Image Denoising by Incorporating Noisy Image Patch and Impulse Response of Low-Pass Filter in CNN Learning
May Thet Tun, Yosuke Sugiura, Tetsuya Shimamura · 2023
In this paper, we present a novel image denoising approach which incorporates the noisy image patch and the impulse response of low-pass filter (LPF) in convolutional neural network (CNN) learning. We demonstrate the performance of the proposed method for spatially variant additive white Gaussian noise (AWGN) and realistic noise removal. Our method outperforms the conventional fast and flexible denoising CNN (FFDNet) in preserving image information and perceptual quality, particularly in high noise variance.