http://iet.metastore.ingenta.com
1887

SAR object classification using the DAE with a modified triplet restriction

SAR object classification using the DAE with a modified triplet restriction

For access to this article, please select a purchase option:

Buy article PDF
$19.95
(plus tax if applicable)
Buy Knowledge Pack
10 articles for $120.00
(plus taxes if applicable)

IET members benefit from discounts to all IET publications and free access to E&T Magazine. If you are an IET member, log in to your account and the discounts will automatically be applied.

Learn more about IET membership 

Recommend Title Publication to library

You must fill out fields marked with: *

Librarian details
Name:*
Email:*
Your details
Name:*
Email:*
Department:*
Why are you recommending this title?
Select reason:
 
 
 
 
 
IET Radar, Sonar & Navigation — Recommend this title to your library

Thank you

Your recommendation has been sent to your librarian.

Although deep learning methods have made great progress in synthetic aperture radar (SAR)-based remote sensing, lack of training data has often been the major obstacle while they are adopted for SAR automatic target recognition. In this study, a new deep network in the form of a restricted three-branch denoising auto-encoder (DAE) is proposed to take the full advantage of limited training samples. In this model, a modified triplet restriction that combines the semi-hard triplet loss with the intra-class distance penalty is devised to learn discriminative features with a small intra-class divergence and a large inter-class divergence. Besides, the reconstruction distortion is measured between the model outputs and the images filtered by the improved Lee Sigma filter rather than the original inputs to suppress clutter in the background. Furthermore, a batch-based triplet loss, which calculates the modified triplet loss in a batch-based manner, is proposed to tackle the difficulties in implementation and reduce its computation complexity. The simplified version of the three-branch Triplet-DAE is subsequently devised as a one-branch DAE restricted by the batch-based triplet loss. Experimental results with the MSTAR data demonstrate the effectiveness of the proposed method on real SAR images.

