Compressed Classification from Learned Measurements

This work proposes a deep compressed learning framework inferring classification directly from the compressive measurements. While classical approaches separately sense, reconstruct signals, and apply classification on these reconstructions, we jointly learn the sensing and classification schemes utilizing a deep neural network with a novel loss function. Our approach employs a data-driven reconstruction network within the compressed learning framework utilizing a weighted loss that combines both in-network reconstruction and classification losses. The proposed network structure also learns the optimal measurement matrices for the goal of enhancing classification performance. Quantitative results demonstrated on CIFAR-10 image dataset show that the proposed framework provides better classification performance and robustness to noise compared to the tested state of the art deep compressed learning approaches.

[1]  Richard G. Baraniuk,et al.  The smashed filter for compressive classification and target recognition , 2007, Electronic Imaging.

[2]  Ali Mousavi,et al.  Learning to invert: Signal recovery via Deep Convolutional Networks , 2017, 2017 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP).

[3]  Hao Xu,et al.  Learning Sparse Representations of High Dimensional Data on Large Scale Dictionaries , 2011, NIPS.

[4]  R. Calderbank,et al.  Compressed Learning : Universal Sparse Dimensionality Reduction and Learning in the Measurement Domain , 2009 .

[5]  Bülent Sankur,et al.  Compressively Sensed Image Recognition , 2018, 2018 7th European Workshop on Visual Information Processing (EUVIP).

[6]  Pavan K. Turaga,et al.  ReconNet: Non-Iterative Reconstruction of Images from Compressively Sensed Measurements , 2016, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[7]  Pavan Turaga,et al.  Convolutional Neural Networks for Noniterative Reconstruction of Compressively Sensed Images , 2017, IEEE Transactions on Computational Imaging.

[8]  J. Haupt,et al.  Compressive Sampling for Signal Classification , 2006, 2006 Fortieth Asilomar Conference on Signals, Systems and Computers.

[9]  Jian Sun,et al.  Deep Residual Learning for Image Recognition , 2015, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[10]  R.G. Baraniuk,et al.  Compressive Sensing [Lecture Notes] , 2007, IEEE Signal Processing Magazine.

[11]  Vinh Nguyen Xuan.  A Deep Learning Framework for Compressed Learning and Signal Reconstruction , 2018 .

[12]  S. Frick,et al.  Compressed Sensing , 2014, Computer Vision, A Reference Guide.

[13]  Deanna Needell,et al.  Constrained Adaptive Sensing , 2015, IEEE Transactions on Signal Processing.

[14]  R. H. Md. Rafi,et al.  Data Driven Measurement Matrix Learning for Sparse Reconstruction , 2019, 2019 IEEE Data Science Workshop (DSW).

[15]  A. Robert Calderbank,et al.  Finding needles in compressed haystacks , 2012, 2012 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP).

[16]  Andrew Zisserman,et al.  Very Deep Convolutional Networks for Large-Scale Image Recognition , 2014, ICLR.

[17]  Richard G. Baraniuk,et al.  DeepCodec: Adaptive sensing and recovery via deep convolutional neural networks , 2017, 2017 55th Annual Allerton Conference on Communication, Control, and Computing (Allerton).

[18]  Michael Elad,et al.  Compressed Learning: A Deep Neural Network Approach , 2016, ArXiv.

[19]  P. Alam.  ‘L’ , 2021, Composites Engineering: An A–Z Guide.

[20]  Bernard Ghanem,et al.  ISTA-Net: Iterative Shrinkage-Thresholding Algorithm Inspired Deep Network for Image Compressive Sensing , 2017, ArXiv.

[21]  Geoffrey E. Hinton,et al.  ImageNet classification with deep convolutional neural networks , 2012, Commun. ACM.

[22]  Nikos Komodakis,et al.  Wide Residual Networks , 2016, BMVC.

[23]  Ali Cafer Gurbuz,et al.  Learning to Sense and Reconstruct A Class of Signals , 2019, 2019 IEEE Radar Conference (RadarConf).

[24]  Guangming Shi,et al.  Adaptive Measurement Network for CS Image Reconstruction , 2017, CCCV.

[25]  Wilhelm Burger,et al.  Digital Image Processing - An Algorithmic Introduction using Java , 2008, Texts in Computer Science.

[26]  Kilian Q. Weinberger,et al.  Densely Connected Convolutional Networks , 2016, 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[27]  Pavan K. Turaga,et al.  Direct inference on compressive measurements using convolutional neural networks , 2016, 2016 IEEE International Conference on Image Processing (ICIP).

[28]  Robiulhossain Mdrafi,et al.  Joint Learning of Measurement Matrix and Signal Reconstruction via Deep Learning , 2020, IEEE Transactions on Computational Imaging.

[29]  M. Elad,et al.  Compressed Learning for Image Classification: A Deep Neural Network Approach , 2018 .

[30]  E.J. Candes,et al.  An Introduction To Compressive Sampling , 2008, IEEE Signal Processing Magazine.