Fuzzy hybrid filter for removal of impulse noise from highly corrupted images
Minggang Ma, Xiao Jiao, Xinquan Tan · 2003
A fuzzy hybrid filter (FHF) is presented which can remove impulse noise from highly corrupted images while preserving details. Primarily, the noise ratio in the filter window is defined and detected. Then we present a new detection algorithm of the corrupted degree of central pixel in the filter window, and fuzzy decision based on fuzzy rules, construct the fuzzy membership function. Finally, a fuzzy hybrid filter (FHF) is designed for the removal of impulse noise from highly corrupted images. Experimental results indicate that our algorithm impressively outperforms other techniques in terms of noise suppression and detail preservation across a wide range of impulse noise corruption, ranging from 5 % to 80 %. Moreover the FHF can effectively restore images corrupted with mixed Gaussian and impulse noise.