Loading [MathJax]/jax/output/HTML-CSS/config.js
full
search.png

Search

close.png

Cancel

arrow
Volume 10, Issue 4
Learning Non-Negativity Constrained Variation for Image Denoising and Deblurring

Tengda Wei, Linshan Wang, Ping Lin, Jialing Chen, Yangfan Wang & Haiyong Zheng

DOI: 10.4208/nmtma.2017.m1653

Numer. Math. Theor. Meth. Appl., 10 (2017), pp. 852-871.

Published online: 2017-10

Preview Purchase PDF 2221 34765

Cited by

google scholar semantic scholar
Export citation
  • Abstract

This paper presents a heuristic Learning-based Non-Negativity Constrained Variation (L-NNCV) aiming to search the coefficients of variational model automatically and make the variation adapt different images and problems by supervised-learning strategy. The model includes two terms: a problem-based term that is derived from the prior knowledge, and an image-driven regularization which is learned by some training samples. The model can be solved by classical ε-constraint method. Experimental results show that: the experimental effectiveness of each term in the regularization accords with the corresponding theoretical proof; the proposed method outperforms other PDE-based methods on image denoising and deblurring.

  • Keywords

Learning idea, TV-based model, constraint, ε-constraint method, image restoration.

  • AMS Subject Headings

35A15, 35Q93

  • Copyright

COPYRIGHT: © Global Science Press

  • Email address
  • BibTex
  • RIS
  • TXT
@Article{NMTMA-10-852, author = {Tengda Wei, Linshan Wang, Ping Lin, Jialing Chen, Yangfan Wang and Haiyong Zheng}, title = {Learning Non-Negativity Constrained Variation for Image Denoising and Deblurring}, journal = {Numerical Mathematics: Theory, Methods and Applications}, year = {2017}, volume = {10}, number = {4}, pages = {852--871}, abstract = {

This paper presents a heuristic Learning-based Non-Negativity Constrained Variation (L-NNCV) aiming to search the coefficients of variational model automatically and make the variation adapt different images and problems by supervised-learning strategy. The model includes two terms: a problem-based term that is derived from the prior knowledge, and an image-driven regularization which is learned by some training samples. The model can be solved by classical ε-constraint method. Experimental results show that: the experimental effectiveness of each term in the regularization accords with the corresponding theoretical proof; the proposed method outperforms other PDE-based methods on image denoising and deblurring.

}, issn = {2079-7338}, doi = {https://doi.org/10.4208/nmtma.2017.m1653}, url = {http://global-sci.org/intro/article_detail/nmtma/10459.html} }
TY - JOUR T1 - Learning Non-Negativity Constrained Variation for Image Denoising and Deblurring AU - Tengda Wei, Linshan Wang, Ping Lin, Jialing Chen, Yangfan Wang & Haiyong Zheng JO - Numerical Mathematics: Theory, Methods and Applications VL - 4 SP - 852 EP - 871 PY - 2017 DA - 2017/10 SN - 10 DO - http://doi.org/10.4208/nmtma.2017.m1653 UR - https://global-sci.org/intro/article_detail/nmtma/10459.html KW - Learning idea, TV-based model, constraint, ε-constraint method, image restoration. AB -

This paper presents a heuristic Learning-based Non-Negativity Constrained Variation (L-NNCV) aiming to search the coefficients of variational model automatically and make the variation adapt different images and problems by supervised-learning strategy. The model includes two terms: a problem-based term that is derived from the prior knowledge, and an image-driven regularization which is learned by some training samples. The model can be solved by classical ε-constraint method. Experimental results show that: the experimental effectiveness of each term in the regularization accords with the corresponding theoretical proof; the proposed method outperforms other PDE-based methods on image denoising and deblurring.

Tengda Wei, Linshan Wang, Ping Lin, Jialing Chen, Yangfan Wang and Haiyong Zheng. (2017). Learning Non-Negativity Constrained Variation for Image Denoising and Deblurring. Numerical Mathematics: Theory, Methods and Applications. 10 (4). 852-871. doi:10.4208/nmtma.2017.m1653
Copy to clipboard
BibteX RIS TXT
The citation has been copied to your clipboard