© The Institution of Engineering and Technology
Recently, image denoising algorithm based on sparse representation has received an increasing amount of attention. Such algorithms proposed a comprehensive sparse representation model, by solving the sparse coding problem and choosing the proper method for dictionary updating to achieve better denoising results. Therefore, the construction of learning dictionary has become one of the key problems that limit the denoising effectiveness. The nonlocally centralised sparse representation denoising algorithm uses principal component analysis method to achieve dictionary updating. Nevertheless, the instability of a single complete dictionary in sparse coding leads to erratic result in the process of the original image restoration. In this study, the authors present a new method named generalised nonlocally centralised sparse representation algorithm. In the proposed method, the authors cluster the training patches extracted from a set of example images into subspaces, and then train dictionaries for subspaces by sparse analysis ksingular value decomposition dictionary, which is utilised to construct coded subblock dictionary to avoid the instable results caused by a single dictionary. Experiments show that the improved method has better signaltonoise ratio and denoising effect compared with other methods.
References


1)

1. Li, T., Wang, W., Xu, L., et al: ‘Image denoising using lowrank dictionary and sparse representation’. Tenth Int. Conf. on Computational Intelligence and Security, Kunming, 2014, pp. 228–232.

2)

2. Giryes, R., Elad, M.: ‘Poisson denoising using sparse representations and dictionary learning’, IEEE Trans. Image Process, 2014, 23, pp. 5057–5069.

3)

3. Yang, J., Zhao, Y.Q., Chan, J.C.W., et al: ‘Coupled sparse denoising and unmixing with lowrank constraint for hyperspectral image’, IEEE Trans. Geosci. Remote Sens., 2016, 54, (3), pp. 1818–1833.

4)

4. Irofti, P.: ‘Sparse denoising with learned composite structured dictionaries’. Int. Conf. on System Theory, Control and Computing (ICSTCC), Cheile Gradistei, 2015, pp. 331–336.

5)

5. Cai, R., Weng, S., Luo, B., et al: ‘Optimizing the framework of image denoising based on sparse and redundant representations’. Chinese Control and Decision Conf. (CCDC), Yinchuan, 2016, pp. 2757–2762.

6)

6. Liu, Z., Yu, L., Zhang, M., et al: ‘Nonlocal structured nonparametric Bayesian dictionary learning for image denoising’. 2016 IEEE 13th Int. Conf. on Signal Processing (ICSP), 2016, pp. 144–148.

7)

7. Buades, A., Coll, B., Morel, J.M.: ‘A nonlocal algorithm for image denoising’. 2005 IEEE Computer Society Conf. on Computer Vision and Pattern Recognition, 2005, , pp. 60–65.

8)

8. Dong, W., Zhang, L., Shi, G.: ‘Centralized sparse representation for image restoration’. 2011 IEEE Int. Conf. on Computer Vision (ICCV), 2011, pp. 1259–1266.

9)

9. Liu, Q., Zhang, C., Guo, Q., et al: ‘Adaptive sparse coding on PCA dictionary for image denoising’, The Visual Computer, 2016, 32, (4), pp. 535–549.

10)

10. Sulam, J., Elad, M.: ‘Expected patch log likelihood with a sparse prior’. Energy Minimization Methods in Computer Vision and Pattern Recognition, 2015 (, pp. 99–111.

11)

11. Mairal, J., Bach, F., Ponce, J., et al: ‘Nonlocal sparse models for image restoration’. IEEE 12th Int. Conf. on Computer Vision, Kyoto, 2009, pp. 2272–2279.

12)

12. Dong, W.S., Zhang, L., Shi, G.M., et al: ‘Nonlocal centralized sparse representation for image restoration’, IEEE Trans. Image Process., 2013, 22, (4), pp. 1620–1630.

13)

13. Yang, J., Wright, J., Huang, T.S.: ‘Image superresolution via sparse representation’, IEEE Trans. Image Process., 2010, 19, (11), pp. 2861–2873.

14)

14. Koutsogiannis, G.S., Soraghan, J.J.: ‘Kernel principal component analysis (KPCA) for the denoising of communication signals’. 11th European Signal Processing Conf., Toulouse, 2002, pp. 1–4.

15)

15. Qayyum, A., Malik, A.S., Naufal, M., et al: ‘Designing of overcomplete dictionaries based on DCT and DWT’. IEEE Student Symp. in Biomedical Engineering & Sciences (ISSBES), Shah Alam, 2015, pp. 134–139.

16)

16. Engan, K., Aase, S.O., Husoy, J.H.: ‘Frame based signal compression using method of optimal directions (MOD)’. Proc. of the 1999 IEEE Int. Symp. on Circuits and Systems, Orlando, FL, 1999, pp. 1–4.

17)

17. Coumar, S.O., Rajesh, P., Sadanandam, S.: ‘Image restoration using filters and image quality assessment using reduced reference metrics’, Circuits Controls Commun., 2013, 115, (6), pp. 1–5.

18)

18. Fu, Y., Lam, A., Sato, I., et al: ‘Adaptive spatialspectral dictionary learning for hyperspectral image restoration’, Int. J. Comput. Vis., 2017, 122, (2), pp. 228–245.

19)

19. Rubinstein, R., Faktor, T., Elad, M.: ‘KSVD dictionarylearning for the sparse sparse model’. IEEE Int. Conf. on Acoustics, Speech and Signal Processing (ICASSP), Kyoto, 2012, pp. 5405–5408.

20)

20. Rubinstein, R., Peleg, T., Elad, M.: ‘KSVD dictionarylearning for the analysis sparse model’, IEEE Trans. Signal Process., 2012, 61, (3), pp. 661–677.

21)

21. Romano, Y., Elad, M.: ‘Improving KSVD denoising by postprocessing its methodnoise’. Image Processing, Melbourne, Australia, 2013, pp. 435–439.

22)

22. Su, P., Liu, T., Sun, Z.: ‘KSVD based image denoising method using image residual information in different frequency bands’. Int. Conf. on Intelligent Computing, 2016, pp. 482–492.

23)

23. Goklani, H.S., Sarvaiya, J.N., Fahad, A.M.: ‘Image reconstruction using orthogonal matching pursuit (OMP) algorithm’. Emerging Technology Trends in Electronics, Communication and Networking, Surat, 2014, pp. 1–5.

24)

24. Dong, W., Shi, G., Li, X.: ‘Nonlocal image restoration with bilateral variance estimation: a lowrank approach’, IEEE Trans. Image Process., 2013, 22, (2), pp. 700–711.

25)

25. Dabov, K., Foi, A., Katkovnik, V., et al: ‘Image denoising by sparse 3d transformdomain collaborative filtering’, IEEE Trans. Image Process., 2007, 16, (8), pp. 2080–2095.
http://iet.metastore.ingenta.com/content/journals/10.1049/ietipr.2017.0783
Related content
content/journals/10.1049/ietipr.2017.0783
pub_keyword,iet_inspecKeyword,pub_concept
6
6