Image restoration by adaptive fuzzy optimal filter

Abstract

Filtering is a fundamental technique in image processing. Median filter is one of the most popular nonlinear filters and it preserves edges in image while removing impulsive noise effectively. Fuzzy filters are a novel class of nonlinear operators that have been demonstrated to be capable of addressing image processing applications. In this paper, we use a new nonlinear fuzzy filter, namely adaptive fuzzy optimal filter, which is shown more powerful for image restoration than conventional median filters. It can also improve the weakness of adaptive fuzzy median filter in the case that the noise rate is below 20%.