The digital CR image can be divided into object region and background region

and useful information about diagnosis is in the former one. In image processing

the two regions should be treated with different ways. In this paper

a fuzzy set is used to describe the two regions

and the membership degree function is established by fuzzy statistics. The image processing model based on the fussy set theory is constructed and the processing method is given in detail. Experimental results show that the proposed methods are both effective and feasible. In the processed images

medical information is impressed clearly

thus improving the diagnosis accuracy.