基于2DPCA-SCN正则化的SAR图像目标识别方法
|
摘要
合成孔径雷达(synthetic aperture radar,SAR)图像解译是一项重大的科学应用挑战,SAR图像目标识别已成为该领域的主要研究方向之一。针对SAR图像识别算法训练参数较多的问题,本文提出一种二维主成分分析(two-dimensional principal component analysis,2DPCA)与L2正则化约束的随机配置网络(stochastic configuration network,SCN)进行集成学习的SAR图像目标识别方法。2DPCA不仅能够有效地提取出目标的特征信息而且通过稀疏表征方式降低数据量,SCN正则化算法参数较少且可以有效避免网络过拟合问题,提高网络的识别率。我们将提出的方法在MSTAR(moving and stationary target acquisition and recognition)数据集上进行实验,结果表明该方法相对传统方法具有更高的识别率。
Synthetic aperture radar(SAR) image interpretation is a major scientific application challenge. SAR image target recognition has become one of the main research directions in this field. In view of the fact that target recognition algorithms for SAR images needs many training parameters. This paper proposes a target recognition method based on the two-dimensional principal component analysis(2 DPCA) and the L2 regularization constrained stochastic configuration network(SCN) for integrated learning. 2 DPCA can not only effectively extract the feature information of the target but also reduce the amount of data through sparse representation. The Regularized SCN algorithm can effectively avoid the over-fitting problem and improve the recognition rate of the network. Based on the moving and stationary target acquisition and recognition(MSTAR) dataset, the experimental results show that the proposed method can achieve a higher recognition rate than the traditional method.
Synthetic aperture radar(SAR) image interpretation is a major scientific application challenge. SAR image target recognition has become one of the main research directions in this field. In view of the fact that target recognition algorithms for SAR images needs many training parameters. This paper proposes a target recognition method based on the two-dimensional principal component analysis(2 DPCA) and the L2 regularization constrained stochastic configuration network(SCN) for integrated learning. 2 DPCA can not only effectively extract the feature information of the target but also reduce the amount of data through sparse representation. The Regularized SCN algorithm can effectively avoid the over-fitting problem and improve the recognition rate of the network. Based on the moving and stationary target acquisition and recognition(MSTAR) dataset, the experimental results show that the proposed method can achieve a higher recognition rate than the traditional method.
引文
[1] 徐丰,王海鹏,金亚秋.深度学习在SAR目标识别与地物分类中的应用[J].雷达学报,2017,6(2):136-148.Xu Feng,Wang Haipeng,Jin Yaqiu.Deep Learning as Applied in SAR Target Recognition and Terrain Classification[J].Journal of Radars,2017,6(2):136-148.(in Chinese)
[2] 张红,王超,张波,等.高分辨率SAR图像目标识别[M].北京:科学出版社,2009:4-7.Zhang Hong,Wang Chao,Zhang Bo,et al.High Resolution SAR Image Target Recognition[M].Beijing:Science Press,2009:4-7.(in Chinese)
[3] 盖旭刚,陈晋汶,韩俊,等.合成孔径雷达的现状与发展趋势[J].飞航导弹,2011(3):82- 86.Gai Xugang,Chen Jinwen,Han Jun,et al.Present Situation and Development Trend of Synthetic Aperture Radar[J].Aerodynamic Missile Journal,2011(3):82- 86.(in Chinese)
[4] Moreira A,Prats-Iraola P,Younis M,et al.A tutorial on synthetic aperture radar[J].IEEE Geoscience and Remote Sensing Magazine,2013,1(1):6- 43.
[5] 田壮壮,占荣辉,胡杰民,等.基于卷积神经网络的SAR图像目标识别研究[J].雷达学报,2016,5(3):320-325.Tian Zhuangzhuang,Zhan Ronghui,Hu Jiemin,et al.SAR ATR Based on Convolutional Neural Network[J].Journal of Radars,2016,5(3):320-325.(in Chinese)
[6] LEcun Y,Bottou L,Bengio Y,et al.Radient-Based Learning Applied to Document Recognition[J].Proceedings of the IEEE,1998,86(11):2278-2324.
[7] Mishra A.Validation of PCA and LDA for SAR ATR[C]//IEEE TENCON:IEEE,2008:1- 6.
[8] Chen T,Chen H.Universal approximation to nonlinear operators by neural networks[J].Neural Computation,2002,14(11):2561-2566.
[9] Kotsiantis S B.Supervised machine learning:a review of classification techniques[C]//Conference on Emerging Artificial Intelligence Applications in Computer Engineering:Real Word Ai Systems with Applications in Ehealth,Hci,Information Retrieval and Pervasive Technologies.IOS Press,2007:3-24.
[10] Igelnik B,Pao Y H.Stochastic choice of basic functions in adaptive function approximation and the functional-link net[J].IEEE Transactions on Neural Networks,1995,6(6):1320-1329.
[11] Wang D,Li M.Stochastic Configuration Networks:Fundamentals and Algorithms[J].IEEE Trans Cybern,2017,47(10):3466-3479.
