A new unsupervised pseudo-siamese network with two filling strategies for image denoising and quality enhancement
Open Access
- 17 November 2021
- journal article
- research article
- Published by Springer Nature in Neural Computing and Applications
- Vol. 35 (31) , 22855-22863
Abstract
Digital image noise may be introduced during acquisition, transmission, or processing and affects readability and image processing effectiveness. The accuracy of established image processing techniques, such as segmentation, recognition, and edge detection, is adversely impacted by noise. There exists an extensive body of work which focuses on circumventing such issues through digital image enhancement and noise reduction, but this work is limited by a number of constraints including the application of non-adaptive parameters, potential loss of edge detail information, and (with supervised approaches) a requirement for clean, labeled, training data. This paper, developed on the principle of Noise2Void, presents a new unsupervised learning approach incorporating a pseudo-siamese network. Our method enables image denoising without the need for clean images or paired noise images, instead requiring only noise images. Two independent branches of the network utilize different filling strategies, namely zero filling and adjacent pixel filling. Then, the network employs a loss function to improve the similarity of the results in the two branches. We also modify the Efficient Channel Attention module to extract more diverse features and improve performance on the basis of global average pooling. Experimental results show that compared with traditional methods, the pseudo-siamese network has a greater improvement on the ADNI dataset in terms of quantitative and qualitative evaluation. Our method therefore has practical utility in cases where clean images are difficult to obtain.Keywords
Funding Information
- National Natural Science Foundation of China (62002304)
- Natural Science Foundation of Fujian Province (2020J05002)
This publication has 43 references indexed in Scilit:
- Competitive advantage of PET/MRIEuropean Journal of Radiology, 2014
- Medical image denoising using adaptive fusion of curvelet transform and total variationComputers and Electrical Engineering, 2013
- Non-Local Means DenoisingImage Processing On Line, 2011
- A New Method to Remove Noise in Magnetic Resonance and Ultrasound ImagesJournal of Scientific Research, 2010
- Wavelet domain non-linear filtering for MRI denoisingMagnetic Resonance Imaging, 2010
- An Introduction to PET-CT ImagingRadioGraphics, 2004
- Bilateral filtering for gray and color images2002
- Wavelet-based Rician noise removal for magnetic resonance imagingIEEE Transactions on Image Processing, 1999
- Image Selective Smoothing and Edge Detection by Nonlinear DiffusionSIAM Journal on Numerical Analysis, 1992
- Filtering noise from images with wavelet transformsMagnetic Resonance in Medicine, 1991