Abstract
Non-local means (NLM) filtering is an effective and popular image denoising algorithm. It estimates the pixel by taking advantage of redundancy present in a whole image or in a predefined fixed search region. NLM algorithm using fixed size search region for all pixels is unable to preserve the important details such as edges and texture in an image due to inclusion of unrelated pixels in the averaging. The selection of variable size search region for each pixel in an image is a critical issue in NLM algorithm. This paper focuses on the selection of optimal size search region for each pixel according to the characteristics of the region. The proposed algorithm adaptively selects an optimal size search region for a pixel based on maximization of ratio of contribution of similar pixels to dissimilar pixels for different search regions centered on that pixel. Experimental results on standard natural test images show that the proposed algorithm performs consistently better than the conventional NLM and other state-of-the-art NLM variants in terms of PSNR (dB), SSIM and visual quality at various noise levels.
Similar content being viewed by others
Notes
Images database link: https://www.cs.cmu.edu/~cil/v-images.html
References
Berkovich H, Malah D, Barzohar M (2013) Non-local means denoising using a content-based search region and dissimilarity kernel. In Image and Signal Processing and Analysis (ISPA), 2013 8th International Symposium on
Brownlee KA, Brownlee KA (1965) Statistical theory and methodology in science and engineering, vol. 150. Wiley, New York
Buades A, Coll B, Morel J-M (2005) A review of image denoising algorithms, with a new one. Multiscale Model Simul 4(2):490–530
Buades A, Coll B, Morel J-M (2010) Image denoising methods. A new nonlocal principle. SIAM Rev 52(1):113–147
Buades A, Coll B, Morel J-M (2011) Non-local means denoising. Image Process On Line 1. doi:10.5201/ipol.2011.bcm_nlm
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
de la Rosa Vargas JI, Villa JJ, Gonz E, Cortez J et al. (2016) A tour of nonlocal means techniques for image filtering. In 2016 International Conference on Electronics, Communications and Computers (CONIELECOMP)
Efros AA, Leung TK (1999) Texture synthesis by non-parametric sampling. In Computer vision, 1999. The Proceedings of the Seventh IEEE International Conference on
Gonzalez RC, Woods RE (2002) Digital image processing. Prentice Hall, Upper Saddle River
Hu J, Luo Y-P (2013) Non-local means algorithm with adaptive patch size and bandwidth. Optik-Int J Light Electron Opt 124(22):5639–5645
Kumar BKS (2013) Image denoising based on non-local means filter and its method noise thresholding. SIViP 7(6):1211–1227
Lefkimmiatis S, Maragos P, Papandreou G (2009) Bayesian inference on multiscale models for Poisson intensity estimation: applications to photon-limited image denoising. IEEE Trans Image Process 18(8):1724–1741
Mahmoudi M, Sapiro G (2005) Fast image and video denoising via nonlocal means of similar neighborhoods. Signal Processing Letters IEEE 12(12):839–842
Makitalo M, Foi A (2011) Optimal inversion of the Anscombe transformation in low-count Poisson image denoising. IEEE Trans Image Process 20(1):99–109
Makitalo M, Foi A (2013) Optimal inversion of the generalized Anscombe transformation for Poisson-Gaussian noise. IEEE Trans Image Process 22(1):91–103
Mallat SG (1989) A theory for multiresolution signal decomposition: the wavelet representation. IEEE Trans Pattern Anal Mach Intell 11(7):674–693
Mallat S (1999) A wavelet tour of signal processing. Academic press, California
Milanfar P (2013) A tour of modern image filtering: new insights and methods, both practical and theoretical. Signal Process Mag IEEE 30(1):106–128
Perona P, Malik J (1990) Scale-space and edge detection using anisotropic diffusion. IEEE Trans Pattern Anal Mach Intell 12(7):629–639
Shreyamsha Kumar BK (2012) Image denoising based on gaussian/bilateral filter and its method noise thresholding. SIViP 7(6):1159–1172
Tasdizen T (2009) Principal neighborhood dictionaries for nonlocal means image denoising. Image Processing IEEE Transactions on 18(12):2649–2660
Thacker NA, Manjon JV, Bromiley PA (2010) Statistical interpretation of non-local means. Comput Vis, IET 4(3):162–172
Tiku M (2004) Noncentral chi-square distribution. In Encyclopedia of statistical sciences. John Wiley & Sons, Inc. doi:10.1002/0471667196.ess1798.pub2
Tomasi C, Manduchi R (1998) Bilateral filtering for gray and color images. In Computer Vision, 1998. Sixth International Conference on
Van De Ville D, Kocher M (2009) SURE-based non-local means. Signal Process Lett, IEEE 16(11):973–976
Van De Ville D, Kocher M (2011) Nonlocal means with dimensionality reduction and SURE-based parameter selection. Image Processing, IEEE Transactions on 20(9):2683–2690
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
Wu Y, Tracey B, Natarajan P, Noonan JP (2013a) James--stein type center pixel weights for non-local means image denoising. Signal Process Lett IEEE 20(4):411–414
Wu Y, Tracey B, Natarajan P, Noonan JP (2013b) Probabilistic non-local means. Signal Process Lett IEEE 20(8):763–766
Xiao F, Zhang Y (2011) A comparative study on thresholding methods in wavelet-based image denoising. Protein Eng 15:3998–4003
Zeng WL, Lu XB (2011) Region-based non-local means algorithm for noise removal. Electron Lett 47(20):1125–1127
Zhang B, Fadili JM, Starck J-L (2008) Wavelets, ridgelets, and curvelets for Poisson noise removal. IEEE Trans Image Process 17(7):1093–1108
Zhu S, Li Y, Li Y (2014) Two-stage non-local means filtering with adaptive smoothing parameter. Optik-Int J Light Electron Optics 125(23):7040–7044
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Verma, R., Pandey, R. A statistical approach to adaptive search region selection for NLM-based image denoising algorithm. Multimed Tools Appl 77, 549–566 (2018). https://doi.org/10.1007/s11042-016-4227-z
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11042-016-4227-z