Multi‐scale Cross‐path Concatenation Residual Network for Poisson denoising. Issue 8 (21st May 2019)

• Record Type: Journal Article
• Authors: Su, Yueming; Lian, Qiusheng; Zhang, Xiaohua; Shi, Baoshun; Fan, Xiaoyu
• Is Part Of: IET image processing. Volume 13:Issue 8(2019)
• Page Start: 1295
• Page End: 1303
• Publication Date: 2019-05-21
• Subjects: image denoising -- stochastic processes -- learning (artificial intelligence) -- Gaussian noise -- image reconstruction -- image resolution -- convolutional neural nets; signal degradation -- Poisson noise -- low‐light imaging field -- deep learning -- convolution neural network -- image denoising -- deep multiscale cross‐path concatenation residual network -- cross‐path concatenation modules -- multiple paths -- cross‐path concatenation operation -- residual learning strategy -- noisy image -- latent clean image -- discriminative Poisson denoising algorithms -- blind Poisson denoising -- peak signal‐to‐noise ratio -- receptive fields -- MC2RNet -- corrupted pixel reconstruction -- multiscale context representations -- visual effect; Image processing -- Periodicals; 621.36705
• DOI: 10.1049/iet-ipr.2018.5941
• Languages: English
• ISSNs: 1751-9659

Abstract

Abstract : The signal degradation due to the Poisson noise is a common problem in the low‐light imaging field. Recently, deep learning employing the convolution neural network for image denoising has drawn considerable attention owing to its favourable denoising performance. On the basis of the fact that the reconstruction of corrupted pixels can be facilitated by the context information in image denoising, the authors propose a deep multi‐scale cross‐path concatenation residual network (MC 2 RNet) which incorporates cross‐path concatenation modules for Poisson denoising. Multiple paths are achieved by the cross‐path concatenation operation and the skip connection. As a consequence, multi‐scale context representations of images under different receptive fields can be learnt by MC 2 RNet. With the residual learning strategy, MC 2 RNet learns the residual between the noisy image and the latent clean image rather than the direct mapping to facilitate model training. Specially, unlike existing discriminative Poisson denoising algorithms that train a model only for the specific noise level, they aim to train a single model for handling Poisson noise with different levels, i.e. blind Poisson denoising. Quantitative experiments demonstrate that the proposed model is superior over the state‐of‐the‐art Poisson denoising approaches in terms of peak signal‐to‐noise ratio and visual effect.