References

    1. 1)
      • 1. Li, Y., Zhou, C., Wang, N.: ‘A survey on feature extraction of SAR images’. Proc. Int. Conf. Computer Application and System Modelling, Taiyuan, China, October 2010, pp. V1-312V1-317.
    2. 2)
      • 2. Zhai, Y., Li, J., Guo, C., et al: ‘SAR automatic target recognition based on local phase quantization plus biomimetic pattern recognition’. Proc. Int. Conf. Signal Processing, Beijing, China, October 2012, pp. 18851889.
    3. 3)
      • 3. Cui, Z., Cao, Z., Yang, J., et al: ‘Target recognition in synthetic aperture radar images via non-negative matrix factorisation’, IET Radar Sonar Navig., 2015, 9, (9), pp. 13761385.
    4. 4)
      • 4. Mishra, A.K., Motaung, T.: ‘Application of linear and nonlinear PCA to SAR ATR’. Proc. Int. Conf. Radioelektronika, Pardubice, Czech Republic, April 2015, pp. 349354.
    5. 5)
      • 5. Yin, K., Lin, J., Zhang, C., et al: ‘A method for automatic target recognition using shadow contour of SAR image’, IETE Technical Review, 2013, 30, (4), pp. 313323.
    6. 6)
      • 6. Dungan, K.E.: ‘Feature-based vehicle classification in wide-angle synthetic aperture radar’. PhD thesis, The Ohio State University, 2010.
    7. 7)
      • 7. Zhang, H., Nasrabadi, N., Zhang, Y., et al: ‘Multi-view automatic target recognition using joint sparse representation’, IEEE Trans. Aerosp. Electron. Syst., 2012, 48, (3), pp. 24812497.
    8. 8)
      • 8. Ding, B., Wen, G., Huang, X., et al: ‘Target recognition in synthetic aperture radar images via matching of attributed scattering centres’, IEEE J. Sel. Top. Appl. Earth Observ. Remote Sens., 2017, 10, (7), pp. 33343347.
    9. 9)
      • 9. Ding, B., Wen, G.: ‘Target reconstruction based on 3-d scattering centre model for robust SAR ATR’, IEEE Trans. Geosci. Remote Sens., 2018, 56, (7), pp. 37723785.
    10. 10)
      • 10. Xie, H., Wang, S., Liu, K., et al: ‘Multilayer feature learning for polarimetric synthetic radar data classification’. Proc. Int. Geoscience Remote Sensing Symp., Quebec City, QC, Canada, July 2014, pp. 28182821.
    11. 11)
      • 11. Ni, J., Xu, Y.: ‘SAR automatic target recognition based on a visual cortical system’. Proc. Int. Congress on Image and Signal Processing, Hangzhou, China, December 2013, pp. 778782.
    12. 12)
      • 12. Cui, Z., Cao, Z., Yang, J., et al: ‘Hierarchical recognition system for target recognition from sparse representations’, Math. Probl. Eng., 2015, 2015, (527095), pp. 16.
    13. 13)
      • 13. Chen, S., Wang, H.: ‘SAR target recognition based on deep learning’. Proc. Int. Conf. Data Science and Advanced Analytics, Shanghai, China, 30 October–1 November 2015, pp. 541547.
    14. 14)
      • 14. Morgan, D.: ‘Deep convolutional neural networks for ATR from SAR imagery’. Proc. SPIE Volume 9475, Algorithms for Synthetic Aperture Radar Imagery XXII, Baltimore, MD, USA, May 2015, pp. 94750F-194750F-13.
    15. 15)
      • 15. Kreucher, C.: ‘Modern approaches in deep learning for SAR ATR’. Proc. SPIE. 9843, Algorithms for Synthetic Aperture Radar Imagery XXIII, Baltimore, MD, USA, May 2016, pp. 98430N-9843198430N-98410.
    16. 16)
      • 16. He, H., Wang, S., Yang, D., et al: ‘SAR target recognition and unsupervised detection based on convolutional neural network’. Proc. Chinese Automation Congress (CAC), Jinan, China, October 2017, pp. 435438.
    17. 17)
      • 17. Li, Y., Wang, J., Xu, Y., et al: ‘DeepSAR-Net: deep convolutional neural networks for SAR target recognition’. Proc. Int. Conf. Big Data Analysis, Beijing, China, March 2017, pp. 740743.
    18. 18)
      • 18. Shao, J., Qu, C., Li, J.: ‘A performance analysis of convolutional neural network models in SAR target recognition’. Proc. SAR in Big Data Era: Models, Methods and Applications, Beijing, China, November 2017, pp. 16.
    19. 19)
      • 19. Zhao, J., Guo, W., Cui, S., et al: ‘Convolutional neural network for SAR image classification at patch level’. Proc. Int. Geoscience Remote Sensing Symp., Beijing, China, July 2016, pp. 945948.
    20. 20)
      • 20. Zhou, Y., Wang, H., Xu, F., et al: ‘Polarimetric SAR image classification using deep convolutional neural networks’, IEEE Geosci. Remote Sens. Lett., 2016, 13, (12), pp. 19351939.
    21. 21)
      • 21. Li, J., Wang, C., Wang, S., et al: ‘Classification of very high-resolution SAR image based on convolutional neural network’. Proc. Int. Workshop on Remote Sensing with Intelligent Processing, Shanghai, China, May 2017, pp. 14.
    22. 22)
      • 22. Malmgren-Hansen, D., Engholm, R., Pedersen, M.O.: ‘Training convolutional neural networks for translational invariance on SAR ATR’. Proc. European Conf. Synthetic Aperture Radar, Hamburg, Germany, June 2016, pp. 459462.
    23. 23)
      • 23. Andrew, P., Andres, R., Clouse, H.S.: ‘Convolutional neural networks for synthetic aperture radar classification’. Proc. SPIE 9843, Algorithms for Synthetic Aperture Radar Imagery XXIII, Baltimore, MD, USA, May 2016, pp. 98430M-198430M-10.
    24. 24)
      • 24. Kechagias-Stamatis, O., Aouf, N., Belloni, C.: ‘SAR automatic target recognition based on convolutional neural networks’. Proc. Int. Conf. Radar Systems, Belfast, UK, October 2017, pp. 14.
    25. 25)
      • 25. Wagner, S.: ‘Combination of convolutional feature extraction and support vector machines for radar ATR’. Proc. Int. Conf. Information Fusion, Salamanca, Spain, July 2014, pp. 16.
    26. 26)
      • 26. Geng, J., Fan, J., Wang, H., et al: ‘High-resolution SAR image classification via deep convolutional auto-encoders’, IEEE Geosci. Remote Sens. Lett., 2015, 12, (11), pp. 23512355.
    27. 27)
      • 27. Li, X., Li, C., Wang, P., et al: ‘SAR ATR based on dividing CNN into CAE and SNN’. Proc. Asia-Pacific Conf. Synthetic Aperture Radar, Singapore, Singapore, September 2015, pp. 676679.
    28. 28)
      • 28. Chen, S., Wang, H., Xu, F., et al: ‘Target classification using the deep convolutional networks for SAR images’, IEEE Trans. Geosci. Remote Sens., 2016, 54, (8), pp. 48064817.
    29. 29)
      • 29. Ding, J., Chen, B., Liu, H., et al: ‘Convolutional neural network with data augmentation for SAR target recognition’, IEEE Geosci. Remote Sens. Lett., 2016, 13, (3), pp. 364368.
    30. 30)
      • 30. Wagner, S., Barth, K., Brüggenwirth, S.: ‘A deep learning SAR ATR system using regularization and prioritized classes’. Proc. Radar Conf., Seattle, WA, USA, May 2017, pp. 772777.
    31. 31)
      • 31. Wagner, S.A.: ‘SAR ATR by a combination of convolutional neural network and support vector machines’, IEEE Trans. Aerosp. Electron. Syst., 2017, 52, (6), pp. 28612872.
    32. 32)
      • 32. Deng, S., Du, L., Li, C., et al: ‘SAR automatic target recognition based on Euclidean distance restricted autoencoder’, IEEE J. Sel. Top. Appl. Earth Observ. Remote Sens., 2017, 10, (7), pp. 33233333.
    33. 33)
      • 33. Lee, J., Wen, J., Ainsworth, T.L., et al: ‘Improved Sigma filter for speckle filtering of SAR imagery’, IEEE Trans. Geosci. Remote Sens., 2009, 47, (1), pp. 202213.
    34. 34)
      • 34. Glorot, X., Bengio, Y.: ‘Understanding the difficulty of training deep feedforward neural networks’. Int. Conf. Artificial Intelligence and Statistics, Chia Laguna Resort, Sardinia, Italy, May 2010, pp. 249256.
    35. 35)
      • 35. Maaten, L.V., Hinton, G.: ‘Visualizing data using t-SNE’, J. Mach. Learn. Res., 2008, 9, (2605), pp. 25792605.
http://iet.metastore.ingenta.com/content/journals/10.1049/iet-rsn.2018.5413
Loading

Related content

content/journals/10.1049/iet-rsn.2018.5413
pub_keyword,iet_inspecKeyword,pub_concept
6
6
Loading
This is a required field
Please enter a valid email address