volume | PIER Journals

Abstract

Convolutional Neural Network (CNN) models applied to synthetic aperture radar automatic target recognition (SAR ATR) universally focus on two important issues: overfitting caused by lack of sufficient training data and independent variations like worse estimates of the aspect angle etc. To this end, we developed a lightweight CNN-based method named SARNet to accomplish the classification task. Firstly, a clock-wise data amplification approach is presented to generate adequate SAR images without requiring many raw SAR images, effectively avoiding overfitting in the course of training. Then a SARNet is devised to process the extracted features from SAR target images and work on classification tasks with parameters fine-tuning under comparative models. To enhance and structurally organize the representation of learned proposed model, various activation functions are explored in this paper. Furthermore, due to the pioneering conducted experiments, training samples in the MSTAR and extended MSTAR database are utilized to demonstrate the robustness and effectiveness of the lightweight model. Experimental results have shown that our proposed model has achieved a 98.30% state-of-the-art accuracy.

1. Liu, S., M. Liu, P. Li, et al. "SAR image denoising via sparse representation in shearlet domain based on continuous cycle spinning," IEEE TRGS, Vol. 55, No. 5, 2985-2992, 2017. Google Scholar

2. Cloude, S. R. and E. Pottier, "An entropy based classification scheme for land applications of polarimetric SAR," IEEE TRGS, Vol. 35, No. 1, 68-78, 1997. Google Scholar

3. Singh, M. and G. Kaur, "SAR image classification using PCA and texture analysis," Information Technology and Mobile Communication, 435-439, Springer, Berlin, Heidelberg, 2011. Google Scholar

4. Mishra, A. K., "Validation of PCA and LDA for SAR ATR," TENCON 2008 — 2008 IEEE Region 10 Conference, 1-6, 2008. Google Scholar

5. Li, Q., G. Qu, and Z. Li, "Matching between SAR images and optical images based on HOG descriptor," International Radar Conference IET, 1-4, 2013. Google Scholar

6. Huan, R. H., Y. Pan, and K. J. Mao, "SAR image target recognition based on NMF feature extraction and Bayesian decision fusion," IITA-GRS, 496-499, 2010. Google Scholar

7. Cao, Z. J., Y. C. Ge, and J. L. Feng, "SAR image classification with a sample reusable domain adaptation algorithm based on SVM classifier," Pattern Recognition, 2017. Google Scholar

8. Khosravi, I., A. Safari, S. Homayouni, et al. "Enhanced decision tree ensembles for land-cover mapping from fully polarimetric SAR data," IJRS, Vol. 38, No. 23, 7138-7160, 2017. Google Scholar

9. Xu, G., M. Xing, L. Zhang, et al. "Bayesian inverse synthetic aperture radar imaging," IEEE GRSL, Vol. 8, No. 6, 1150-1154, 2011. Google Scholar

10. Huo, W., Y. Huang, J. Pei, et al. "Virtual SAR target image generation and similarity," 2016 IEEE International Geoscience and Remote Sensing Symposium (IGARSS), 914-917, IEEE, 2016. Google Scholar

11. Zhao, Q. and J. C. Principe, "Support vector machine for SAR automatic target recognition,", Vol. 37, No. 2, 643-654, 2001. Google Scholar

12. Sun, Y., Z. Liu, and J. Li, "Adaptive boosting for SAR automatic target recognition," IEEE TGRS, Vol. 43, No. 1, 112-125, 2007. Google Scholar

13. Zhou, J. X., Z. G. Shi, X. Cheng, and Q. Fu, "Automatic target recognition of SAR images based on global scattering center model," IEEE TGRS, Vol. 49, No. 10, 3713-3729, 2011. Google Scholar

14. Park, J., S. H. Park, and K. T. Kim, "New discrimination features for SAR automatic target recognition," IEEE GRSL, Vol. 10, No. 3, 476-480, 2013. Google Scholar

15. Dong, G., N. Wang, and G. Kuang, "Sparse representation of monogenic signal: With application to target recognition in SAR images," IEEE GRSL, Vol. 21, No. 8, 952-956, 2014. Google Scholar

16. Clemente, C., et al. "Pseudo-Zernike based multi-pass automatic target recognition from multichannel SAR," IET RSN, Vol. 9, No. 4, 457-466, 2015. Google Scholar

17. Mishra, A. K. and B. Mulgrew, "Bistatic SAR ATR using PCA-based features," Automatic Target Recognition XVI, Vol. 6234, International Society for Optics and Photonics, 2006. Google Scholar

18. Ash, J. N., "Joint imaging and change detection for robust exploitation in interrupted SAR environments," Algorithms for Synthetic Aperture Radar Imagery XX, Vol. 8746, 87460J, International Society for Optics and Photonics, 2013. Google Scholar

19. Zhang, Y. D., L. Wu, and G. Wei, "A new classifier for polarimetric SAR images," Progress In Electromagnetics Research, Vol. 94, 83-104, 2009. Google Scholar

20. Zhai, Y., J. Li, J. Gan, and Z. Ying, "A multi-scale local phase quantization plus biomimetic pattern recognition method for SAR automatic target recognition," Progress In Electromagnetics Research, Vol. 135, 105-122, 2013. Google Scholar

21. Mishra, B. and J. Susaki, "Coupling of thresholding and region growing algorithm for change detection in SAR images," Progress In Electromagnetics Research, Vol. 143, 519-544, 2013. Google Scholar

22. Gao, G., X. Qin, and S. Zhou, "Modeling SAR images based on a generalized gamma distribution for texture component," Progress In Electromagnetics Research, Vol. 137, 669-685, 2013. Google Scholar

