Deep learning for SAR image formation

The recent success of deep learning has lead to growing interest in applying these methods to signal processing problems. This paper explores the applications of deep learning to synthetic aperture radar (SAR) image formation. We review deep learning from a perspective relevant to SAR image formation. Our objective is to address SAR image formation in the presence of uncertainties in the SAR forward model. We present a recurrent auto-encoder network architecture based on the iterative shrinkage thresholding algorithm (ISTA) that incorporates SAR modeling. We then present an off-line training method using stochastic gradient descent and discuss the challenges and key steps of learning. Lastly, we show experimentally that our method can be used to form focused images in the presence of phase uncertainties. We demonstrate that the resulting algorithm has faster convergence and decreased reconstruction error than that of ISTA.

[1]  Kurt Hornik,et al.  Approximation capabilities of multilayer feedforward networks , 1991, Neural Networks.

[2]  Birsen Yazici,et al.  Passive Imaging of Moving Targets Using Sparse Distributed Apertures , 2012, SIAM J. Imaging Sci..

[3]  Hean-Teik Chuah,et al.  A Comparison of Autofocus Algorithms for SAR Imagery , 2005 .

[4]  G. Lewicki,et al.  Approximation by Superpositions of a Sigmoidal Function , 2003 .

[5]  Kurt Hornik,et al.  Multilayer feedforward networks are universal approximators , 1989, Neural Networks.

[6]  Minh N. Do,et al.  SAR Image Autofocus By Sharpness Optimization: A Theoretical Study , 2007, IEEE Transactions on Image Processing.

[7]  Heinz H. Bauschke,et al.  Fixed-Point Algorithms for Inverse Problems in Science and Engineering , 2011, Springer Optimization and Its Applications.

[8]  Müjdat Çetin,et al.  A Sparsity-Driven Approach for Joint SAR Imaging and Phase Error Correction , 2012, IEEE Transactions on Image Processing.

[9]  Geoffrey E. Hinton,et al.  Deep Learning , 2015, Nature.

[10]  Birsen Yazici,et al.  Passive Synthetic Aperture Hitchhiker Imaging of Ground Moving Targets—Part 2: Performance Analysis , 2014, IEEE Transactions on Image Processing.

[11]  Philip Schniter,et al.  Onsager-corrected deep learning for sparse linear inverse problems , 2016, 2016 IEEE Global Conference on Signal and Information Processing (GlobalSIP).

[12]  Birsen Yazici,et al.  Bistatic Synthetic Aperture Radar Imaging of Moving Targets Using Ultra-Narrowband Continuous Waveforms , 2014, SIAM J. Imaging Sci..

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

[14]  Yann LeCun,et al.  Discriminative Recurrent Sparse Auto-Encoders , 2013, ICLR.

[15]  Kurt Hornik,et al.  Universal approximation of an unknown mapping and its derivatives using multilayer feedforward networks , 1990, Neural Networks.

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

[17]  Birsen Yazici,et al.  Ground Moving Target Imaging Using Ultranarrowband Continuous Wave Synthetic Aperture Radar , 2013, IEEE Transactions on Geoscience and Remote Sensing.

[18]  Yoshua Bengio,et al.  Scaling learning algorithms towards AI , 2007 .

[19]  B. Yazıcı,et al.  Passive imaging of moving targets exploiting multiple scattering using sparse distributed apertures , 2012 .

[20]  Razvan Pascanu,et al.  On the difficulty of training recurrent neural networks , 2012, ICML.

[21]  Charles V. Jakowatz,et al.  Phase gradient autofocus-a robust tool for high resolution SAR phase correction , 1994 .

[22]  Birsen Yazici,et al.  Antenna motion errors in bistatic SAR imagery , 2015 .

[23]  Gregory F. Cooper,et al.  The Computational Complexity of Probabilistic Inference Using Bayesian Belief Networks , 1990, Artif. Intell..

[24]  Patrick L. Combettes,et al.  Proximal Splitting Methods in Signal Processing , 2009, Fixed-Point Algorithms for Inverse Problems in Science and Engineering.

[25]  Can Evren Yarman,et al.  Doppler synthetic aperture hitchhiker imaging , 2008, 2008 IEEE Radar Conference.

[26]  B. Yazıcı,et al.  Passive imaging using distributed apertures in multiple-scattering environments , 2010 .

[27]  Yoshua. Bengio,et al.  Learning Deep Architectures for AI , 2007, Found. Trends Mach. Learn..

[28]  Jason Weston,et al.  Joint Learning of Words and Meaning Representations for Open-Text Semantic Parsing , 2012, AISTATS.

[29]  Can Evren Yarman,et al.  Synthetic Aperture Hitchhiker Imaging , 2008, IEEE Transactions on Image Processing.

[30]  Marc'Aurelio Ranzato,et al.  Fast Inference in Sparse Coding Algorithms with Applications to Object Recognition , 2010, ArXiv.

[31]  M. Cheney,et al.  Synthetic aperture inversion , 2002 .

[32]  Yann LeCun,et al.  Learning Fast Approximations of Sparse Coding , 2010, ICML.

[33]  Birsen Yazici,et al.  Passive Synthetic Aperture Hitchhiker Imaging of Ground Moving Targets—Part 1: Image Formation and Velocity Estimation , 2014, IEEE Transactions on Image Processing.

[34]  Can Evren Yarman,et al.  Multistatic Synthetic Aperture Radar Image Formation , 2009, IEEE Transactions on Image Processing.

[35]  Surya Ganguli,et al.  Identifying and attacking the saddle point problem in high-dimensional non-convex optimization , 2014, NIPS.

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

[37]  Xiaoling Zhang,et al.  An autofocus approach for model error correction in compressed sensing SAR imaging , 2012, 2012 IEEE International Geoscience and Remote Sensing Symposium.

[38]  Birsen Yazici,et al.  Bistatic Doppler-SAR DPCA imaging of ground moving targets , 2014, 2014 IEEE Radar Conference.

[39]  Lu Wang,et al.  Structured sparsity-driven autofocus algorithm for high-resolution radar imagery , 2016, Signal Process..