Classification via weighted kernel CNN: application to SAR target recognition

ABSTRACT The conventional convolutional neural network (CNN) has proven to be effective for synthetic aperture radar (SAR) target recognition. However, the relationship between different convolutional kernels is not taken into account. The lack of the relationship limits the feature extraction capability of the convolutional layer to a certain extent. To address this problem, this paper presents a novel method named weighted kernel CNN (WKCNN). WKCNN integrates a weighted kernel module (WKM) into the common CNN architecture. The WKM is proposed to model the interdependence between different kernels, and thus to improve the feature extraction capability of the convolutional layer. The WKM consists of variables and activations. The variable represents the weight of the convolutional kernel. The activation is a mapping function which is used to determine the range of the weight. To adjust the variable adaptively, back propagation (BP) algorithm for the WKM is derived. The training of the WKM is driven by optimizing the cost function according to the BP algorithm, and three training modes are presented and analysed. SAR target recognition experiments are conducted on the moving and stationary target acquisition and recognition (MSTAR) dataset, and the results show the superiority of the proposed method.

[1]  Baiyuan Ding,et al.  Target recognition in SAR images by exploiting the azimuth sensitivity , 2017 .

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

[3]  Eric R. Keydel,et al.  MSTAR extended operating conditions: a tutorial , 1996, Defense, Security, and Sensing.

[4]  Yi Li,et al.  Deformable Convolutional Networks , 2017, 2017 IEEE International Conference on Computer Vision (ICCV).

[5]  Geoffrey E. Hinton,et al.  ImageNet classification with deep convolutional neural networks , 2012, Commun. ACM.

[6]  Vassilis Athitsos,et al.  Selective unsupervised feature learning with Convolutional Neural Network (S-CNN) , 2016, 2016 23rd International Conference on Pattern Recognition (ICPR).

[7]  Jian Yang,et al.  SAR Target Recognition via Supervised Discriminative Dictionary Learning and Sparse Representation of the SAR-HOG Feature , 2016, Remote. Sens..

[8]  Raman K. Mehra,et al.  Nonparametric error estimation techniques applied to MSTAR data sets , 1998, Defense, Security, and Sensing.

[9]  Guigang Zhang,et al.  Deep Learning , 2016, Int. J. Semantic Comput..

[10]  Jimmy Ba,et al.  Adam: A Method for Stochastic Optimization , 2014, ICLR.

[11]  Nitish Srivastava,et al.  Dropout: a simple way to prevent neural networks from overfitting , 2014, J. Mach. Learn. Res..

[12]  Naveen Kumar,et al.  An Overview of Automatic Target Recognition Systems for Underwater Mine Classification , 2016 .

[13]  Huanxin Zou,et al.  Deep Convolutional Highway Unit Network for SAR Target Classification With Limited Labeled Training Data , 2017, IEEE Geoscience and Remote Sensing Letters.

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

[15]  Peter F. McGuire,et al.  Automatic Target Recognition in Synthetic Aperture Radar Imagery: A State-of-the-Art Review , 2016, IEEE Access.

[16]  Rong Zhang,et al.  SAR target classification based on multiscale sparse representation , 2016, ISPRS International Conference on Computer Vision and Remote Sensing.

[17]  Nikhil Ketkar,et al.  Introduction to PyTorch , 2021, Deep Learning with Python.

[18]  G. Anagnostopoulos.  SVM-based target recognition from synthetic aperture radar images using target region outline descriptors , 2009 .

[19]  Vladlen Koltun,et al.  Multi-Scale Context Aggregation by Dilated Convolutions , 2015, ICLR.

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

[21]  John A. Saghri.  SAR automatic target recognition using maximum likelihood template-based classifiers , 2008, Optical Engineering + Applications.

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

[23]  Ning Li,et al.  SAR ATR based on displacement- and rotation-insensitive CNN , 2016 .

[24]  Qiang Guo,et al.  Convolutional feature learning and Hybrid CNN-HMM for scene number recognition , 2016, Neurocomputing.

[25]  M. W. Roth.  NEURAL NETWORKS FOR AUTOMATIC TARGET RECOGNITION , 2015 .

[26]  Thomas Wiatowski,et al.  A Mathematical Theory of Deep Convolutional Neural Networks for Feature Extraction , 2015, IEEE Transactions on Information Theory.

[27]  Sergey Ioffe,et al.  Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift , 2015, ICML.

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

[29]  Lawrence D. Jackel,et al.  Handwritten Digit Recognition with a Back-Propagation Network , 1989, NIPS.

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

[31]  Gangyao Kuang,et al.  Classification on the Monogenic Scale Space: Application to Target Recognition in SAR Image , 2015, IEEE Transactions on Image Processing.

[32]  T. Shimizu,et al.  Protective activity of glycolipid antigen against infection by Leptospira interrogans serovar canicola. , 1990, Journal of general microbiology.

[33]  Mark E. Oxley,et al.  Neural networks for automatic target recognition , 1995, Neural Networks.

[34]  Na Wang,et al.  Sparse Representation of Monogenic Signal: With Application to Target Recognition in SAR Images , 2014, IEEE Signal Processing Letters.

[35]  Dumitru Erhan,et al.  Going deeper with convolutions , 2014, 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[36]  Zongxu Pan,et al.  Transfer Learning with Deep Convolutional Neural Network for SAR Target Classification with Limited Labeled Data , 2017, Remote. Sens..