Orthogonality is superiority in piecewise-polynomial signal segmentation and denoising

Segmentation and denoising of signals often rely on the polynomial model which assumes that every segment is a polynomial of a certain degree and that the segments are modeled independently of each other. Segment borders (breakpoints) correspond to positions in the signal where the model changes its polynomial representation. Several signal denoising methods successfully combine the polynomial assumption with sparsity. In this work, we follow on this and show that using orthogonal polynomials instead of other systems in the model is beneficial when segmenting signals corrupted by noise. The switch to orthogonal bases brings better resolving of the breakpoints, removes the need for including additional parameters and their tuning, and brings numerical stability. Last but not the least, it comes for free!

[1]  Seyed-Ahmad Ahmadi,et al.  V-Net: Fully Convolutional Neural Networks for Volumetric Medical Image Segmentation , 2016, 2016 Fourth International Conference on 3D Vision (3DV).

[2]  John G. McWhirter,et al.  Polynomial matrix QR decomposition for the decoding of frequency selective multiple-input multiple-output communication channels , 2012, IET Signal Process..

[3]  G. Walter,et al.  Wavelets and Other Orthogonal Systems , 2018 .

[4]  B. Torrésani,et al.  Structured Sparsity: from Mixed Norms to Structured Shrinkage , 2009 .

[5]  Pavel Rajmic,et al.  Piecewise-polynomial signal segmentation using convex optimization , 2018, Kybernetika.

[6]  Michael Elad,et al.  Analysis versus synthesis in signal priors , 2006, 2006 14th European Signal Processing Conference.

[7]  Stephen Arnold,et al.  Polynomial Smoothing of Time Series With Additive Step Discontinuities , 2012, IEEE Transactions on Signal Processing.

[8]  T. Pock,et al.  Second order total generalized variation (TGV) for MRI , 2011, Magnetic resonance in medicine.

[9]  Pavel Rajmic,et al.  Exact risk analysis of wavelet spectrum thresholding rules , 2003, 10th IEEE International Conference on Electronics, Circuits and Systems, 2003. ICECS 2003. Proceedings of the 2003.

[10]  Mladen Victor Wickerhauser.  Mathematics for Multimedia , 2003 .

[11]  D. Donoho,et al.  Simultaneous cartoon and texture image inpainting using morphological component analysis (MCA) , 2005 .

[12]  Mathews Jacob,et al.  Recovery of Discontinuous Signals Using Group Sparse Higher Degree Total Variation , 2015, IEEE Signal Processing Letters.

[13]  Michael Unser,et al.  Splines Are Universal Solutions of Linear Inverse Problems with Generalized TV Regularization , 2016, SIAM Rev..

[14]  Paolo Zaffino,et al.  Deep Neural Networks for Fast Segmentation of 3D Medical Images , 2016, MICCAI.

[15]  Michael Elad,et al.  Sparsity Based Methods for Overparametrized Variational Problems , 2014, SIAM J. Imaging Sci..

[16]  Pierre Vandergheynst,et al.  UNLocBoX A matlab convex optimization toolbox using proximal splitting methods , 2014, ArXiv.

[17]  Fabian J. Theis,et al.  TREVOR HASTIE, ROBERT TIBSHIRANI, AND MARTIN WAINWRIGHT. Statistical Learning with Sparsity: The Lasso and Generalizations. Boca Raton: CRC Press. , 2018, Biometrics.

[18]  Pavel Rajmic,et al.  Image Edges Resolved Well When Using an Overcomplete Piecewise-Polynomial Model , 2018, 2018 12th International Conference on Signal Processing and Communication Systems (ICSPCS).

[19]  R. Tibshirani.  Regression Shrinkage and Selection via the Lasso , 1996 .

[20]  Michael Elad,et al.  Optimally sparse representation in general (nonorthogonal) dictionaries via ℓ1 minimization , 2003, Proceedings of the National Academy of Sciences of the United States of America.

