Local sparse representation for astronomical image denoising
来源期刊：中南大学学报(英文版)2013年第10期
论文作者：YANG A-feng(杨阿锋) LU Min(鲁敏) TENG Shu-hua(滕书华) SUN Ji-xiang(孙即祥)
文章页码：2720 - 2727
Key words：astronomical image denoising; local sparse representation (LSR); dictionary learning; alternating optimization
Abstract: Motivated by local coordinate coding (LCC) theory in nonlinear manifold learning, a new image representation model called local sparse representation (LSR) for astronomical image denoising was proposed. Borrowing ideas from surrogate function and applying the iterative shrinkage-thresholding algorithm (ISTA), an iterative shrinkage operator for LSR was derived. Meanwhile, a fast approximated LSR method by first performing a K-nearest-neighbor search and then solving a l1 optimization problem was presented under the guarantee of denoising performance. In addition, the LSR model and adaptive dictionary learning were incorporated into a unified optimization framework, which explicitly established the inner connection of them. Such processing allows us to simultaneously update sparse coding vectors and the dictionary by alternating-optimization method. The experimental results show that the proposed method is superior to the traditional denoising method and reaches state-of-the-art performance on astronomical image.
YANG A-feng(杨阿锋), LU Min(鲁敏), TENG Shu-hua(滕书华), SUN Ji-xiang(孙即祥)
(School of Electronic Science and Engineering, National University of Defense Technology, Changsha 410073, China)
Abstract:Motivated by local coordinate coding (LCC) theory in nonlinear manifold learning, a new image representation model called local sparse representation (LSR) for astronomical image denoising was proposed. Borrowing ideas from surrogate function and applying the iterative shrinkage-thresholding algorithm (ISTA), an iterative shrinkage operator for LSR was derived. Meanwhile, a fast approximated LSR method by first performing a K-nearest-neighbor search and then solving a l1 optimization problem was presented under the guarantee of denoising performance. In addition, the LSR model and adaptive dictionary learning were incorporated into a unified optimization framework, which explicitly established the inner connection of them. Such processing allows us to simultaneously update sparse coding vectors and the dictionary by alternating-optimization method. The experimental results show that the proposed method is superior to the traditional denoising method and reaches state-of-the-art performance on astronomical image.
Key words:astronomical image denoising; local sparse representation (LSR); dictionary learning; alternating optimization
