Research On Algorithm For Image Segmentation Based On Rough Set Theory

Rough set theory is proposed by a Polish mathematician-Pawlak Z. in 1982, which is a new mathematical tool to deal with vagueness and uncertainty problems. The main advantage of rough set theory is that it has no use for any preliminary or additional information about data. Image segmentation is the basis of feature extraction and object recognition of image. Therefore, by means of the advantage of rough set theory, some investigations were done, and some main problems are mainly researched in the paper as follows:(1) Based on rough set theory, the definition of the lower approximation and upper approximation of a image are given, and the Liang's entropy of a image is also be proposed. Furthermore, the maximum entropy principle is used to determine the size of window and the value of threshold. And a rough set based image segmentation algorithm is designed, whose validation is indicated by the results of experiment show the validation of(2) A new histon is proposed, it depends on a kind of distance function that consider the weight of the RGB components in color image. Moreover, a roughness based algorithm is given, which segment RGB images. The experiment show the proposed algorithm is valide.Finally, achieved main results in the paper are useful to image segmentation based on rough set theory, and some problems farther needed and remarkable research direction for the future are indicated.