Removal of salt and pepper noise using robust M-filter
Vinod Kumar, G Nanalya · 2016
In recent days, camera manufacturers continue to pack increasing numbers of pixels per unit area, an increase in noise sensitivity manifests itself in the form of noisier image. Camera manufacturers, therefore, depend on image denoising algorithms to reduce the effects of such noise artifacts in the resultant image. The most common type of noise is impulse noise, in which the affected pixels are replaced by noise values which will be either zero or one. First, the noise pixel can be identified by switching method. Second, the median filter applied on every corrupted pixel. Third, compare the neighboring pixel value on input image with the median filtered image, if the two pixels having a difference then apply the M-filter on that. This paper mainly deals with the detection of noisy pixels and applying robust M-filter (lorentzian) method to improve the peak signal to noise ratio (PSNR), structural similarity index (SSIM) and mean square error (MSE) value. The experimental results show that the proposed algorithm significantly outperforms other state-of-the-art image denoising methods such as adaptive median filter, decision based filter in terms of both objective measure and visual evaluation.