MLP在雷达参数反演和SAR图像分类中的应用
|
摘要
本文主要介绍了神经网络在微波遥感领域的地面参数反演和地物识别与分类上的应用，并用ENVISAT-ASAR数据和AirSAR数据做了实际分析。
近些年来，使用神经网络进行参数的反演和地物的识别与分类是一种重要和先进的方法。他充分利用了神经网络的特性，解决了遥感领域参数反演和分类的许多复杂的问题和操作。
神经网络与散射模型相结合使得准确和实时的进行参数反演和分类成为可能。这里我们使用的神经网络是用快速学习算法(FL)训练的多层前馈网络(MLP)，训练速度和精度与传统神经网络相比都有较大提高，网络结构是全连接的。反演使用的模拟数据是用IEM模型得到的，并用来训练神经网络。因此，训练数据可以被认为是从一个完全知道随机粗糙表面得到的数据。神经网络的数据是各个角度和极化的后向散射系数(σ~0(θ))，输出是地面的散射参数，包括层介电常数(ε)、地表相关长度(kι)、和地表粗糙度(kσ)。
同样，神经网络也能够被应用的地物识别与分类中。本次研究中，还是使用与刚才一样的全连接的、使用FL算法训练的MLP网络，网络的训练数据是从各个确定目标地物得到的各个极化的后向散射系数。训练好的神经网络被应用到ENVISAT-ASAR数据和AirSAR数据中进行地物分类。并把得到的结果再与其他的分类的方法的结果做相关的比较。
This paper describes the application of neural networks to surface parameters retrieval and targets classification from multi-polarization ENVISAT-ASAR datas and AirSAR datas. It is an important advancement to use neural networks to perform inversion and classification in Remote Sensing recently. The combination of a scattering model (SM) and neural networks make it possible to perform inversion and classification accurately and in real time. The used neural network is multilayer perceptron (MLP) with fast learning (FL), which is fully interconnected network. Simulated data sets based on the Integration Equation Model (IEM) are used to train the neural network. Accordingly, the training data sets may be viewed as taken from a completely known randomly rough surface. The input to the neural network is the set of values (σ0(θ)) with angles and polarizations, the output of the neural network is the set of surface scattering parameters. The layer permittivity (ε), surface correlation length (kl) and surface r
oughness (kσ) are retrieved from σ0(θ) using the trained MLP.
As above, this method can be used into the classifier. For this aim, we suggest the proposed fully interconnected MLP with FL, in which the training data sets are values (σ0(θ)) with polarizations from some identified targets. The trained neural networks is used in target classification in the ENVISAT-ASAR datas and AirSAR datas. And finally, the results of proposed method are compared with that of the unsupervised classification one, the in situ test data are from Zhaoqing in Guangdong Province and Taichung in Taiwan Province in China.
近些年来，使用神经网络进行参数的反演和地物的识别与分类是一种重要和先进的方法。他充分利用了神经网络的特性，解决了遥感领域参数反演和分类的许多复杂的问题和操作。
神经网络与散射模型相结合使得准确和实时的进行参数反演和分类成为可能。这里我们使用的神经网络是用快速学习算法(FL)训练的多层前馈网络(MLP)，训练速度和精度与传统神经网络相比都有较大提高，网络结构是全连接的。反演使用的模拟数据是用IEM模型得到的，并用来训练神经网络。因此，训练数据可以被认为是从一个完全知道随机粗糙表面得到的数据。神经网络的数据是各个角度和极化的后向散射系数(σ~0(θ))，输出是地面的散射参数，包括层介电常数(ε)、地表相关长度(kι)、和地表粗糙度(kσ)。
同样，神经网络也能够被应用的地物识别与分类中。本次研究中，还是使用与刚才一样的全连接的、使用FL算法训练的MLP网络，网络的训练数据是从各个确定目标地物得到的各个极化的后向散射系数。训练好的神经网络被应用到ENVISAT-ASAR数据和AirSAR数据中进行地物分类。并把得到的结果再与其他的分类的方法的结果做相关的比较。
This paper describes the application of neural networks to surface parameters retrieval and targets classification from multi-polarization ENVISAT-ASAR datas and AirSAR datas. It is an important advancement to use neural networks to perform inversion and classification in Remote Sensing recently. The combination of a scattering model (SM) and neural networks make it possible to perform inversion and classification accurately and in real time. The used neural network is multilayer perceptron (MLP) with fast learning (FL), which is fully interconnected network. Simulated data sets based on the Integration Equation Model (IEM) are used to train the neural network. Accordingly, the training data sets may be viewed as taken from a completely known randomly rough surface. The input to the neural network is the set of values (σ0(θ)) with angles and polarizations, the output of the neural network is the set of surface scattering parameters. The layer permittivity (ε), surface correlation length (kl) and surface r
oughness (kσ) are retrieved from σ0(θ) using the trained MLP.
