A noncausal linear prediction based switching median filter for the removal of salt and pepper noise

In this paper, we propose a switching based median filter for the removal of salt and pepper noise in gray scale images. The filter is based on the concept of substitution of noisy pixels prior to estimation. It effectively suppresses the impulse noise in two stages. First, the noisy pixels are detected by using the signal dependent rank-ordered mean (SD-ROM) filter. In the second stage, the noisy pixels are first substituted by the first-order 2D noncausal linear prediction technique and subsequently replaced by the median value. Extensive simulations are carried out to validate the proposed method. Experimental results show improvements both visually and quantitatively compared to other switching based median filters for the removal of salt-and-pepper noise at different densities.