Abstract
The present work proposes a review and comparison of different Kernel functionals and neighborhood geometry for Nonlocal Means (NLM) in the task of digital image filtering. Some different alternatives to change the classical exponential kernel function used in NLM methods are explored. Moreover, some approaches that change the geometry of the neighborhood and use dimensionality reduction of the neighborhood or patches onto principal component analysis (PCA) are also analyzed, and their performance is compared with respect to the classic NLM method. Mainly, six approaches were compared using quantitative and qualitative evaluations, to do this an homogeneous framework has been established using the same simulation platform, the same computer, and same conditions for the initializing parameters. According to the obtained comparison, one can say that the NLM filtering could be improved when changing the kernel, particularly for the case of the Tukey kernel. On the other hand, the excellent performance given by recent hybrid approaches such as NLM SAP, NLM PCA (PH), and the BM3D SAPCA lead to establish that significantly improvements to the classic NLM could be obtained. Particularly, the BM3D SAPCA approach gives the best denoising results, however, the computation times were the longest.
Similar content being viewed by others
References
Berlinet A, Devroye L (1994) A comparison of kernel density estimates. Publications de l’Intitut de Statistique de l’Université, de Paris 38(3):3–59
Buades A, Coll B, Morel JM (2005) A non-local algorithm for image denoising. In: Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition, vol 2, San Diego, USA, pp 60–65
Buades A, Coll B, Morel JM (2010) Image denoising methods. A new Nonlocal principle. SIAM Rev 52(1):113–147
Buades A, Coll B, Morel JM (2008) Nonlocal image and movie denoising. Int J Comput Vision 76(2):123–139
Buades A, Coll B, Morel JM (2011) Self-similarity-based image denoising. Commun ACM 54(5):109–117
Colom M, Buades A, Morel J-M (2014) Nonparametric noise estimation method for raw images. J Opt Soc Am A 31(4):863–871
Dabov K, Foi A, Katkovnik V, Egiazarian K (2007) Image denoising by sparse 3-D Transform-Domain collaborative filtering. IEEE Trans on Image Processing 16(8):2080–2095
Dabov K, Foi A, Katkovnik V, Egiazarian K (2009) BM3D Image denoising with Shape-Adaptive Principal Componet Analysis. In: Proceedings of the Workshop on Signal Processing with Adaptive Sparse Structured Rep. (SPARS’09), Saint-Malo, France
Danielyan A, Foi A (2009) Noise variance estimation in nonlocal tranform domain. In: Proceedings of the International Workshop on Local and Non-Local Approx. in Image Proc. (NLNA2009), Tuusula, Finland, pp 41–45
De la Rosa JI, Fleury G (2002) On the Kernel selection for Minimum-Entropy estimation. In: Proceedings of the IEEE Instrumentation and Measurement Technology Conference, Anchorage, AK (USA), pp 1205–1210
De la Rosa JI, Fleury G, Davoust M-E (2003) Minimum-entropy, pdf approximation and Kernel selection for measurement estimation. IEEE Trans Instrum Meas 52(4):1009–1020
De la Rosa JI, Villa JJ, De la Rosa E, González E, Gutíerrez O, Escalante N, Ivanov R, Fleury G (2013) MAP Entropy estimation: Applications in robust image filtering. Journal of the European Optical Society-Rapid Publication 8:13047
Deledalle C-A, Denis L, Tupin F (2009) Iterative weighted Maximum Likelihood denoising with probabilistic patch-based weights. IEEE Trans on Image Processing 18 (12):2661–2672
Deledalle C-A, Duval V, Salmon J (2012) Non-local Methods with Shape-Adaptive Patches (NLM-SAP). J Math Imaging Vision 43(2):103–120
Deledalle C-A, Salmon J, Dalalyan AS (2011) Image denoising with patch based PCA: Local versus global. In: Proceedings of the British Machine Vision Conference (BMVC). University of Dundee
Devroye L (1992) A note on the usefulness of superkernels in density estimation. Ann Stat 20:2037–2056
Devroye L (1989) The double kernel method in density estimation. Ann Inst Henri Poincare 25:533– 580
Devroye L (1997) Universal smoothing factor selection in density estimation: theory and practice. Test 6:223–320
Devroye L, Krzyzȧk A (1999) On the Hilbert kernel density estimate. Statistics and Probability Letters 44:299–308
Dinesh PJ, Govindan VK, Abraham TM (2009) Robust estimation approach for nonlocal-means denoising based on structurally similar patches. International Journal of Open Problems in Computer Science and Mathematics 2:293–310
