An efficient image denoising approach to remove random valued impulse noise by truncating data inside sliding window
Satrughan Kumar, Jigyendra Sen Yadav, Yashwant Kurmi, Arpita Baronia · 2020
Random impulse noise disrupts the homogeneity of the image and deteriorates the other features such as edges, layer depth and sharpness. To alleviate this problem, this paper presents a new algorithm that detects and removes the random valued impulse noise by devising the local statistics inside the sliding window. The corrupt pixel is identified by examining the data under the current window using Gaussian probability distribution function. Meanwhile, the median of the entire data and the mean value of truncated data inside the current window estimates the optimum intensity value at the position where the pixel is disrupted by noise. Simulations show that the proposed method effectively reduces the noise at various level of noise density and yields a better intensity estimate in restored image.