As above, this method can be used into the classifier. For this aim, we suggest the proposed fully interconnected MLP with FL, in which the training data sets are values (σ0(θ)) with polarizations from some identified targets. The trained neural networks is used in target classification in the ENVISAT-ASAR datas and AirSAR datas. And finally, the results of proposed method are compared with that of the unsupervised classification one, the in situ test data are from Zhaoqing in Guangdong Province and Taichung in Taiwan Province in China.
引文
[1] Mu-Song Chen and Michael T.Manry, Conventional Modeling of the Multilayer Perceptron Using Polynomial Basis Functions, IEEE Transaction on Neural Network, Vol 4, No 1, Jan 1993, pp164-166
[2] Mu-Song Chen and Michael T.Manry, Power Series Analyses of Back-Propagation Neural Networks, IEEE, 1991, Ⅰ-295—Ⅰ-300
[3] Mu-Song Chen and Michael T.Manry, Back-Propagation Representation Theorem Using Power Series, International Joint Conf on Neural Networks, IJCNN, Vol Ⅰ, 1990, Ⅰ-643—Ⅰ-648
[4] A.S.Pandya and R.Szabo, A fast learning algorithm for neural network application, Proceeding of IEEE conference on SMC, Oct, 1991, pp 1569-1573
[5] G.Govind and P.A.Ramamoorthy, Multi-layered neural networks and Volterra Series: The missing link, In Proc. Int. Conf. Systems Engineering, Pittsburgh, PA, Aug, 1990
[6] G.W.Davis and M.L.Gasperi, ANN Modeling of Volterra systems, In Proc. IJCNN'91, Seattle WA, pp Ⅱ 727-734
[7] S.A.Barton, A matrix method for optimizing a neural networks, Neural computation, Vol 3, No.3, 1991, pp450-459
[8] P.Gallinari, S.Thiria, and F.Fogelman Soulie, Multilayer perceptrons and data analysis, in Proc. IJCNN, Vol 1, 1988, pp391-399
[9] W.Schmidt, A non-iterative method for training feed forward networks, in Proc. IJCNN, Vol Ⅱ, 1991, pp 19-24
[10] Xian pingjiang, Analysis Optimization of Neural Networks for Remote Sensing, Remote Sensing Reviews, 1994
[11] Jonathan Lee, A Neural Network Approach to Cloud Classification, IEEE, Transactions on Geoscience and Remote Sensing, VOL 28, NO 5, Sep, 1990, pp846-855
[12] J.P.Fitch, Ship Wake-Detection Procedure Using Conjugate Gradient Trained Artifical Neural Networks, IEEE, Transactions on Geoscience and Remote
Sensing, VOL 29, NO 5, Sep, 1991, pp718-726
[13] Philip D.Heermann, Classification of Multispectral Remote Sensing Data Using a Back-Propagation Neural Networks, IEEE, Transactions on Geoscience and Remote Sensing, VOL 30, NO 1, JAN, 1992, pp81-88
[14] H. Bischof, Multispectral Classification of Landsat-Images Using Neural Networks, IEEE, Transactions on Geoscience and Remote Sensing, VOL 30, NO 3, MAY, 1992, pp482-490
[15] Classification of Mutltispectral Image Data by the Binary Diamond Neural Networks and by Nonparametric, Pixel by Pixel methods, IEEE, Transactions on Geoscience and Remote Sensing, VOL 31, NO 3, MAY, 1993, pp606-617
[16] Daniel T. Davis, Retrieval of Snow Parameters by Iterative Inversion of Neural Networks IEEE, Transactions on Geoscience and Remote Sensing, VOL 31, NO 4, JULY, 1993, pp842-852
[17] Yu Chang Tzeng, A Dynamic Learning Neural Networks for Remote Sensing Applications, IEEE, Transactions on Geoscience and Remote Sensing, VOL 32, NO 5, SEPT, 1994, pp 1096-1102
