Sparse Overcomplete Denoising: Aggregation Versus Global Optimization
Denoising is often addressed via sparse coding with respect to an overcomplete dictionary. There are two main approaches when the dictionary is composed of translates of an orthonormal basis. The first, traditionally employed by techniques such as wavelet cycle spinning, separately seeks sparsity w.r.t. each translate of the orthonormal basis, solving multiple partial optimizations and obtaining a collection of sparse approximations of the noisefree image, which are aggregated together to obtain a final estimate. The second approach, recently employed by convolutional sparse representations, instead seeks sparsity over the entire dictionary via a global optimization. It is tempting to view the former approach as providing a suboptimal solution of the latter. In this letter, we analyze whether global sparsity is a desirable property, and under what conditions the global optimization provides a better solution to the denoising problem. In particular, our experimental analysis shows that the two approaches attain comparable performance in case of natural images and global optimization outperforms the simpler aggregation of partial estimates only when the image admits an extremely sparse representation. Here, we explain this phenomenon by separately studying the bias and variance of these solutions, and by noting that the variance of the global solution increasesmore »
 Authors:

^{[1]};
^{[1]};
^{[2]};
^{[3]}
 Politecnico di Milano, Milano (Italy)
 Tampere Univ. of Technology, Tampere (Finland)
 Los Alamos National Lab. (LANL), Los Alamos, NM (United States)
 Publication Date:
 Report Number(s):
 LAUR1723531
Journal ID: ISSN 10709908
 Grant/Contract Number:
 AC5206NA25396
 Type:
 Accepted Manuscript
 Journal Name:
 IEEE Signal Processing Letters
 Additional Journal Information:
 Journal Volume: 24; Journal Issue: 10; Journal ID: ISSN 10709908
 Publisher:
 IEEE Signal Processing Society
 Research Org:
 Los Alamos National Lab. (LANL), Los Alamos, NM (United States)
 Sponsoring Org:
 USDOE Laboratory Directed Research and Development (LDRD) Program
 Country of Publication:
 United States
 Language:
 English
 Subject:
 97 MATHEMATICS AND COMPUTING; Computer Science; Information Science; Mathematics; Convolutional sparse coding; denoising; overcomplete representations; sparse representations
 OSTI Identifier:
 1467204
Carrera, Diego, Boracchi, Giacomo, Foi, Alessandro, and Wohlberg, Brendt Egon. Sparse Overcomplete Denoising: Aggregation Versus Global Optimization. United States: N. p.,
Web. doi:10.1109/LSP.2017.2734119.
Carrera, Diego, Boracchi, Giacomo, Foi, Alessandro, & Wohlberg, Brendt Egon. Sparse Overcomplete Denoising: Aggregation Versus Global Optimization. United States. doi:10.1109/LSP.2017.2734119.
Carrera, Diego, Boracchi, Giacomo, Foi, Alessandro, and Wohlberg, Brendt Egon. 2017.
"Sparse Overcomplete Denoising: Aggregation Versus Global Optimization". United States.
doi:10.1109/LSP.2017.2734119. https://www.osti.gov/servlets/purl/1467204.
@article{osti_1467204,
title = {Sparse Overcomplete Denoising: Aggregation Versus Global Optimization},
author = {Carrera, Diego and Boracchi, Giacomo and Foi, Alessandro and Wohlberg, Brendt Egon},
abstractNote = {Denoising is often addressed via sparse coding with respect to an overcomplete dictionary. There are two main approaches when the dictionary is composed of translates of an orthonormal basis. The first, traditionally employed by techniques such as wavelet cycle spinning, separately seeks sparsity w.r.t. each translate of the orthonormal basis, solving multiple partial optimizations and obtaining a collection of sparse approximations of the noisefree image, which are aggregated together to obtain a final estimate. The second approach, recently employed by convolutional sparse representations, instead seeks sparsity over the entire dictionary via a global optimization. It is tempting to view the former approach as providing a suboptimal solution of the latter. In this letter, we analyze whether global sparsity is a desirable property, and under what conditions the global optimization provides a better solution to the denoising problem. In particular, our experimental analysis shows that the two approaches attain comparable performance in case of natural images and global optimization outperforms the simpler aggregation of partial estimates only when the image admits an extremely sparse representation. Here, we explain this phenomenon by separately studying the bias and variance of these solutions, and by noting that the variance of the global solution increases very rapidly as the original signal becomes less and less sparse.},
doi = {10.1109/LSP.2017.2734119},
journal = {IEEE Signal Processing Letters},
number = 10,
volume = 24,
place = {United States},
year = {2017},
month = {7}
}