Image restoration using fuzzy logic with auxiliary BDND filter: A modified approach

This paper proposes a modified approach to filter Salt & Pepper Noise by applying Fuzzy logic with auxiliary BDND algorithm and FSM filtering for grayscale images. Main purpose of this algorithm is to provide an efficient method to differentiate unwanted noise pixels with image edge pixels to restore the details and textures in grayscale images. The approach constitutes two phases: one phase detects the corrupted pixels in the image with the help of modified boundary discriminative noise detection approach along with IF-Then rules to locate salt and pepper noise with higher precession whereas other phase is noise cancellation step which uses fuzzy switching median filter to remove noisy pixels explored by noise detection phase. Extensive simulation result conducted on grayscale images shows that the proposed filter outperforms several existing filters.