[12] Mishra A,Mulgrew B.Bistatic SAR ATR[J].IET Radar Sonar & Navigation,2007,1(6):459- 469.
[13] Pei J,Huang Y,Huo W,et al.SAR Imagery Feature Extraction Using 2DPCA-Based Two-Dimensional Neighborhood Virtual Points Discriminant Embedding[J].IEEE Journal of Selected Topics in Applied Earth Observations & Remote Sensing,2017,9(6):2206-2214.
[14] Yang J,Zhang D,Frangi A F,et al.Two-dimensional PCA:a New Approach to Appearance-based Face Representation and Recognitio[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2004,26(1):131-137.
[15] Chen S,Wang H,Xu F,et al.Target classification using the deep convolutional networks for SAR images[J].IEEE Transactions on Geoscience and Remote Sensing,2016,54(8):4806- 4817.
[16] Ding B,Wen G,Zhong J,et al.Robust method for the matching of attributed scattering centers with application to synthetic aperture radar automatic target recognition[J].Journal of Applied Remote Sensing,2016,10(1):016010.
[17] Zhang H,Nasrabadi N M,Zhang Y,et al.Multi-View Automatic Target Recognition using Joint Sparse Representation[J].IEEE Transactions on Aerospace and Electronic Systems,2012,48(3):2481- 2497.
[2] 张红,王超,张波,等.高分辨率SAR图像目标识别[M].北京:科学出版社,2009:4-7.Zhang Hong,Wang Chao,Zhang Bo,et al.High Resolution SAR Image Target Recognition[M].Beijing:Science Press,2009:4-7.(in Chinese)
[3] 盖旭刚,陈晋汶,韩俊,等.合成孔径雷达的现状与发展趋势[J].飞航导弹,2011(3):82- 86.Gai Xugang,Chen Jinwen,Han Jun,et al.Present Situation and Development Trend of Synthetic Aperture Radar[J].Aerodynamic Missile Journal,2011(3):82- 86.(in Chinese)
[4] Moreira A,Prats-Iraola P,Younis M,et al.A tutorial on synthetic aperture radar[J].IEEE Geoscience and Remote Sensing Magazine,2013,1(1):6- 43.
[5] 田壮壮,占荣辉,胡杰民,等.基于卷积神经网络的SAR图像目标识别研究[J].雷达学报,2016,5(3):320-325.Tian Zhuangzhuang,Zhan Ronghui,Hu Jiemin,et al.SAR ATR Based on Convolutional Neural Network[J].Journal of Radars,2016,5(3):320-325.(in Chinese)
[6] LEcun Y,Bottou L,Bengio Y,et al.Radient-Based Learning Applied to Document Recognition[J].Proceedings of the IEEE,1998,86(11):2278-2324.
[7] Mishra A.Validation of PCA and LDA for SAR ATR[C]//IEEE TENCON:IEEE,2008:1- 6.
[8] Chen T,Chen H.Universal approximation to nonlinear operators by neural networks[J].Neural Computation,2002,14(11):2561-2566.
[9] Kotsiantis S B.Supervised machine learning:a review of classification techniques[C]//Conference on Emerging Artificial Intelligence Applications in Computer Engineering:Real Word Ai Systems with Applications in Ehealth,Hci,Information Retrieval and Pervasive Technologies.IOS Press,2007:3-24.
[10] Igelnik B,Pao Y H.Stochastic choice of basic functions in adaptive function approximation and the functional-link net[J].IEEE Transactions on Neural Networks,1995,6(6):1320-1329.
[11] Wang D,Li M.Stochastic Configuration Networks:Fundamentals and Algorithms[J].IEEE Trans Cybern,2017,47(10):3466-3479.
[12] Mishra A,Mulgrew B.Bistatic SAR ATR[J].IET Radar Sonar & Navigation,2007,1(6):459- 469.
[13] Pei J,Huang Y,Huo W,et al.SAR Imagery Feature Extraction Using 2DPCA-Based Two-Dimensional Neighborhood Virtual Points Discriminant Embedding[J].IEEE Journal of Selected Topics in Applied Earth Observations & Remote Sensing,2017,9(6):2206-2214.
[14] Yang J,Zhang D,Frangi A F,et al.Two-dimensional PCA:a New Approach to Appearance-based Face Representation and Recognitio[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2004,26(1):131-137.
[15] Chen S,Wang H,Xu F,et al.Target classification using the deep convolutional networks for SAR images[J].IEEE Transactions on Geoscience and Remote Sensing,2016,54(8):4806- 4817.
[16] Ding B,Wen G,Zhong J,et al.Robust method for the matching of attributed scattering centers with application to synthetic aperture radar automatic target recognition[J].Journal of Applied Remote Sensing,2016,10(1):016010.
[17] Zhang H,Nasrabadi N M,Zhang Y,et al.Multi-View Automatic Target Recognition using Joint Sparse Representation[J].IEEE Transactions on Aerospace and Electronic Systems,2012,48(3):2481- 2497.