23. Cheng, J., G. Gao, W. Ding, X. Ku, and J. Sun, "An improved scheme for parameter estimation of G◦ distribution model in high-resolution SAR images," Progress In Electromagnetics Research, Vol. 134, 23-46, 2013. Google Scholar

24. Ni, W. P., W. D. Yan, J. Z. Wu, et al. "Moment feature analysis and multi-threshold segmentation of MSTAR image," JOIG, Vol. 18, No. 10, 2018. Google Scholar

25. Fu, F. C., "SAR target recognition method based on target region matching," EO&C, Vol. 4, 2018. Google Scholar

26. Ding, J., B. Chen, H. Liu, et al. "Convolutional neural network with data amplification for SAR target recognition," IEEE, Vol. 13, No. 3, 364-368, 2016. Google Scholar

27. Chen, S., H. Wang, F. Xu, et al. "Target classification using the deep convolutional networks for SAR images," IEEE TGRS, Vol. 54, No. 8, 4806-4817, 2016. Google Scholar

28. Zhao, W. and S. Du, "Spectral-spatial feature extraction for hyperspectral image classification: A dimension reduction and deep learning approach," IEEE TGRS, Vol. 54, No. 8, 4544-4554, 2016. Google Scholar

29. Marmanis, D., M. Datcu, T. Esch, et al. "Deep learning earth observation classification using ImageNet pertained networks," IEEE GRSL, Vol. 13, No. 1, 105-109, 2016. Google Scholar

30. AbdAlmageed, W., Y. Wu, S. Rawls, et al. "Face recognition using deep multi-pose representations," 2016 IEEE Winter Conference on Applications of Computer Vision (WACV), 1-9, IEEE, 2016. Google Scholar

31. Morgan, D. A. E., "Deep convolutional neural networks for ATR from SAR imagery," Algorithms for Synthetic Aperture Radar Imagery XXII, Vol. 9475, 94750F, International Society for Optics and Photonics, 2015. Google Scholar

32. Profeta, A., A. Rodriguez, and H. S. Clouse, "Convolutional neural networks for synthetic aperture radar classification," Algorithms for Synthetic Aperture Radar Imagery XXIII, 9843–98430M, International Society for Optics and Photonics, 2016. Google Scholar

33. Wilmanski, M., C. Kreucher, and J. Lauer, "Modern approaches in deep learning for SAR ATR," Algorithms for Synthetic Aperture Radar Imagery XXIII, International Society for Optics and Photonics, 9843–98430N, 2016. Google Scholar

34. Ødegaard, N., A. O. Knapskog, C. Cochin, et al. "Classification of ships using real and simulated data in a convolutional neural network," 2016 IEEE Radar Conference, 1-6, IEEE, 2016. Google Scholar

35. Liu, C., C. W. Qu, et al. "Target classification of SAR images based on convolution neural network migration learning," Modern Radar, Vol. 3, 2018. Google Scholar

36. Krizhevsky, A., I. Sutskever, and G. E. Hinton, "ImageNet classification with deep convolutional neural networks," International Conference on Neural Information Processing Systems, Curran Associates Inc., 2012. Google Scholar

37. Lecun, Y. L., et al. "Gradient-based learning applied to document recognition," Proceedings of the IEEE, Vol. 86, No. 11, 2278-2324, 1998. Google Scholar

38. He, K. M., et al. "Deep residual learning for image recognition," 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2016. Google Scholar

39. Qi, H. J., Y. H. Wang, J. Ding, et al. "SAR target recognition based on Multi-information dictionary learning and sparse representation," J. Syst. Eng. Electron., Vol. 37, No. 6, 1280-1287, 2015. Google Scholar

40. Wu, T., et al. "Study on SAR target recognition based on Support Vector Machine," IEEE Conference on Synthetic Aperture Radar, Asian-Pacific, 856-859, 2010. Google Scholar

41. Zhai, Y. K., J. Li, J. Y. Gan, et al. "A multi-scale local phase quantization plus biomimetic pattern recognition method for SAR automatic target recognition," Progress In Electromagnetics Research, Vol. 135, No. 1, 105-122, 2013. Google Scholar

42. Wang, L., F. Zhang, W. Li, et al. "A method of SAR target recognition based on gabor filter and local texture feature extraction," JOR, Vol. 4, No. 6, 658-665, 2015. Google Scholar

43. Zhang, H., N. M. Nasrabadi, Y. Zhang, et al. "Multi-view automatic target recognition using joint sparse representation," IEEE Transactions on Aerospace and Electronic Systems, Vol. 48, No. 3, 2481-2497, 2012. Google Scholar

44. Tian, Z. Z., R. H. Zhan, J. M. Hu, et al. "SAR ATR based on convolutional neural network," Journal of Radars, Vol. 5, No. 3, 320-325, 2016. Google Scholar

45. Sun, Y., Z. Liu, S. Todorovic, et al. "Adaptive boosting for SAR automatic target recognition," IEEE Transactions on Aerospace and Electronic Systems, Vol. 43, No. 1, 112-125, 2007. Google Scholar

46. Liu, K. P., Z. L. Ying, and Y. K. Zhai, "SAR image target recognition based on unsupervised k-means feature and data amplification," JOSP, Vol. 33, No. 3, 456-458, 2017 (in Chinese). Google Scholar

Dr. Yikui Zhai

Dr. Wenbo Deng

Dr. Yanqing Zhu

Dr. Ying Xu

Dr. Bing Sun

Dr. Jingwen Li

Dr. Qirui Ke

Dr. Yihang Zhi

Dr. Vincenzo Pirui