Based On Rough Set Theory Of Image Retrieval

Image is applied in information technology more frequently as one information carrier along with the development of hardware. Thus CBIR become hot research spots of people.This paper is formed of two parts. The first part is a kind of store image retrieval base on color. Another is about morphology operators based on rough set theory.The first part avoids the similar degree defining of pictures . solving classification of pictures and search problem through relation theory. Colors are arranged by area in order. We use main color as the characteristic of the picture and consider the space distribution of colors at the same time. Then based on relation theory, we obtain a method of image reduction。The second part starts with the relation between mathematics morphology filter operators and Approximation Operators, pointing out morphological dilate and erode, open and close can be regarded as dual approximation operators in rough set theory, and presents some new operators. Good qualities are found of the new operators.