research-article

Synthetic Images Augmentation for Robust SAR Target Recognition

Authors:

Zhe LiuAuthors Info & Claims

ICVIP '21: Proceedings of the 2021 5th International Conference on Video and Image Processing

Pages 19 - 25

https://doi.org/10.1145/3511176.3511180

Published: 12 March 2022 Publication History

Abstract

Synthetic aperture radar (SAR) image simulation can generate synthetic SAR images as complements of measured SAR images in SAR automatic target recognition (ATR) problem. However, most deep learning-based SAR ATR models can hardly perform well when trained on fully synthetic images, due to the infidelity clutter in synthetic images and the noise sensitivity of deep neural network. Therefore, to train clutter robust SAR ATR models, we propose an augmentation method based on clutter reconstruction. Our augmentation is modeled from the perspective of signal and noise. The speckle noise is reconstructed based on the homomorphic transform of the multiplicative signal model. We also consider the spatial correlation of noise by using adjustable convolution kernels. Meanwhile, a power-laws transform is applied to reconstruct background reflectivity based on signal-to-clutter ratio (SCR) measurement. Based on the SAMPLE dataset, experimental results on cross-class classification show that typical deep learning ATR models trained by our method can achieve over 95% accuracy. It improves more than 10% compared to the baseline. Tests on clutter corrupted images indicating our method is effective and generic.

References

[1]

Sergey Abramov, Victoriya Abramova, Vladimir Lukin, Nikolay Ponomarenko, Benoit Vozel, Kacem Chehdi, Karen Egiazarian, and Jaakko Astola. 2014. Methods for Blind Estimation of Speckle Variance in SAR Images: Simulation Results and Verification for Real-Life Data. (2014).

[2]

H. H. Arsenault and G. April. 1976. Properties of speckle integrated with a finite aperture and logarithmically transformed. Journal of the Optical Society of America 66, 11 (1976), 1160–1163.

[3]

S. Auer, S. Hinz, and R. Bamler. 2010. Ray-Tracing Simulation Techniques for Understanding High-Resolution SAR Images. IEEE Transactions on Geoscience and Remote Sensing 48, 3 (2010), 1445–1456.

[4]

Carole Belloni, Alessio Balleri, Nabil Aouf, Jean-Marc Le Caillec, and Thomas Merlet. 2021. Explainability of Deep SAR ATR Through Feature Analysis. IEEE Trans. Aerospace Electron. Systems 57, 1 (2021), 659–673.

[5]

Miriam Cha, Arjun Majumdar, H.T. Kung, and Jarred Barber. 2018. Improving Sar Automatic Target Recognition Using Simulated Images Under Deep Residual Refinements. In 2018 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). 2606–2610.

[6]

Sizhe Chen, Haipeng Wang, Feng Xu, and Ya-Qiu Jin. 2016. Target Classification Using the Deep Convolutional Networks for SAR Images. IEEE Transactions on Geoscience and Remote Sensing 54, 8 (2016), 4806–4817.

[7]

Shiyong Cui, Gottfried Schwarz, and Mihai Datcu. 2016. A Benchmark Evaluation of Similarity Measures for Multitemporal SAR Image Change Detection. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing 9, 3 (2016), 1101–1118.

[8]

Zongyong Cui, Mingrui Zhang, Zongjie Cao, and Changjie Cao. 2019. Image Data Augmentation for SAR Sensor via Generative Adversarial Nets. IEEE Access 7(2019), 42255–42268.

[9]

Jun Ding, Bo Chen, Hongwei Liu, and Mengyuan Huang. 2016. Convolutional Neural Network With Data Augmentation for SAR Target Recognition. IEEE Geoscience and Remote Sensing Letters 13, 3 (2016), 364–368.

[10]

Shaoyan Du, Jun Hong, Yu Wang, Kaichu Xing, and Tian Qiu. 2021. Physical-Related Feature Extraction From Simulated SAR Image Based on the Adversarial Encoding Network for Data Augmentation. IEEE Geoscience and Remote Sensing Letters(2021), 1–5.

[11]

Maoguo Gong, Meng Jia, Linzhi Su, Shuang Wang, and Licheng Jiao. 2014. Detecting changes of the Yellow River Estuary via SAR images based on a local fit-search model and kernel-induced graph cuts. International Journal of Remote Sensing 35 (2014), 4009–4030.

[12]

Jiayi Guo, Bin Lei, Chibiao Ding, and Yueting Zhang. 2017. Synthetic Aperture Radar Image Synthesis by Using Generative Adversarial Nets. IEEE Geoscience and Remote Sensing Letters 14, 7 (2017), 1111–1115.

