AN ADVANCED NEURAL NETWORK-BASED APPROACH FOR MILITARY GROUND VEHICLE RECOGNITION IN SAR AERIAL IMAGERY

The paper presents a novel neural network approach for automatic target recognition (ATR) in the synthetic aperture radar (SAR) aerial imagery; this is applied to identify military ground vehicles. The proposed ATR algorithm consists of a processing cascade with the following stages: (a) object detection using a pulse-coupled neural network (PCNN) segmentation module; (b) a first feature selection module using Gabor filtering (GF); (c) a second feature selection module using principal component analysis (PCA); (d) a support vector machine (SVM) classifier improved by using virtual training data generation (VTDG) with concurrent self-organization maps (CSOM). The proposed model has been applied for the recognition of three classes of military ground vehicles of the former Soviet Union represented by the set of 2987 images of the MSTAR public release database. The experimental results have confirmed the method effectiveness, leading to a total success rate of 97.36%.

[1]  Heggere S. Ranganath,et al.  Perfect image segmentation using pulse coupled neural networks , 1999, IEEE Trans. Neural Networks.

[2]  Tetsuo Kirimoto,et al.  Accurate and Robust Automatic Target Recognition Method for SAR Imagery with SOM-Based Classification , 2012, IEICE Trans. Commun..

[3]  Victor-Emil Neagoe,et al.  A new approach for accurate classification of hyperspectral images using Virtual Sample Generation by Concurrent Self-Organizing Maps , 2013, 2013 IEEE International Geoscience and Remote Sensing Symposium - IGARSS.

[4]  Joni-Kristian Kämäräinen,et al.  Fundamental frequency Gabor filters for object recognition , 2002, Object recognition supported by user interaction for service robots.

[5]  Joni-Kristian Kämäräinen,et al.  Simple Gabor feature space for invariant object recognition , 2004, Pattern Recognit. Lett..

[6]  Heggere S. Ranganath,et al.  Object detection using pulse coupled neural networks , 1999, IEEE Trans. Neural Networks.

[7]  Nicholas M Sandirasegaram.  Automatic Target Recognition in SAR Imagery Using a MLP Neural Network , 2002 .

[8]  Radford M. Neal.  Pattern Recognition and Machine Learning , 2007, Technometrics.

[9]  Joni-Kristian Kämäräinen,et al.  Robustness of Gabor Feature Parameter Selection , 2002, MVA.

[10]  Saman A. Zonouz,et al.  Identification Using Encrypted Biometrics , 2013, CAIP.

[11]  Francesca Bovolo,et al.  Concurrent Self-Organizing Maps for Supervised/Unsupervised Change Detection in Remote Sensing Images , 2014, IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing.

[12]  Yang Ruliang,et al.  SAR Target Recognition Based on MRF and Gabor Wavelet Feature Extraction , 2008, IGARSS 2008 - 2008 IEEE International Geoscience and Remote Sensing Symposium.

[13]  Juha A. Karvonen,et al.  Baltic Sea ice SAR segmentation and classification using modified pulse-coupled neural networks , 2004, IEEE Transactions on Geoscience and Remote Sensing.

[14]  V. Neagoe,et al.  Concurrent self-organizing maps -a powerfuu artificial neural tool for biometric technology , 2004, Proceedings World Automation Congress, 2004..

[15]  Victor-Emil Neagoe,et al.  Automatic target recognition in SAR imagery using pulse-coupled neural network segmentation cascaded with virtual training data generation CSOM-based classifier , 2015, 2015 IEEE International Geoscience and Remote Sensing Symposium (IGARSS).

[16]  Bülent Sankur,et al.  Survey over image thresholding techniques and quantitative performance evaluation , 2004, J. Electronic Imaging.

[17]  Zhipeng Liu,et al.  Adaptive boosting for SAR automatic target recognition , 2007, IEEE Transactions on Aerospace and Electronic Systems.