Robust Noisiness Measure Based Improved Generalized Fuzzy Peer Group for Removal of Mixed Noise From Color Image

In this letter, a novel method has been proposed to estimate noisiness of a pixel in an image. The proposed noisiness is based on a robust noisiness measure, which is evaluated using noise and edge information of locality of the pixel. Furthermore, this noisiness is used to develop a fuzzy peer group for removing mixed Gaussian and impulse noises from color images. Also, an iterative approach to the proposed method is presented. The proposed approach is then compared with the state-of-the-art algorithms. Results shows that the proposed method outperform existing algorithms significantly.