[13]

Kaiming He, Xiangyu Zhang, Shaoqing Ren, and Jian Sun. 2016. Deep Residual Learning for Image Recognition. In 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). 770–778.

[14]

Nathan Inkawhich, Matthew J. Inkawhich, Eric K. Davis, Uttam K. Majumder, Erin Tripp, Chris Capraro, and Yiran Chen. 2021. Bridging a Gap in SAR-ATR: Training on Fully Synthetic and Testing on Measured Data. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing 14(2021), 2942–2955. https://doi.org/10.1109/JSTARS.2021.3059991

[15]

Nathan A. Inkawhich, Eric K. Davis, Matthew J. Inkawhich, Uttam K. Majumder, and Yiran Chen. 2021. Training SAR-ATR Models for Reliable Operation in Open-World Environments. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing 14(2021), 3954–3966. https://doi.org/10.1109/JSTARS.2021.3068944

[16]

Diederik P. Kingma and Jimmy Ba. 2014. Adam: A Method for Stochastic Optimization. arXiv preprint arXiv:1412.6980(2014).

[17]

Youngchul Kwak, Woo-Jin Song, and Seong-Eun Kim. 2019. Speckle-Noise-Invariant Convolutional Neural Network for SAR Target Recognition. IEEE Geoscience and Remote Sensing Letters 16, 4 (2019), 549–553.

[18]

Benjamin Lewis, Theresa Scarnati, and Ernest Parke. 2020. Investigating the saliency of SAR image chips. In Algorithms for Synthetic Aperture Radar Imagery XXVII, Edmund Zelnio and Frederick D. Garber (Eds.). Vol. 11393. International Society for Optics and Photonics, SPIE, 83 – 95. https://doi.org/10.1117/12.2558364

[19]

Benjamin Lewis, Theresa Scarnati, Elizabeth Sudkamp, John Nehrbass, Stephen Rosencrantz, and Edmund Zelnio. 2019. A SAR dataset for ATR development: the Synthetic and Measured Paired Labeled Experiment (SAMPLE). In Algorithms for Synthetic Aperture Radar Imagery XXVI, Vol. 10987.

[20]

David Malmgren-Hansen, Anders Kusk, Jorgen Dall, Allan Aasbjerg Nielsen, Rasmus Engholm, and Henning Skriver. 2017. Improving SAR Automatic Target Recognition Models With Transfer Learning From Simulated Data. IEEE Geoscience and Remote Sensing Letters 14, 9 (2017), 1484–1488.

[21]

D. Malmgren-Hansen and M. Nobel-Jorgensen. 2015. Convolutional neural networks for SAR image segmentation. In 2015 IEEE International Symposium on Signal Processing and Information Technology (ISSPIT).

[22]

Shengren Niu, Xiaolan Qiu, Bin Lei, Chibiao Ding, and Kun Fu. 2020. Parameter Extraction Based on Deep Neural Network for SAR Target Simulation. IEEE Transactions on Geoscience and Remote Sensing 58, 7 (2020), 4901–4914.

[23]

Shengren Niu, Xiaolan Qiu, Bin Lei, and Kun Fu. 2021. A SAR Target Image Simulation Method With DNN Embedded to Calculate Electromagnetic Reflection. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing 14(2021), 2593–2610.

[24]

Joseph O’Sullivan. 2002. Performance Complexity Study of Several Approaches to Automatic Target Recognition from SAR Images. IEEE Trans. Aerospace Electron. Systems 38, 2 (2002), 632–648.

[25]

R. Schumacher and J. Schiller. 2005. Non-cooperative target identification of battlefield targets - classification results based on SAR images. In IEEE International Radar Conference, 2005.167–172.

[26]

Spencer R. Sellers, Peter J. Collins, and Julie Ann Jackson. 2020. Augmenting Simulations for SAR ATR Neural Network Training. In 2020 IEEE International Radar Conference (RADAR). 309–314.

[27]

Karen Simonyan and Andrew Zisserman. 2014. Very Deep Convolutional Networks for Large-Scale Image Recognition. In arXiv: Computer Vision and Pattern Recognition.

[28]

Nitish Srivastava, Geoffrey Hinton, Alex Krizhevsky, Ilya Sutskever, and Ruslan Salakhutdinov. 2014. Dropout: a simple way to prevent neural networks from overfitting. Journal of Machine Learning Research 15, 1 (2014), 1929–1958.