[18] Sucharita Gopal, Remote Sensing of Forest Change Using Artifical Neural Networks, IEEE, Transactions on Geoscience and Remote Sensing, VOL 34, NO 2, MARCH, 1996, pp398-402
[19] Li Li, Microwave Radiometric Technique to Retrieve Vapor, Liquid and Ice, Part Ⅰ—Development of a Neural Network-Based Inversion methods, IEEE, Transactions on Geoscience and Remote Sensing, VOL 35, NO 2, MARCH, 1997, pp224-236
[20] Dimitris Tsintikidis, A Neural Network Approach to Estimating Rainfall from Spaceborne Microwave Data, IEEE, Transactions on Geoscience and Remote Sensing, VOL 35, NO 5, SEPT, 1997, pp1079-1093
[21] Laurent Ferro-Famil, Unsupervised Classification of Multifrequency and Fully Polarimetric SAR Images Based on the H/A/Alpha—Wishart Classifier, IEEE, Transactions on Geoscience and Remote Sensing, VOL 39, NO 11, NOV, 2001, pp2332-2341
[22] Jong-Sen Lee, Unsupervised Classification Using Polarimetric Decomposition and the Complex Wishart Classifier, IEEE, Transactions on Geoscience and Remote Sensing, VOL 37, NO 5, SEP, 1999, pp2249-2258
[23] Jiancheng Shi and Jeff Dozier, Estimation of Snow Water Equivalence Using SIR-C/X-SAR, Part Ⅱ: Interring Snow Depth and Particle Size, IEEE, Transactions on Geoscience and Remote Sensing, VOL 38, NO 6, NOV, 2000, pp2475-2487
[24] Yisok Oh, Kamal Sarabandi, An Empirical Model and an Inversion Technique for Radar Scattering from Bare Soil Surfaces, IEEE, Transactions on Geoscience and Remote Sensing, 1992
[25] Pascale C. Dubois, Jakob van Zyl, Measuring Soil Moisture with Imaging Radars, IEEE, Transactions on Geoscience and Remote Sensing, 1995, pp915-925
[26] 郭华东等，雷达对地观测理论与应用，科学出版社，2000
[27] 郭华东等，中国雷达遥感图像分析，科学出版社，1999
[28] 周成虎等，遥感影像地学理解与分析，科学出版社，1999
[29] 容观澳等，计算机图像处理，清华大学出版社，1998
[30] 舒宁，微波遥感原理，武汉大学出版社，2003
[31] 章孝灿，黄智才，遥感数字图像处理，浙江大学出版社，1997
[32] [美]S．Weisberg，应用线性回归，中国统计出版社，1998
[33] 吴喜之，现代回归模型诊断，中国统计出版社，2003
[34] 李小文，先验知识在遥感反演中的作用，中国科学，D辑，1998，28(1)67—72
[35] 杨劲松，合成孔径雷达图像的近海面风场的反演，遥感学报，VOL 5，NO 1，JAN，2000．1，pp13—16
[36] 陆桂华，遗传算法在马斯京根模型参数估计中的应用，何海大学学报，VOL29，NO 4，JUL，2001．7，pp9—12
[37] 张佳华，利用遥感反演的页面指数研究中国东部的生态系统对东亚季风的影响，自然科学进展，2002．10，pp1098—1100
[38] 李国胜，东海真成光深度的遥感反演与影响机理研究，自然科学进展，2003．1，pp90—94
[39] 王耀南，小波神经网络的遥感图像分类，中国图像图形学报，1999．5，pp368—371
[40] 孙丹峰，自组织神经网络在遥感土地覆盖分类中的应用，遥感学报，1999．5，pp139—143
[41] 陈华，陈书海，K-means算法在遥感分类中的应用，红外与激光工程，2000．4，pp26—30
[42] 边肇祺，张学工，模式识别，清华大学出版社，1999
[2] Mu-Song Chen and Michael T.Manry, Power Series Analyses of Back-Propagation Neural Networks, IEEE, 1991, Ⅰ-295—Ⅰ-300
[3] Mu-Song Chen and Michael T.Manry, Back-Propagation Representation Theorem Using Power Series, International Joint Conf on Neural Networks, IJCNN, Vol Ⅰ, 1990, Ⅰ-643—Ⅰ-648
[4] A.S.Pandya and R.Szabo, A fast learning algorithm for neural network application, Proceeding of IEEE conference on SMC, Oct, 1991, pp 1569-1573
[5] G.Govind and P.A.Ramamoorthy, Multi-layered neural networks and Volterra Series: The missing link, In Proc. Int. Conf. Systems Engineering, Pittsburgh, PA, Aug, 1990
[6] G.W.Davis and M.L.Gasperi, ANN Modeling of Volterra systems, In Proc. IJCNN'91, Seattle WA, pp Ⅱ 727-734
[7] S.A.Barton, A matrix method for optimizing a neural networks, Neural computation, Vol 3, No.3, 1991, pp450-459
