An improved stochastic restoration method using digital filter for medical X-ray images contaminated by quantum mottles

The degradation due to the quantum mottles of X-ray images causes many difficulties in medical diagnosis. In this paper, by especially introducing the general type image processing method of X-ray images on the basis of the actual physical formation process, a new restoration method for the noisy image is proposed. More specifically, as a noise removal method for the quantum mottles, a digital filter of a stochastic type considering the specific statistical properties reflecting the variability of variance of quantum mottles is newly derived on the basis of Bayes' theorem suitable for the recursive image processing. Finally in order to confirm the effectiveness of the proposed method, it is applied to one of the actual medical images. (author)