Enhanced CNN for image denoising. Issue 1 (1st March 2019)

• Record Type: Journal Article
• Authors: Tian, Chunwei; Xu, Yong; Fei, Lunke; Wang, Junqian; Wen, Jie; Luo, Nan
• Is Part Of: CAAI transactions on intelligence technology. Volume 4:Issue 1(2019)
• Page Start: 17
• Page End: 23
• Publication Date: 2019-03-01
• Subjects: image restoration -- image representation -- learning (artificial intelligence) -- image denoising -- neural nets -- convolution; image restoration CNN -- image denoising -- enhanced CNN -- flexible architectures -- deep convolutional neural networks -- deep network architecture -- Deeper networks -- performance saturation -- convolutional neural denoising network -- residual learning -- batch normalisation techniques -- training difficulties -- dilated convolutions -- authors; B6135 Optical, image and video signal processing -- C5260B Computer vision and image processing techniques -- C5290 Neural computing techniques; Artificial intelligence -- Periodicals; Computer science -- Periodicals; Artificial intelligence; Computer science; Electronic journals; Periodicals; 006.305
• DOI: 10.1049/trit.2018.1054
• Languages: English
• ISSNs: 2468-6557

Abstract

Abstract : Owing to the flexible architectures of deep convolutional neural networks (CNNs) are successfully used for image denoising. However, they suffer from the following drawbacks: (i) deep network architecture is very difficult to train. (ii) Deeper networks face the challenge of performance saturation. In this study, the authors propose a novel method called enhanced convolutional neural denoising network (ECNDNet). Specifically, they use residual learning and batch normalisation techniques to address the problem of training difficulties and accelerate the convergence of the network. In addition, dilated convolutions are used in the proposed network to enlarge the context information and reduce the computational cost. Extensive experiments demonstrate that the ECNDNet outperforms the state‐of‐the‐art methods for image denoising.