Advanced Techniques for Robust SAR ATR: Mitigating Noise and Phase Errors

We present advanced Deep Learning (DL) techniques for robust Synthetic Aperture Radar (SAR) automatic target recognition (ATR) in the presence of noise and signal phase errors. Our research focuses on ensuring robust performance of SAR ATR algorithms under noise and adversarial attacks. Robust DL-based SAR ATR is paramount in operational scenarios such as disaster relief, search and rescue, and highly accurate object classification for autonomous vehicles. Our contributions, as described in this paper, include algorithm development and implementation of an advanced deep learning technique known as adversarial training (AT) to mitigate the detrimental effects of sophisticated noise and phase errors. Our research demonstrated that 1) AT improves performance under extended operating conditions, in some cases improving up to 10% over models without AT. 2) The use of AT improves performance when sinusoidal or wideband phase noise is present, in some cases gaining 40% in accuracy that would be lost in the presence of noise. 3) We find the model architecture has significant impact on robustness, with more complex networks showing a greater improvement from AT. 4) The availability of multi-polarization data is always advantageous. To our knowledge no one has provided an extensive analysis of the impact of adversarial machine learning (ML) on SAR image classification. Thus, this paper serves as a comprehensive research revealing the impact of adversarial attack and how to mitigate it.

[1]  A.K. Mishra,et al.  Validation of PCA and LDA for SAR ATR , 2008, TENCON 2008 - 2008 IEEE Region 10 Conference.

[2]  Pei Huang,et al.  A robust decision fusion strategy for SAR target recognition , 2018 .

[3]  Simon A. Wagner,et al.  SAR ATR by a combination of convolutional neural network and support vector machines , 2016, IEEE Transactions on Aerospace and Electronic Systems.

[4]  Hidetoshi Furukawa,et al.  Deep Learning for Target Classification from SAR Imagery: Data Augmentation and Translation Invariance , 2017, ArXiv.

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

[6]  Michael Lee Bryant,et al.  Standard SAR ATR evaluation experiments using the MSTAR public release data set , 1998, Defense, Security, and Sensing.

[7]  Dan Boneh,et al.  The Space of Transferable Adversarial Examples , 2017, ArXiv.

[8]  Odysseas Kechagias-Stamatis,et al.  Target recognition for synthetic aperture radar imagery based on convolutional neural network feature fusion , 2018, Journal of Applied Remote Sensing.

[9]  R. Huan,et al.  Decision fusion strategies for SAR image target recognition , 2011 .

[10]  Lee C. Potter,et al.  Attributed scattering centers for SAR ATR , 1997, IEEE Trans. Image Process..

[11]  Gongjian Wen,et al.  Combination of global and local filters for robust SAR target recognition under various extended operating conditions , 2019, Inf. Sci..

[12]  Baiyuan Ding,et al.  An Efficient and Robust Framework for SAR Target Recognition by Hierarchically Fusing Global and Local Features , 2018, IEEE Transactions on Image Processing.

[13]  Joan Bruna,et al.  Intriguing properties of neural networks , 2013, ICLR.

[14]  Hongwei Liu,et al.  Attributed Scattering Center Extraction Algorithm Based on Sparse Representation With Dictionary Refinement , 2017, IEEE Transactions on Antennas and Propagation.

[15]  Xiangyu Zhang,et al.  ShuffleNet: An Extremely Efficient Convolutional Neural Network for Mobile Devices , 2017, 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition.

[16]  Chan Gook Park,et al.  Additional feature CNN based automatic target recognition in SAR image , 2017, 2017 Fourth Asian Conference on Defence Technology - Japan (ACDT).

[17]  Lee C. Potter,et al.  Civilian vehicle radar data domes , 2010, Defense + Commercial Sensing.

[18]  Yao Zhao,et al.  Adversarial Attacks and Defences Competition , 2018, ArXiv.

[19]  Armin W. Doerry,et al.  Anatomy of a SAR impulse response. , 2007 .

[20]  Hai Li,et al.  Feature Space Perturbations Yield More Transferable Adversarial Examples , 2019, 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).

[21]  S. Quegan,et al.  Understanding Synthetic Aperture Radar Images , 1998 .

[22]  Yue Liu,et al.  Data Augmentation Based on Attributed Scattering Centers to Train Robust CNN for SAR ATR , 2019, IEEE Access.

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

[24]  David A. Wagner,et al.  Towards Evaluating the Robustness of Neural Networks , 2016, 2017 IEEE Symposium on Security and Privacy (SP).

[25]  Aleksander Madry,et al.  Learning Perceptually-Aligned Representations via Adversarial Robustness , 2019, ArXiv.

[26]  Haipeng Wang,et al.  Target Classification Using the Deep Convolutional Networks for SAR Images , 2016, IEEE Transactions on Geoscience and Remote Sensing.

[27]  Aleksander Madry,et al.  Robustness May Be at Odds with Accuracy , 2018, ICLR.

[28]  Jihong Fan,et al.  Target Reconstruction Based on Attributed Scattering Centers with Application to Robust SAR ATR , 2018, Remote. Sens..

[29]  Aleksander Madry,et al.  Towards Deep Learning Models Resistant to Adversarial Attacks , 2017, ICLR.

[30]  Jonathon Shlens,et al.  Explaining and Harnessing Adversarial Examples , 2014, ICLR.