Twofold dynamic attention guided deep network and noise-aware mechanism for image denoising

Convolutional neural networks are given extensive attention towards noise removal due to their good performance over traditional denoising algorithms. With shallow conventional neural networks, the feature extraction ability is not profound. While employing deeper networks, network performance improves with the cost of additional computational requirements. In this paper, we propose an attention-guided twofold denoising network to remove the noise present in the image. The proposed network incorporates dilation convolution to enlarge the receptive fields and improves the feature extraction ability. Also, the presence of attention mechanism strengthens the extracted features and restores the image details during the noise removal. To demonstrate the superiority of the twofold structure, the proposed network is compared with the state-of-the-art denoising models. The experimental results prove that the proposed deep network achieves good peak signal to noise ratio and structural similarity index for different noise levels. (c) 2023 The Author(s). Published by Elsevier B.V. on behalf of King Saud University. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).