Generalized Fractional Derivative Based Adaptive Algorithm for Image Denoising

This paper presents a new image denoising algorithm based on fractional filters. The fractional filters are derived using a newly introduced fractional operator. The proposed algorithm identifies the noisy pixels based on pixel-density and upgrades them by an adaptive fractional integral mask. To maintain the correlation and recover the lost information, the noise-free pixels are also processed by an adaptive fractional differential mask. We formulate the order function for the fractional mask with the help of gradient features and variance of the image. The algorithm is applied to standard images of different characteristics. The experimental results are compared with some other existing techniques. Evaluation parameters and visual perceptions show that the proposed method performs better than most of the discussed methods. The proposed approach is applicable for image denoising due to its applicability over different types of noises and denoising performance.