Digital Library

Cited By

Do JYoo MLee JMoon HKim S(2024)Semi-Supervised SAR Image Classification via Adaptive Threshold SelectionJournal of the Korea Institute of Military Science and Technology10.9766/KIMST.2024.27.3.31927:3(319-328)Online publication date: 5-Jun-2024
https://doi.org/10.9766/KIMST.2024.27.3.319
Han FDong HSi LZhang L(2024)Improving SAR Automatic Target Recognition via Trusted Knowledge Distillation From Simulated DataIEEE Transactions on Geoscience and Remote Sensing10.1109/TGRS.2024.336047062(1-14)Online publication date: 2024
https://doi.org/10.1109/TGRS.2024.3360470
Li JYu ZYu LCheng PChen JChi C(2023)A Comprehensive Survey on SAR ATR in Deep-Learning EraRemote Sensing10.3390/rs1505145415:5(1454)Online publication date: 5-Mar-2023
https://doi.org/10.3390/rs15051454
Show More Cited By

Recommendations

SAR automatic target recognition based on K-means and data augmentation
ICIIP '16: Proceedings of the 1st International Conference on Intelligent Information Processing

Synthetic aperture radar (SAR) automatic target recognition (ATR) has been receiving more and more attention in the past two decades. A lot of methods have been proposed and studied for radar target recognition. Among some of these methods, they use the ...
Robust moving targets detection and velocity estimation using multi-channel and multi-look SAR images

For the sake of the robustness to ground moving targets indication (GMTI) for synthetic aperture radar (SAR) in heterogeneous environment, a new approach using multi-channel and multi-look SAR images is presented. This approach has been shown to attain ...
Spatially Variant Apodization for Squinted Synthetic Aperture Radar Images

Spatially variant apodization (SVA) is a nonlinear sidelobe reduction technique that improves sidelobe level and preserves resolution at the same time. This method implements a bidimensional finite impulse response filter with adaptive taps depending on ...

Comments

Information & Contributors

Information

Published In

cover image ACM Other conferences

ICVIP '21: Proceedings of the 2021 5th International Conference on Video and Image Processing

December 2021

219 pages

ISBN:9781450385893

DOI:10.1145/3511176

Copyright © 2021 ACM.

Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. Copyrights for components of this work owned by others than ACM must be honored. Abstracting with credit is permitted. To copy otherwise, or republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee. Request permissions from [email protected]

Publisher

Association for Computing Machinery

New York, NY, United States

Publication History

Published: 12 March 2022

Permissions

Request permissions for this article.

Request Permissions

Check for updates

Author Tags

Qualifiers

Research-article
Research
Refereed limited

Conference

ICVIP 2021

ICVIP 2021: 2021 The 5th International Conference on Video and Image Processing

December 22 - 25, 2021

CA, Hayward, USA

Contributors

Other Metrics

View Article Metrics

Bibliometrics & Citations

Bibliometrics

Article Metrics

4
Total Citations
View Citations
112
Total Downloads

Downloads (Last 12 months)33
Downloads (Last 6 weeks)2

Reflects downloads up to 06 Nov 2024

Other Metrics

View Author Metrics

Citations

Cited By

Do JYoo MLee JMoon HKim S(2024)Semi-Supervised SAR Image Classification via Adaptive Threshold SelectionJournal of the Korea Institute of Military Science and Technology10.9766/KIMST.2024.27.3.31927:3(319-328)Online publication date: 5-Jun-2024
https://doi.org/10.9766/KIMST.2024.27.3.319
Han FDong HSi LZhang L(2024)Improving SAR Automatic Target Recognition via Trusted Knowledge Distillation From Simulated DataIEEE Transactions on Geoscience and Remote Sensing10.1109/TGRS.2024.336047062(1-14)Online publication date: 2024
https://doi.org/10.1109/TGRS.2024.3360470
Li JYu ZYu LCheng PChen JChi C(2023)A Comprehensive Survey on SAR ATR in Deep-Learning EraRemote Sensing10.3390/rs1505145415:5(1454)Online publication date: 5-Mar-2023
https://doi.org/10.3390/rs15051454
Sun YWang YLiu HHu LZhang CWang S(2023)Gradual Domain Adaptation with Pseudo-Label Denoising for SAR Target Recognition When Using Only Synthetic Data for TrainingRemote Sensing10.3390/rs1503070815:3(708)Online publication date: 25-Jan-2023
https://doi.org/10.3390/rs15030708

View Options

Get Access

Login options

Check if you have access through your login credentials or your institution to get full access on this article.

Full Access

Get this Publication

View options

PDF

View or Download as a PDF file.

eReader

View online with eReader.

HTML Format

View this article in HTML Format.

Media

Figures

Other

Tables

View Table of Contents