[21]  Michael Unser,et al.  Splines: a perfect fit for signal and image processing , 1999, IEEE Signal Process. Mag..

[22]  Michael Elad,et al.  The Cosparse Analysis Model and Algorithms , 2011, ArXiv.

[23]  Wang-Q Lim,et al.  Compactly supported shearlets are optimally sparse , 2010, J. Approx. Theory.

[24]  Jean-Philippe Vert,et al.  The group fused Lasso for multiple change-point detection , 2011, 1106.4199.

[25]  Ivan W. Selesnick,et al.  Sparsity-Assisted Signal Smoothing , 2015 .

[26]  Karl Kunisch,et al.  On Infimal Convolution of TV-Type Functionals and Applications to Video and Image Reconstruction , 2014, SIAM J. Imaging Sci..

[27]  John G. McWhirter,et al.  Relevance of polynomial matrix decompositions to broadband blind signal separation , 2017, Signal Process..

[28]  Pavel Rajmic,et al.  Piecewise-polynomial curve fitting using group sparsity , 2016, 2016 8th International Congress on Ultra Modern Telecommunications and Control Systems and Workshops (ICUMT).

[29]  Laurent Condat,et al.  A Primal–Dual Splitting Method for Convex Optimization Involving Lipschitzian, Proximable and Linear Composite Terms , 2012, Journal of Optimization Theory and Applications.

[30]  Ivan W. Selesnick,et al.  Convex 1-D Total Variation Denoising with Non-convex Regularization , 2015, IEEE Signal Processing Letters.

[31]  P ? ? ? ? ? ? ? % ? ? ? ? , 1991 .

[32]  Ivan W. Selesnick,et al.  Sparsity-assisted signal smoothing (revisited) , 2017, 2017 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP).

[33]  Alfred M. Bruckstein,et al.  On Globally Optimal Local Modeling: From Moving Least Squares to Over-parametrization , 2013, Innovations for Shape Analysis, Models and Algorithms.

[34]  C. O’Brien.  Statistical Learning with Sparsity: The Lasso and Generalizations , 2016 .

[35]  Pavel Rajmic,et al.  On the limitation of convex optimization for sparse signal segmentation , 2016, 2016 39th International Conference on Telecommunications and Signal Processing (TSP).

[36]  R. Tibshirani.  Adaptive piecewise polynomial estimation via trend filtering , 2013, 1304.2986.

[37]  Muhammad Usman,et al.  Signal processing with adaptive sparse structured representations (SPARS'09) , 2009 .

[38]  Pavel Rajmic,et al.  Piecewise-Polynomial signal segmentation using reweighted convex optimization , 2017, 2017 40th International Conference on Telecommunications and Signal Processing (TSP).

[39]  Kristian Bredies,et al.  A TGV-Based Framework for Variational Image Decompression, Zooming, and Reconstruction. Part I: Analytics , 2015, SIAM J. Imaging Sci..

[40]  Vítezslav Veselý,et al.  Change Point Detection by Sparse Parameter Estimation , 2011, Informatica.

[41]  Stephen P. Boyd,et al.  Enhancing Sparsity by Reweighted ℓ1 Minimization , 2007, 0711.1612.

[42]  Laurent Condat,et al.  A Direct Algorithm for 1-D Total Variation Denoising , 2013, IEEE Signal Processing Letters.

[43]  Laurent Condat,et al.  A Generic Proximal Algorithm for Convex Optimization—Application to Total Variation Minimization , 2014, IEEE Signal Processing Letters.

[44]  Paolo Prandoni,et al.  Signal Processing for Communications , 2008, Communication and information sciences.

[45]  Jun Geng,et al.  Multiple Change-Points Estimation in Linear Regression Models via Sparse Group Lasso , 2015, IEEE Transactions on Signal Processing.

[46]  Stephen P. Boyd,et al.  1 Trend Filtering , 2009, SIAM Rev..