Feature extraction is a key step and difficult point in SAR image target recognition. This paper presents a novel method based on Poisson Gamma Belief Network (PGBN) for SAR image target recognition. As a deep Bayesian generative network, the PGBN model obtains a more structured multi-layer feature representation from the complex SAR image data using the high nonlinearity of the Gamma distribution, and the multi-layer feature representation effectively improves SAR image target recognition performance. In order to obtain a higher recognition rate and efficiency of training, this paper further proposes a method for classifying PGBN model based on the Naive Bayes rule. The experimental results about MSTAR dataset show that the feature extracted by this new method has better structure information, and it has better performance for SAR image target recognition.
ZHANG Hong, WANG Chao, and ZHANG Bo. High Resolution SAR Images Target Recognition[M]. Beijing: Science Press, 2009: 5.2 Section.
[2]
保铮, 邢孟道, 王彤. 雷达成像技术[M]. 北京: 电子工业出版社, 2004: 1.1 节.
BAO Zheng, XING Mengdao, and WANG Tong. Radar Imaging Technology[M]. Beijing: Publishing House of Electronics Industry, 2004: 1.1 Section.
[3]
HE Zhiguo, LU Jun, and YAO Kuanggang. A fast SAR target recognition approach using PCA features[C]. International Conference on Image and Graphics, Chengdu, China, 2007: 580-585.
[4]
LIN C, PENG F, WANG B H, et al. Research on PCA and KPCA self-fusion based MSTAR SAR automatic target recognition algorithm[J]. Journal of Electronic Science and Technology, 2012, 10(4): 352-357.
HUAN Ruohong and YANG Ruliang. SAR images feature extraction and target recognition based on ICA and SVM[J]. Computer Engineering, 2008, 34(13): 24-25.
[6]
LEE D D and SEUNG H S. Learning the parts of objects by non-negative matrix factorization[J]. Nature, 1999, 401(6755): 788-791.
[7]
LEE D D and SEUNG H S. Algorithms for non-negative matrix factorization[C]. Neural Information Processing Systems, Denver, CO, USA, 2000: 556-562.
LONH Honglin, PI Yiming, and CAO Zongjie . Non-negative matrix factorization for target recognition[J]. Acta Electronica Sinica, 2010, 38(6): 1425-1429.
[9]
ZHOU Mingyuan and CARIN Lawrence. Beta-negative binomial process and Poisson factor analysis[C]. Artificial Intelligence and Statistics. La Palma, Canary Islads, Spain, 2012: 1462-1471.
[10]
孙洪. 高分辨率SAR图像目标识别[M]. 北京: 电子工业出版社, 2013: 5.1 节.
SUN Hong. Processing of Synthetic Aperture Radar Images [M]. Beijing: Publishing House of Electronics Industry, 2013: 5.1 Section.
[11]
ZHOU Mingyuan, CONG Yulai, and CHEN Bo. The Poisson Gamma belief network[C]. Neural Information Processing Systems, Montreal, Canda, 2015: 562-570.
[12]
CHEN Y, ZHAO X, and JIA X. Spectral-Spatial classification of hyperspectral data based on deep belief network[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2015, 8(6): 1-12. doi: 10.1109/JSTAR.2015.2388577.
ZHANG Xuefeng. Study of radar target recognition and outlier rejection based on high range resolution profiles[D]. [Ph.D. dissertation], Xi dian University, 2016: 71-73.
[14]
LIU X, LIU R, MA J, et al. Privacy-preserving patent-centric clinical decision support system on NaÏve Bayes classification [J]. IEEE Journal of Biomedical & Health Informatics, 2016, 20(2): 655-668. doi: 10.1109/JBHI.2015.2407157.
[15]
CHAN Chihchung and LIN Chinjen. LIBSVM: A library for support vector machines[J]. ACM Transactions on Intelligent Systems and Technology, 2011, 2(3): 1-27. doi: 10.1145/ 1961189.1961199.
[16]
GE Hinton. A practical guide training restricted boltzmann machines[J]. Momentum, 2010, 9(1): 599-619. doi: 10.007/ 978-3-642-35289-8_32.
DING Jun, LIU Hongwei, and CHEN Bo. Application of similar constraints deep belief networks in SAR image target recognition[J]. Journal of Electronics & Information Technology, 2016, 38(1): 91-103. doi: 10.11999/JEIT150366.
DING Jun, LIU Hongwei, WANG Yinghua, et al. The method of SAR target recognition with joint shadow region and target region image[J]. Journal of Electronics & Information Technology, 2015, 37(3): 594-600. doi: 10.11999/ JEIT140713.