Escalante N, Villa J, De la Rosa I, De la Rosa E, González-Ramírez E, Gutiérrez O, Olvera C, Araiza M (2013) 2-D Continuous Wavelet transform for ESPI phase-maps denoising. Opt Lasers Eng 51(9):1060–1065
Goossens B, Luong H, Pizurica A, Philips W (2008) An improved non-local denoising algorithm. In: Proceedings of International Workshop on Local and Non-Local Approximation in Image Processing (NLNA2008), Lausanne, Switzerland, pp 143–156
http://www.mathworks.com/matlabcentral/fileexchange/13176-non-local-means-filter
Juditsky A, Nemirovsky A (2010) Nonparemetric denoising signals of unknown local structure, II: Nonparametric function recovery. Appl Comput Harmon Anal 29:354–367
Lebrun M, Buades A, Morel J-M (2013) Implemantation of the Non-local Bayes (NL-bayes) image denoising algorithm. Image Processing On Line (IPOL):1–42
Lin L, Lingfu K (2009) Image denoising base on Non-local means with Wiener filtering in wavelet domain. In: Proceedings of the Fifth IEEE International Conference on Intelligent Information Hiding and Multimedia Signal Processing, pp 471–474
Liu YL, Wang J, Chen X, Guo Y-W, Peng Q-S (2008) A robust and fast non-local means algorithm for image denoising. J Comput Sci Technol 23(2):270–279
Loader CM (1999) Bandwidth selection: classical or plug-in? Ann Stat 27 (3):415–438
Maleki A, Narayan M, Baraniuk RG (2013) Anisotropic nonlocal means denoising. Appl Comput Harmon Anal 35:452–482
Masry E (1983) Probability density estimation from sampled data. IEEE Trans Inf Theory 29(5):697– 709
Milanfar P (2013) A tour of modern image filtering. IEEE Signal Process Mag 30(1):106–128
Ponomarenko NN, Lukin VV, Zriakhov MS, Kaarna A, Astola J (2007) An automatic approach to lossy compression of AVIRIS images. In: Proceedings of the IEEE International Geoscience and Remote Sensing Symposium-IGARSS, pp 472–475
Pyatykh S, Hesser J, Zheng L (2013) Image noise level estimation by principal component analysis. IEEE Trans on Image Processing 22(2):687–699
Sutour C, Aujol J-F, Deledalle C-A (2016) Automatic estimation of the noise level function for adaptive blind denoising. In: Proceedings of the 24th IEEE European Signal Processing Conference (EUSIPCO), pp 76–80
Sutour C, Deledalle CA, Aujol JF (2014) Adaptive regularization of the NL-means: Application to image and video denoising. IEEE Trans on Image Processing 23(8):3506–3521
Takeda H, Farsiu S, Milanfar P (2007) Kernel regression for image processing and reconstruction. IEEE Trans on Image Processing 16(2):349–366
Talebi H, Zhu X, Milanfar P (2013) How to SIAF-ly boost denoising performance. IEEE Trans on Image Processing 22(4):1470–1485
Terrell GP (1990) The maximal smoothing principle in density estimation. J Am Stat Assoc 85:470–477
Terrell GP, Scott DW (1985) Oversmoothed nonparametric density estimation. J Am Stat Assoc 80:209–214
Tian J, Yu W-Y, Xie S-L (2008) On the kernel function selection of nonlocal filtering for image denoising. In: Proceedings of the IEEE Seventh International Conference on Machine Learning and Cybernetics, pp 2964–2969
Van De Ville D, Kocher M (2009) SURE-Based non-Local means. IEEE Signal Process Lett 16(11):973–976
Van De Ville D, Kocher M (2011) Non-local means with dimensionality reduction and SURE-based parameter selection. IEEE Trans on Image Processing 20(9):2683–2690
You SJ, Cho NI (2011) A new image denoising method based on the wavelet domain Nonlocal means filtering. In: Proceedings of the ICASSP, vol 2011, pp 1141–1144
You SJ, Cho NI (2013) An adaptive bandwidth nonlocal means image denoising in wavelet domain. EURASIP Journal on Image and Video Processing 60:1–22
Acknowledgments
The first author acknowledge all support given by the Universidad Autónoma de Zacatecas (UAZ) during the years 2014-2015 to realize one academic year (invited professor-researcher) of leave at the Instituto Tecnológico de Sonora (ITSON). Additional acknowledgements for the support given by CONACyT, particularly the postgraduate programs in Electrical Engineering Sciences of both institutions (UAZ and ITSON). Also, our deep gratitude to the Secretaría de Educación Pública (SEP) of Mexico, this work was partially supported through the PIFI 2014-2015 Mexican Program from UAZ. Finally, we acknowledge the valuable comments and recommendations made by two anonymous reviewers.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
de la Rosa, J.I., Villa-Hernández, J., Cortez, J. et al. On the comparison of different kernel functionals and neighborhood geometry for nonlocal means filtering. Multimed Tools Appl 77, 1205–1235 (2018). https://doi.org/10.1007/s11042-016-4322-1
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11042-016-4322-1