invansc package v3.00 16.3.2016 [for Matlab v.7 (or later)] ------------------------------------------------------------------- Lucio Azzari, Markku Mäkitalo, Alessandro Foi The Matlab software included in this package implements the transformations and filtering procedures for Poisson and Poisson-Gaussian data presented in the papers [1] M. Mäkitalo and A. Foi, "On the inversion of the Anscombe transformation in low-count Poisson image denoising", Proc. Int. Workshop on Local and Non-Local Approx. in Image Process., LNLA 2009, Tuusula, Finland, pp. 26-32, August 2009. doi:10.1109/LNLA.2009.5278406 [2] M. Mäkitalo and A. Foi, "Optimal inversion of the Anscombe transformation in low-count Poisson image denoising", IEEE Trans. Image Process., vol. 20, no. 1, pp. 99-109, January 2011. doi:10.1109/TIP.2010.2056693 [3] M. Mäkitalo and A. Foi, "A closed-form approximation of the exact unbiased inverse of the Anscombe variance-stabilizing transformation", IEEE Trans. Image Process., vol. 20, no. 9, pp. 2697-2698, September 2011. doi:10.1109/TIP.2011.2121085 [4] M. Mäkitalo and A. Foi, "Optimal inversion of the generalized Anscombe transformation for Poisson-Gaussian noise", IEEE Trans. Image Process., vol. 22, no. 1, pp. 91-103, January 2013. doi:10.1109/TIP.2012.2202675 [5] L. Azzari and A. Foi, "Variance Stabilization for Noisy+Estimate Combination in Iterative Poisson Denoising", submitted, March 2016. Software and publications can be downloaded from http://www.cs.tut.fi/~foi/invansc/ ------------------------------------------------------------------- Contents ------------------------------------------------------------------- demo_Poisson_experiments_table.m - script reproducing the results in Table 1 of the LNLA2009 paper [1]. demo_GenAnscombe_denoising.m - script reproducing the results in Table 1 of the paper [4] demo_iterVSTpoisson.m - script reproducing the results in Table 1 and Table 2 of paper [5]. Poisson_denoising_Anscombe_exact_unbiased_inverse.m - function for denoising Poisson images using the Anscombe variance-stabilizing transformation, the BM3D filter, and the exact unbiased inverse of the Anscombe transformation iterVSTpoisson.m - function for denoising Poisson images using the iterative VST framework [5] Anscombe_forward.m - Anscombe variance-stabilizing transformation GenAnscombe_forward.m - generalized Anscombe variance-stabilizing transformation Anscombe_inverse_asympt_unbiased.m - asymptotically unbiased inverse of the Anscombe transformation Anscombe_inverse_exact_unbiased.m - exact unbiased inverse of the Anscombe transformation GenAnscombe_inverse_exact_unbiased.m - exact unbiased inverse of the generalized Anscombe transformation Anscombe_inverse_closed_form.m - closed-from approximation of exact unbiased inverse of the Anscombe transformation GenAnscombe_inverse_closed_form.m - closed-from approximation of exact unbiased inverse of the generalized Anscombe transformation Anscombe_inverse_MMSE.m - MMSE inverse of the Anscombe transformation bin_B_h.m - binning function debin_Binv_h.m - debinning function Anscombe_vectors.mat - precomputed expectation vectors used for defining the exact unbiased inverse of the Anscombe transformation GenAnscombe_vectors.mat - precomputed expectation vectors and matrix used for defining the exact unbiased inverse of the generalized Anscombe transformation MMSEcurves.mat - precomputed curves used for defining the MMSE inverse of the Anscombe transformation Anscombe_lambda.mat - precomputed expectation vectors and matrix for defining the exact unbiased inverse of noisy+estimate combinations [5] paramsFromQfun.mat - function handle to set parameters in iterVSTpoisson.m [5] ./images (folder) - test images used for the simulation experiments [2] ./images/images_readme.txt - information about the test images contained in images.mat and used for the simulation experiments ./images_for_table_1 (folder) - test images used for Table 1 of [5] ./images_for_table_2 (folder) - test images used for Table 2 of [5] LEGAL_NOTICE.txt - TUT license and disclaimers ReadMe_Contents.txt - this file ------------------------------------------------------------------ Installation and requirements ------------------------------------------------------------------ This software is designed to run on *) MS Windows, Linux, or Mac OSX (32-bit or 64-bit CPU) *) Matlab v.7 or later The denoising functions are configured to use: * BM3D denoising filter http://www.cs.tut.fi/~foi/GCF-BM3D/ The demo scripts use also: * Statistics Toolbox (to generate Poisson data "poissrnd"), * Image Processing Toolbox (only for visualization with "imshow"). ------------------------------------------------------------------- Change log ------------------------------------------------------------------- v3 (16 March 2016) + files for iterative VST method [5] v2.03 (10 August 2012) + bugfix (handling of default value of alpha=1) v2.02 (9 August 2012) + bugfix (sigma normalized by alpha in the codes of the inverses) v2.01 (2 October 2011) + updated some comments. v2.00 (1 October 2011) + added routines for the Poisson-Gaussian case based on the generalized Anscombe transformation [4]. + added closed-form approximation of exact unbiased inverse [2]. v1.02 (5 November 2009) + bugfix ('negative values' in Anscombe_inverse_asympt_unbiased.m). v1.01 (5 November 2009) + added MMSE inverse. v1.00 (9 September 2009) * Initial release [1]. ------------------------------------------------------------------- Disclaimer ------------------------------------------------------------------- Any unauthorized use of these routines for industrial or profit-oriented activities is expressively prohibited. By downloading and/or using any of these files, you implicitly agree to all the terms of the TUT limited license, as specified in the document LEGAL_NOTICE.txt (included in this package) and online at http://www.cs.tut.fi/~foi/invansc/legal_notice.html Tampere University of Technology - 2011-2016 -----------------------------------------------------------------------------------