SAR object classification using the DAE with a modified triplet restriction

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.

[1]  Kerry E. Dungan.  Feature-based Vehicle Classification in Wide-angle Synthetic Aperture Radar , 2010 .

[2]  Jie Geng,et al.  High-Resolution SAR Image Classification via Deep Convolutional Autoencoders , 2015, IEEE Geoscience and Remote Sensing Letters.

[3]  H. Scott Clouse,et al.  Convolutional neural networks for synthetic aperture radar classification , 2016, SPIE Defense + Security.

[4]  Yang Li,et al.  DeepSAR-Net: Deep convolutional neural networks for SAR target recognition , 2017, 2017 IEEE 2nd International Conference on Big Data Analysis (ICBDA)(.

[5]  Simon Wagner,et al.  A deep learning SAR ATR system using regularization and prioritized classes , 2017, 2017 IEEE Radar Conference (RadarConf).

[6]  Thomas L. Ainsworth,et al.  Improved Sigma Filter for Speckle Filtering of SAR Imagery , 2009, IEEE Transactions on Geoscience and Remote Sensing.

[7]  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.

[8]  Haipeng Wang,et al.  SAR target recognition based on deep learning , 2014, 2014 International Conference on Data Science and Advanced Analytics (DSAA).

[9]  Hongliang Ren,et al.  Hierarchical Recognition System for Target Recognition from Sparse Representations , 2015 .

[10]  Xuan Li,et al.  SAR ATR based on dividing CNN into CAE and SNN , 2015, 2015 IEEE 5th Asia-Pacific Conference on Synthetic Aperture Radar (APSAR).

[11]  Shuyang Wang,et al.  SAR target recognition and unsupervised detection based on convolutional neural network , 2017, 2017 Chinese Automation Congress (CAC).

[12]  David Malmgren-Hansen,et al.  Training Convolutional Neural Networks for Translational Invariance on SAR ATR , 2016 .

[13]  Yu Zhou,et al.  Polarimetric SAR Image Classification Using Deep Convolutional Neural Networks , 2016, IEEE Geoscience and Remote Sensing Letters.

[14]  Nabil Aouf,et al.  SAR automatic target recognition based on convolutional neural networks , 2017 .

[15]  Gongjian Wen,et al.  Target Recognition in Synthetic Aperture Radar Images via Matching of Attributed Scattering Centers , 2017, IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing.

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

[17]  Bo Zhang,et al.  Classification of very high resolution SAR image based on convolutional neural network , 2017, 2017 International Workshop on Remote Sensing with Intelligent Processing (RSIP).

[18]  Simon Wagner.  Combination of convolutional feature extraction and support vector machines for radar ATR , 2014, 17th International Conference on Information Fusion (FUSION).

[19]  Thomas S. Huang,et al.  Multi-View Automatic Target Recognition using Joint Sparse Representation , 2012, IEEE Transactions on Aerospace and Electronic Systems.

[20]  David Morgan,et al.  Deep convolutional neural networks for ATR from SAR imagery , 2015, Defense + Security Symposium.

[21]  Jia Cheng Ni,et al.  SAR automatic target recognition based on a visual cortical system , 2013, 2013 6th International Congress on Image and Signal Processing (CISP).

[22]  Jianyu Yang,et al.  Target recognition in synthetic aperture radar images via non-negative matrix factorisation , 2015 .

[23]  Baiyuan Ding,et al.  Target Reconstruction Based on 3-D Scattering Center Model for Robust SAR ATR , 2018, IEEE Transactions on Geoscience and Remote Sensing.

[24]  A. Mishra,et al.  Application of linear and nonlinear PCA to SAR ATR , 2015, 2015 25th International Conference Radioelektronika (RADIOELEKTRONIKA).

[25]  Shiyong Cui,et al.  Convolutional Neural Network for SAR image classification at patch level , 2016, 2016 IEEE International Geoscience and Remote Sensing Symposium (IGARSS).

[26]  Lin Jin,et al.  A method for automatic target recognition using shadow contour of SAR image , 2013 .

[27]  Yikui Zhai,et al.  SAR automatic target recognition based on Local Phase Quantization plus Biomimetic Pattern Recognition , 2012, 2012 IEEE 11th International Conference on Signal Processing.

[28]  Jianwei Li,et al.  A performance analysis of convolutional neural network models in SAR target recognition , 2017, 2017 SAR in Big Data Era: Models, Methods and Applications (BIGSARDATA).

[29]  Hongwei Liu,et al.  Convolutional Neural Network With Data Augmentation for SAR Target Recognition , 2016, IEEE Geoscience and Remote Sensing Letters.

[30]  Hongwei Liu,et al.  SAR Automatic Target Recognition Based on Euclidean Distance Restricted Autoencoder , 2017, IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing.

[31]  Yoshua Bengio,et al.  Understanding the difficulty of training deep feedforward neural networks , 2010, AISTATS.

[32]  Geoffrey E. Hinton,et al.  Visualizing Data using t-SNE , 2008 .

[33]  Shuang Wang,et al.  Multilayer feature learning for polarimetric synthetic radar data classification , 2014, 2014 IEEE Geoscience and Remote Sensing Symposium.

[34]  Wang Ning,et al.  A survey on feature extraction of SAR Images , 2010, 2010 International Conference on Computer Application and System Modeling (ICCASM 2010).