Rough set theory based segmentation of color images

The theory of rough sets has gained importance and applicability in diverse areas of research, especially in data mining, knowledge discovery, artificial intelligence and information systems analysis. Rough sets have also been used in imaging, however, the application of rough sets for color image analysis has yet to be fully investigated. An interesting strategy for color image segmentation using rough set theory has been presented. A new concept of encrustation of the histogram, called histon, has been proposed for the visualization of multi-dimensional color information in on integrated fashion and its applicability in boundary region analysis has been shown. The histon correlates with the upper approximation of a set such that all elements belonging to this set are clarified as possibly belonging to the same segment or segments showing similar color value. The proposed encrustation provides a direct means of segregating pool of inhomogeneous regions into its components. Experimental results for various images have been presented.