A serial attention module‐based deep convolutional neural network for mixed Gaussian‐impulse removal. Issue 6 (16th February 2023)

• Record Type: Journal Article
• Authors: Jiang, Jielin; Yang, Kang; Xu, Xiaolong; Cui, Yan
• Is Part Of: IET image processing. Volume 17:Issue 6(2023)
• Page Start: 1837
• Page End: 1851
• Publication Date: 2023-02-16
• Subjects: batch normalization -- convolutional neural network -- serial attention module; Image processing -- Periodicals; 621.36705
• DOI: 10.1049/ipr2.12759
• Languages: English
• ISSNs: 1751-9659

Abstract

Abstract: The removal of mixed noise is a challenging task because the attenuation of the noise distribution cannot be described precisely. The coupling of additive white Gaussian noise and impulse noise (IN) is a typical case. At present, most methods use a two‐phase strategy, that is, IN detection coupled with additive white Gaussian noise removal, often leading to poor denoising results with an increase in the ratio of IN. In this paper, an effective convolutional neural network (CNN) model is proposed, namely a serial attention module‐based CNN (SACNN), for mixed noise removal. In contrast to the existing two‐phase methods, SACNN unifies the denoising process into a single CNN framework. In SACNN, residual learning and batch normalization are used to train the model, which speeds up the convergence and improves the mixed noise removal performance. Meanwhile, the serial attention module is applied to better preserve the texture details. The experimental results reveal that SACNN achieves superior quality metrics and visual appearance when compared to several leading approaches. Abstract : This paper proposes a serial attention module based deep CNN method. SACNN only needs one step to achieve the different types of mixed noise removal. The serial attention module is introduced into SACNN model, which makes SACNN model allocate resources more effectively. According to the importance of the attention object, the resources are redistributed, the more important units are divided more, and the less important or bad units are divided less, so as to better retain the image details.