An efficient method for salt-and-pepper noise removal based on shearlet transform and noise detection

The conflicting demands for simultaneous salt-and-pepper noise removal and edge preservation still present an outstanding challenge. In this paper we propose an efficient method for salt-and-pepper noise removal. This method is based on shearlet transform with the help of a logic mask, which is generated by the modified boundary discrimination noise detection (MBDND) algorithm and its inverse. We test our method on four images with noise density ranging from 10% to 95% and compare it against seven other efficient methods by calculating the peak-signal-to-noise ratio (PSNR), the measure of structure similarity (SSIM), and image enhancement factor (IEF). In addition, comparisons of visual quality within the region of interests (ROIs) are also carried out. The significant improvement provided by our method is demonstrated by both numerical evaluation and visual quality.