Abstract
The existence of noise is inevitable in real-world applications of digital image processing. Theoretically, image restoration is the process to recover high-quality images from noisy images using adequate techniques. One of the pivotal applications of natural image restoration is the noise reduction (denoising). In this study, clustering-based natural image denoising using dictionary learning algorithm in wavelet domain is proposed (CDLW). This algorithm is exploiting the second-generation wavelet clustering coefficients in the decomposition levels. The use of second-generation wavelet transform in the proposed algorithm has its purposes which promotes the sparsity and multiresolution representations. The significance of second-generation wavelet transform is utilized in order to promote the hierarchical property of CDLW, while sparsity with self-similarity of the image source is utilized to connect the clustered coefficients. Extensive experiments have been conducted in order to show the objective and subjective competitive performance and have shown convincing improvements over the best state-of-the-art denoising methods. Inspired by the nonlocal features of the proposed algorithm structure, the time complexity comparisons showed that CDLW has the most efficient performance in the execution time compared to the rest of algorithms under investigation.
Similar content being viewed by others
References
Ahmed N, Natarajan T, Rao KR (1974) Discrete cosine transform. IEEE Trans Comput 100(1):90–93
Cai N, Zhou Y, Wang S, Ling BWK, Weng S (2016) Image denoising via patch-based adaptive Gaussian mixture prior method. SIViP 10(6):993–999
Candès EJ, Donoho DL (2002) Recovering edges in ill-posed inverse problems: optimality of curvelet frames. Ann Stat 30(3):784–842
Candes E, Donoho D (2004) New tight frames of curvelets and the problem of approximating piecewise C2 images with piecewise C2 edges. Commun Pure Appl Math 57:219–266
Chatterjee P, Milanfar P (2009) Clustering-based denoising with locally learned dictionaries. IEEE Trans Image Process 18(7):1438–1451
Dabov K, Foi A, Katkovnik V, Egiazarian K (2007) Image denoising by sparse 3-D transform-domain collaborative filtering. IEEE Trans Image Process 16(8):2080–2095
Ding Y, Selesnick IW (2015) Artifact-free wavelet denoising: non-convex sparse regularization, convex optimization. IEEE Signal Process Lett 22(9):1364–1368
Donoho DL (1999) Wedgelets: nearly minimax estimation of edges. Ann Stat 27(3):859–897
Efron B, Hastie T, Johnstone I, Tibshirani R (2004) Least angle regression. Ann Stat 32(2):407–499
Elad M (2012) Sparse and redundant representation modeling—What next? IEEE Signal Process Lett 19(12):922–928
Elad M, Aharon M (2006) Image denoising via sparse and redundant representations over learned dictionaries. IEEE Trans Image Process 15(12):3736–3745
Foi A, Katkovnik V, Egiazarian K (2007) Pointwise shape-adaptive DCT for high-quality denoising and deblocking of grayscale and color images. IEEE Trans Image Process 16(5):1395–1411
Gao J, Wang Q (2016) BM3D image denoising algorithm based on k-means clustering. In: China academic conference on printing & packaging and media Technology, Springer, Singapore, pp 265–272
Gao S, Tsang IW-H, Chia L, Zhao P (2010) Local features are not lonely–Laplacian sparse coding for image classification. In: IEEE conference on computer vision and pattern recognition (CVPR)
Goossens B, Luong H, Pizurica A, Philips W (2008) An improved non-local denoising algorithm. In: Local and non-local approximation in image processing, international workshop, proceedings, 2008
Katkovnik V, Foi V, Egiazarian V, Astola J (2010) From local kernel to nonlocal multiple-model image denoising. Int J Comput Vis 86(1):1–32
Khmag A, Abd Rahman R, Al-Haddad SAR, Hashim SJ, Zarina MN, Abdulmawla AM (2015) Image denoising algorithm using second generation wavelet transformation and principle component analysis. J Med Imag Health Inf 5(6):1261–1266
