Abstract
Data-driven techniques are becoming popular in imaging. There is growing interest in learning signal models for various applications. We describe recent research that developed efficient, scalable, and effective data-driven models and methodologies for imaging and image processing. Efficient sparsifying transform learning methods have been proposed, often incorporating various properties such as union-of-transforms, rotation invariance, etc. Transform learning-based approaches achieve high-quality results for X-ray computed tomography (CT) or magnetic resonance image (MRI) reconstruction from limited data. We also discuss recent work on efficient methods for synthesis dictionary learning, including in combination with low-rank models. Newly proposed dictionary-based algorithms provide promising results for dynamic MRI reconstruction from limited measurements.
© 2017 Optical Society of America
PDF ArticleMore Like This
Didem Dogan and Figen S. Oktem
CW4B.4 Computational Optical Sensing and Imaging (COSI) 2020
Matthew Tancik, Tristan Swedish, Guy Satat, and Ramesh Raskar
IW2B.6 Imaging Systems and Applications (IS) 2018
Xingchen He, Lianshan Yan, Lin Jiang, Anlin Yi, Zhengyu Pu, Youren Yu, Wei Pan, and Bin Luo
SM4J.6 CLEO: Science and Innovations (CLEO:S&I) 2022