Adaptive type-2 fuzzy median filter design for removal of impulse noise

Abstract

Image detail preservation and impulse noise attenuation are difficult to achieve simultaneously in the area of image restoration design. This study proposes a novel adaptive type-2 fuzzy median (A2FM) filter to accomplish these objectives. The novel filter is based mainly on the uncertainty of the capacity of the type-2 fuzzy sets to provide the advantages of two filters: the Arakawa proposed fuzzy median (FM) filter and the general type-1 based FM (type-1 FM) filter. The proposed method of designing an A2FM filter, adopts a powerful scheme for overcoming restrictions of the membership function in the type-1 FM filter. It also comprises a flexibility inference mechanism to reduce the memory usage of the FM filter. Extensive simulation results reveal that the A2FM filter not only exhibits desirable robustness in suppressing noise but also outperforms other proposed filtering approaches.