Representation Learning via Cauchy Convolutional Sparse Coding

In representation learning, Convolutional Sparse Coding (CSC) enables unsupervised learning of features by jointly optimising both an \(\ell_2\)-norm fidelity term and a sparsity enforcing penalty. This work investigates using a regularisation term derived from an assumed Cauchy prior for the coefficients of the feature maps of a CSC generative model. The sparsity penalty term resulting from this prior is solved via its proximal operator, which is then applied iteratively, element-wise, on the coefficients of the feature maps to optimise the CSC cost function. The performance of the proposed Iterative Cauchy Thresholding (ICT) algorithm in reconstructing natural images is compared against the common choice of \(\ell_1\)-norm optimised via soft and hard thresholding. ICT outperforms IHT and IST in most of these reconstruction experiments across various datasets, with an average PSNR of up to 11.30 and 7.04 above ISTA and IHT respectively.

[1]  Michael Elad,et al.  Image Denoising Via Sparse and Redundant Representations Over Learned Dictionaries , 2006, IEEE Transactions on Image Processing.

[2]  Michael Elad,et al.  Sparse Coding with Anomaly Detection , 2013, Journal of Signal Processing Systems.

[3]  Tieyong Zeng,et al.  Variational Approach for Restoring Blurred Images with Cauchy Noise , 2015, SIAM J. Imaging Sci..

[4]  T. Blumensath,et al.  Iterative Thresholding for Sparse Approximations , 2008 .

[5]  Michael Unser,et al.  Dictionary Learning Based on Sparse Distribution Tomography , 2017, ICML.

[6]  M. Kramer Nonlinear principal component analysis using autoassociative neural networks , 1991 .

[7]  I. Daubechies,et al.  An iterative thresholding algorithm for linear inverse problems with a sparsity constraint , 2003, math/0307152.

[8]  Alin Achim,et al.  Image denoising using bivariate α-stable distributions in the complex wavelet domain , 2005, IEEE Signal Processing Letters.

[9]  Heng Tao Shen,et al.  Principal Component Analysis , 2009, Encyclopedia of Biometrics.

[10]  Michael Elad,et al.  Compression of facial images using the K-SVD algorithm , 2008, J. Vis. Commun. Image Represent..

[11]  David J. Field,et al.  Sparse coding with an overcomplete basis set: A strategy employed by V1? , 1997, Vision Research.

[12]  Pascal Vincent,et al.  Representation Learning: A Review and New Perspectives , 2012, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[13]  Peilin Liu,et al.  A Survey on Nonconvex Regularization-Based Sparse and Low-Rank Recovery in Signal Processing, Statistics, and Machine Learning , 2018, IEEE Access.

[14]  Jie Li,et al.  Convolutional sparse coding classification model for image classification , 2016, 2016 IEEE International Conference on Image Processing (ICIP).

[15]  Ling Shao,et al.  Simultaneous Super-Resolution and Cross-Modality Synthesis of 3D Medical Images Using Weakly-Supervised Joint Convolutional Sparse Coding , 2017, 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[16]  Terrence J. Sejnowski,et al.  An Information-Maximization Approach to Blind Separation and Blind Deconvolution , 1995, Neural Computation.

[17]  Jon Atli Benediktsson,et al.  Adaptive Spectral–Spatial Compression of Hyperspectral Image With Sparse Representation , 2017, IEEE Transactions on Geoscience and Remote Sensing.

[18]  I. Horev,et al.  Adaptive image compression using sparse dictionaries , 2012, 2012 19th International Conference on Systems, Signals and Image Processing (IWSSIP).

[19]  A. Bruckstein,et al.  K-SVD : An Algorithm for Designing of Overcomplete Dictionaries for Sparse Representation , 2005 .

[20]  Pascal Frossard,et al.  Dictionary Learning , 2011, IEEE Signal Processing Magazine.

[21]  Changsheng Xu,et al.  Low-Rank Sparse Coding for Image Classification , 2013, 2013 IEEE International Conference on Computer Vision.

[22]  M. Omair Ahmad,et al.  Spatially Adaptive Wavelet-Based Method Using the Cauchy Prior for Denoising the SAR Images , 2007, IEEE Transactions on Circuits and Systems for Video Technology.

[23]  N. Canagarajah,et al.  Segmentation of noisy colour images using cauchy distribution in the complex wavelet domain , 2011 .

[24]  Alin Achim,et al.  Convergence Guarantees for Non-Convex Optimisation With Cauchy-Based Penalties , 2020, IEEE Transactions on Signal Processing.

[25]  Andrew Y. Ng,et al.  The Importance of Encoding Versus Training with Sparse Coding and Vector Quantization , 2011, ICML.

[26]  Dmitry Malioutov,et al.  Iterative log thresholding , 2013, 2014 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP).

[27]  Alin Achim,et al.  Novel Bayesian multiscale method for speckle removal in medical ultrasound images , 2001, IEEE Transactions on Medical Imaging.

[28]  M. Elad,et al.  $rm K$-SVD: An Algorithm for Designing Overcomplete Dictionaries for Sparse Representation , 2006, IEEE Transactions on Signal Processing.

[29]  Yiqiu Dong,et al.  Cauchy Noise Removal by Nonconvex ADMM with Convergence Guarantees , 2017, Journal of Scientific Computing.

[30]  Qingwei Gao,et al.  Directionlet-based denoising of SAR images using a Cauchy model , 2013, Signal Process..

[31]  Graham W. Taylor,et al.  Deconvolutional networks , 2010, 2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition.

[32]  Rui Fan,et al.  Super-resolution CT Image Reconstruction Based on Dictionary Learning and Sparse Representation , 2018, Scientific Reports.

[33]  J. Jennifer Ranjani,et al.  Dual-Tree Complex Wavelet Transform Based SAR Despeckling Using Interscale Dependence , 2010, IEEE Transactions on Geoscience and Remote Sensing.