Removing a mixture of Gaussian and impulsive noise using the total variation functional and split Bregman iterative method

Abstract

We apply the split Bregman iterative method to minimise the total variation of a piecewise polynomial function to remove Gaussian and impulsive noise from an image. We compare these numerical results with another approach based on the gradient penalty. Both approaches use a finite element method. Numerical results show that the method based on the total variation functional is superior only for one class of images. References

• Plataniotis, K. N. and Venetsanopoulos, A. N. Color Image Processing and Applications. Springer, 2000. doi:10.1007/978-3-662-04186-4
• Chan, T. F. and Shen, J. Image Processing And Analysis: Variational, PDE, Wavelet, And Stochastic Methods. SIAM, 2005. doi:10.1137/1.9780898717877
• Blanchet, G. and Charbit, M. Digital Signal and Image Processing using Matlab. Wiley, 2010. doi:10.1002/9780470612385.
• Preusser, T. and Rumpf, M. An adaptive finite element method for large scale image processing. J. Vis. Commun. Image R. 11:183–195 2000. doi:10.1006/jvci.1999.0444
• Ferrant, M., Warfield, S. K., Nabavi, A., Jolesz, F. A. and Kikinis, R. Registration of 3D intraoperative MR images of the brain using a finite-element biomechanical model. IEEE T. Med. Imaging 20:1384–1397, 2001. doi:10.1007/978-3-540-40899-4_3
• Besdok, E. Impulsive noise suppression from images by using Anfis interpolant and Lillietest. EURASIP J. Appl. Sig. P. 2004:526574, 2004. doi:10.1155/S1110865704403126
• Wang, Z., Qi, F. and Zhou, F. A discontinuous finite element method for image denoising. In Image Analysis and Recognition vol. 4141 of Lecture Notes in Computer Science. Springer Berlin/Heidelberg, 2006. doi:10.1007/11867586_11
• Demaret, L. and Iske, A. Adaptive image approximation by linear splines over locally optimal Delaunay triangulations. IEEE Signal Proc. Lett. 13:281–284, 2006. doi:10.1109/LSP.2006.870358
• Aizenberg, I. and Butakoff, C. Effective impulse detector based on rank-order criteria. IEEE Signal Proc. Lett. 11:363–366, 2004. doi:10.1109/LSP.2003.822925.
• Garnett, R., Huegerich, T., Chui, C. and He, W. A universal noise removal algorithm with an impulse detector. IEEE T. Image Process. 14:1747–1754, 2005. doi:10.1109/TIP.2005.857261
• Donoho, D. L. De-noising by soft-thresholding. IEEE T. Inform. Theory 41:613–627, 1995. doi:10.1109/18.382009.
• Donoho, D. L. and Johnstone, J. M. Ideal spatial adaptation by wavelet shrinkage. Biometrika 81:425–455, 1994. doi: 10.1093/biomet/81.3.425
• Luisier, F., Blu, T. and Unser, M. A new SURE approach to image denoising: interscale orthonormal wavelet thresholding. IEEE T. Image Process. 16:593–606, 2007. doi:10.1109/TIP.2007.891064
• Xu, Q., Ma, L., Li, M., Wang, W., Cai, J., Brunelli, R. and Messelodi, S. Fuzzy weighted average filtering for mixture noises. In Third International Conference on Image and Graphics pp. 18–21, 2004. doi:10.1109/ICIG.2004.70
• Lamichhane, B. P. Finite element techniques for removing mixture of gaussian and impulsive noise. IEEE T. Signal Process. 57:2538–2547, 2009. doi:10.1109/TSP.2009.2016272
• Duchon, J. Splines minimizing rotation-invariant semi-norms in Sobolev spaces. In Constructive Theory of Functions of Several Variables, Lecture Notes in Mathematics 571:85–100. Springer-Verlag Berlin, 1977. doi:10.1007/BFb0086566
• Wahba, G. Spline Models for Observational Data, vol. 59, of Series in Applied Mathematic. SIAM, Philadelphia, 1990. doi:10.1137/1.9781611970128
• Quarteroni, A. and Valli, A. Numerical approximation of partial differential equations. Springer–Verlag, Berlin, 1994. doi:10.1007/978-3-540-85268-1
• Goldstein, T. and Osher, S. The split Bregman method for L1-regularized problems. SIAM J. Imaging Sci. 2:323–343, 2009. doi:10.1137/080725891
• Wang, Y., Yang, J., Yin, W. and Zhang, Y. A new alternating minimization algorithm for total variation image reconstruction. SIAM J. Imaging Sci. 1:248–272, 2008. doi:10.1137/080724265
• van den Doel, K., Ascher, U. M. and Haber, E. The lost honour of l2-based regularization. In Large Scale Inverse Problems Radon Series on Computational and Applied Mathematics 13:181–203. De Gruyter, 2012. http://www.degruyter.com/viewbooktoc/product/182025