Development of Salt-and-Pepper denoising techniques

This paper proposes various customized methods to remove Salt and Pepper noise from the noisy image. Adaptive filtering and edge preservation based techniques are proposed. Adaptive noise candidate detection is also employed. Proposed techniques preserves high frequency contents and prevents image blurring. Adaptive mean and median filters with variable window sizes are proposed. Techniques were designed to achieve restoration of about 80% of highly corrupted images with the PSNR improvement of 4-25 dB. Techniques developed are of low complexity.