendeley helps you to discover research relevant for your work.