Khmag A, Ramli A, Al-Haddad SAR, Hashim SJ (2016a) Additive noise reduction in natural images using second-generation wavelet transform hidden Markov models. IEEJ Trans Electr Electron Eng 11(3):339–347
Khmag A, Ramli AR, Al-haddad SAR, Yusoff S, Kamarudin NH (2016b) Denoising of natural images through robust wavelet thresholding and genetic programming. Vis Comput 33(9):1141–1154
Khmag A, Al-Haddad SAR, Ramlee RA, Kamarudin NH, Malallah FL (2018a) Natural image noise removal using non local means and hidden Markov models in stationary wavelet transform domain. Multimed Tools Appl 77(15):20065–20086
Khmag A, Ramli A, Al-haddad SAR, Kamarudin N (2018b) Natural image noise level estimation based on local statistics for blind noise reduction. The Visual Computer 34(4):575–587
Le Pennec E, Mallat S (2005a) Sparse geometric image representations with bandelets. IEEE Trans Image Process 14(4):423–438
Le Pennec E, Mallat S (2005b) Bandelet image approximation and compression. Multiscale Model Simul 4(3):992–1039
Liu H, Xiong R, Zhang J, Gao W (2015) Image denoising via adaptive soft-thresholding based on non-local samples. In: Proceedings of the IEEE conference on computer vision and pattern recognition, p 484–492
Maggu J, Majumdar A (2016) Alternate formulation for transform learning. In: Proceedings of the tenth Indian conference on computer vision, graphics and image processing. ACM, p 50
Mairal J, Sapiro G, Elad M (2007) Learning multiscale sparse representations for image and video restoration, DTIC document
Ophir B, Lustig M, Elad M (2011) Multi-scale dictionary learning using wavelets. Sel Top IEEE J Signal Process 5(5):1014–1024
Portilla J, Strela V, Wainwright M, Simoncelli E (2003) Image denoising using scale mixtures of Gaussians in the wavelet domain. IEEE Trans Image Process 12(11):1338–1351
Shao L, Zhang H, De Haan G (2008) An overview and performance evaluation of classification-based least squares trained filters. IEEE Trans Image Process 17(10):1772–1782
Shao L, Yan R, Li X, Liu Y (2014) From heuristic optimization to dictionary learning: a review and comprehensive comparison of image denoising algorithms. IEEE Trans Cybern 44(7):1001–1013
Shapiro JM (1993) Embedded image coding using zerotrees of wavelet coefficients. IEEE Trans Signal Process 41(12):3445–3462
Shapiro L, Stockman GC (2001) Computer vision. Prentice Hall, Upper Saddle River
Smith SM, Brady JM (1997) SUSAN—a new approach to low level image processing. Int J Comput Vis 23(1):45–78
Wang Y-Q (2016) Small neural networks can denoise image textures well: a useful complement to BM3D. Image Process Line 6:1–7
Wang Z, Bovik AC, Sheikh HR, Simoncelli EP (2004) Image quality assessment: from error visibility to structural similarity. IEEE Trans Image Process 13(4):600–612
Wang Y, He Z, Zi Y (2010) Enhancement of signal denoising and multiple fault signatures detecting in rotating machinery using dual-tree complex wavelet transform. Mech Syst Signal Process 24(1):119–137
SIPI, U., The USC-SIPI image database. http://sipi.usc.edu/database/. Accessed 3 Aug 2018
Yan R, Shao L, Liu Y (2013) Nonlocal hierarchical dictionary learning using wavelets for image denoising. IEEE Trans Image Process 22(12):4689–4698
Acknowledgements
The authors are grateful to the referees for their critical and valuable suggestions.
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Conflict of interest
The authors declare that there is no conflict of interests regarding the publishing of this paper.
Additional information
Communicated by V. Loia.
Rights and permissions
About this article
Cite this article
Khmag, A., Ramli, A.R. & Kamarudin, N. Clustering-based natural image denoising using dictionary learning approach in wavelet domain. Soft Comput 23, 8013–8027 (2019). https://doi.org/10.1007/s00500-018-3438-9
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00500-018-3438-9