Enhanced Method of Removing Salt and Pepper Noise in Images Using an Adaptive Dual Threshold Fast Median Filter
El Mehdi Cherrat, Rachid Alaoui, Hassane Bouzahir, Wissam Jenkal · International Review on Computers and Software (IRECOS) · 2016
This paper presents an enhanced adaptive dual threshold to remove salt and pepper noise in images using fast median filter. In the first stage, the adaptive dual threshold, a recent method used in signal processing, is employed to detect efficiently the pixels that are likely to be corrupted more by the salt and pepper noise. In the proposed algorithm, this technique is adaptive in the sense that the thresholding coefficient is automatically computed. In the second stage, the fast median filter will be applied only to the corrupted pixel. This pixel is changed either by the median value or by its neighbouring pixel value in the selected window. The performance evaluation of this algorithm is assessed using the peak signal-to-noise ratio (PSNR), the image enhancement factor (IEF) and correlation factor (CF). The simulation results have shown that the proposed method is significantly superior over some recently published works in terms of salt and pepper noise removal, especially when the images are corrupted by a high density noise.