Classification of Reconstructed SAR Images Based on Convolutional Neural Network

Document Type : Original Article

Authors

1 Communications and Electronics Department Faculty of Electronic Engineering, Menoufia University Menouf, Egypt

2 Department of The Robotics and intelligent Mschines, Faculty of Artificial Intelligence, Kafrelsheikh University, Kafr El-Sheikh, Egypt

3 Electronics and Communications Engineering Department, Faculty of Engineering Minya University Minya, Egypt

4 Industrial Electronics andd Automatic Control Engineering Department Faculty of Electronic Engineering,Manoufia University: Menouf,Egypt

Abstract

Synthetic aperture radar (SAR) is a very important radar imaging type in which the utilization of antenna movement with respect to the target to be detected is considered. Detecting of target existence through noisy received images is a very critical and challenging point. Classification through deep learning presented in the form of Convolutional Neural Network (CNN) is a very good choice to enhance the decision performance reducing the error rate and false alarms. The main aim of this paper is to use a reliable classification technique in order to detect target existence through noisy received SAR images. Training data set for CNN is collected through a simulation in which realistic SAR images can be generated and used for SAR Automatic Target Recognition (ATR). CNNs are performed on images to classify the existence of targets. The accuracy of this approach is 100%.  

Keywords


[1]     S. A. Hovanessian, “Introduction to synthetic array and imaging radars,” Artech  House, Dedham, 1980.
[2]      J. C. Curlander, and R. N. McDounough, “Synthetic aperture radar, systems and  signal processing,” John Wiley & Sons, New York, 1991.
[3]     D. S. Garmatyuk, and R. M. Narayanan, “SAR imaging using fully random bandlimited signals”, Antennas and Propagation Society International Symposium, IEEE Vol. 4, pp.1948–1951, 2000
[4]     Chen Baixiao, Yang Minglei , Wang Yi , Dang Xiaofang, ”The applications and future of synthetic impulse and aperture radar”, CIE International Conference on Radar 2016, pp. 1-5, 2016.
[5]     C.P. Schwegmann, W. Kleynhans, B.P. Salmon, L.W. Mdakane, R.G.V. Meyer, J. Janoth, P. Lumsdon, "Transfer learning for multi-frequency synthetic aperture radar applications", Geoscience and Remote Sensing Symposium IGARSS 2018 - 2018 IEEE International, pp. 4403-4406, 2018
[6]     Chenguang Guo, Jiancheng Xu, Wenyao Xu, "Highly efficient design of SDRAM-based CTM for real-time SAR imaging system", Circuits Devices & Systems IET, vol. 13, no. 5, pp. 656-660, 2019
[7]     Matthew Schlutz. “Synthetic aperture radar imaging simulated in matlab”. Master's Thesis, California Polytechnic State University San Luis Obispo California, 2009.
[8]     S. Zhu, G. Liao, Yi Qu, Z. Zhou, and X. Liu, “ Ground moving targets imaging algorithm for synthetic aperture radar,” Geoscience and Remote Sensing, IEEE Trans. on, Vol. 49, pp. 462-477, 2011.
[9]      Glossary of remote sensing terms. Natural Resources Canada. [Online] Canada Centre for Remote Sensing, November 21, 2005.
[10]  Goodman, Joseph. “Introduction to fourier optics” Third Edition. Englewood, CO:Roberts & Company Publishers, 2005.
[11]  Soumekh, Mehrdad “Synthetic aperture radar signal processing with matlab algorithms” New York, NY: Wiley & Sons, Inc., 1999.
[12]  Y. Hara, R. G. Atkins, S. H. Yueh, R. T. Shin and J. A. Kong, "Application of neural networks to radar image classification," in IEEE Transactions on Geoscience and Remote Sensing, vol. 32, no. 1, pp. 100-109, Jan. 1994.
[13]  Topouzelis, K.N. "Oil spill detection by SAR images: dark formation detection, feature extraction and classification algorithms". Sensors,vol  8,pp. 6642-6659, 2008.
[14]  P. Yu, A. K. Qin and D. A. Clausi, "Unsupervised polarimetric SAR image segmentation and classification using region growing with edge penalty," in IEEE Transactions on Geoscience and Remote Sensing, vol. 50, no. 4, pp. 1302-1317, April 2012.
[15]  S Ochilov and D. A. Clausi, "Operational SAR sea-ice image classification," in IEEE Transactions on Geoscience and Remote Sensing, vol. 50, no. 11, pp. 4397. -4408, Nov. 2012
[16]  Phil Kim.,''MATLAB Deep Learning: With Machine Learning, Neural Networks and Artificial Intelligence'' Book, pp. 121-132, 2017.
[17]  Alex Krizhevsky, Ilya Sutskever, Geoffrey E. Hinton,''Imagenet classification with  deep convolutional neural networks'', NIPS'12 Proceedings of the 25th International Conference on Neural Information Processing Systems, vol 1, pp1097-1105,2012.
[18]  Y. K. Chan, and S. Y. Lim, “Synthetic Aperture Radar (SAR) signal generation,”Progress In Electromagnetics Research B, Vol. 1, pp. 269–290, 2008.
[19]  LeCun, Y., Bottou, L., Bengio, Y., & Haffner, P. ''Gradient-based learning applied to document recognition''. Proceedings of the IEEE,vol 86, pp 2278-2324., 1998.
[20]  Ranzato, M. A., Huang, F. J., Boureau, Y. L., & LeCun, Y. ''Unsupervised learning of invariant feature hierarchies with applications to object recognition,” In Computer Vision and Pattern Recognition, CVPR'07. IEEE Conference ,pp. 1-8, 2007.
[21]  Srivastava, N., Hinton, G., Krizhevsky, A., Sutskever, I., & Salakhutdinov, R. ''Dropout: A simple way to prevent neural networks from overfitting''. The Journal of Machine Learning Research,vol 15,pp 1929-1958,2014.
 
Volume 28, ICEEM2019-Special Issue
ICEEM2019-Special Issue: 1st International Conference on Electronic Eng., Faculty of Electronic Eng., Menouf, Egypt, 7-8 Dec.
2019
Pages 122-125