A fast and reliable median filter for extremely corrupted images

This paper proposes a fast and reliable median filter for denoising extremely corrupted images by impulsive noise. A switching median filter with boundary discriminative noise detection (BDND) is very effective and outperforms all previously proposed median-based filters. However, it is very time-consuming in calculation. The proposed method uses similar scheme as in BDND, but it is much more reliable and faster than BDND method. The proposed method uses an efficient sorting algorithm both in noise detection and adaptive filtering stages. The performance of the proposed method is demonstrated through computer simulations in comparison with the BDND method.