[8] P.Gallinari, S.Thiria, and F.Fogelman Soulie, Multilayer perceptrons and data analysis, in Proc. IJCNN, Vol 1, 1988, pp391-399
[9] W.Schmidt, A non-iterative method for training feed forward networks, in Proc. IJCNN, Vol Ⅱ, 1991, pp 19-24
[10] Xian pingjiang, Analysis Optimization of Neural Networks for Remote Sensing, Remote Sensing Reviews, 1994
[11] Jonathan Lee, A Neural Network Approach to Cloud Classification, IEEE, Transactions on Geoscience and Remote Sensing, VOL 28, NO 5, Sep, 1990, pp846-855
[12] J.P.Fitch, Ship Wake-Detection Procedure Using Conjugate Gradient Trained Artifical Neural Networks, IEEE, Transactions on Geoscience and Remote
Sensing, VOL 29, NO 5, Sep, 1991, pp718-726
[13] Philip D.Heermann, Classification of Multispectral Remote Sensing Data Using a Back-Propagation Neural Networks, IEEE, Transactions on Geoscience and Remote Sensing, VOL 30, NO 1, JAN, 1992, pp81-88
[14] H. Bischof, Multispectral Classification of Landsat-Images Using Neural Networks, IEEE, Transactions on Geoscience and Remote Sensing, VOL 30, NO 3, MAY, 1992, pp482-490
[15] Classification of Mutltispectral Image Data by the Binary Diamond Neural Networks and by Nonparametric, Pixel by Pixel methods, IEEE, Transactions on Geoscience and Remote Sensing, VOL 31, NO 3, MAY, 1993, pp606-617
[16] Daniel T. Davis, Retrieval of Snow Parameters by Iterative Inversion of Neural Networks IEEE, Transactions on Geoscience and Remote Sensing, VOL 31, NO 4, JULY, 1993, pp842-852
[17] Yu Chang Tzeng, A Dynamic Learning Neural Networks for Remote Sensing Applications, IEEE, Transactions on Geoscience and Remote Sensing, VOL 32, NO 5, SEPT, 1994, pp 1096-1102
[18] Sucharita Gopal, Remote Sensing of Forest Change Using Artifical Neural Networks, IEEE, Transactions on Geoscience and Remote Sensing, VOL 34, NO 2, MARCH, 1996, pp398-402
[19] Li Li, Microwave Radiometric Technique to Retrieve Vapor, Liquid and Ice, Part Ⅰ—Development of a Neural Network-Based Inversion methods, IEEE, Transactions on Geoscience and Remote Sensing, VOL 35, NO 2, MARCH, 1997, pp224-236
[20] Dimitris Tsintikidis, A Neural Network Approach to Estimating Rainfall from Spaceborne Microwave Data, IEEE, Transactions on Geoscience and Remote Sensing, VOL 35, NO 5, SEPT, 1997, pp1079-1093
[21] Laurent Ferro-Famil, Unsupervised Classification of Multifrequency and Fully Polarimetric SAR Images Based on the H/A/Alpha—Wishart Classifier, IEEE, Transactions on Geoscience and Remote Sensing, VOL 39, NO 11, NOV, 2001, pp2332-2341
[22] Jong-Sen Lee, Unsupervised Classification Using Polarimetric Decomposition and the Complex Wishart Classifier, IEEE, Transactions on Geoscience and Remote Sensing, VOL 37, NO 5, SEP, 1999, pp2249-2258
[23] Jiancheng Shi and Jeff Dozier, Estimation of Snow Water Equivalence Using SIR-C/X-SAR, Part Ⅱ: Interring Snow Depth and Particle Size, IEEE, Transactions on Geoscience and Remote Sensing, VOL 38, NO 6, NOV, 2000, pp2475-2487
[24] Yisok Oh, Kamal Sarabandi, An Empirical Model and an Inversion Technique for Radar Scattering from Bare Soil Surfaces, IEEE, Transactions on Geoscience and Remote Sensing, 1992
[25] Pascale C. Dubois, Jakob van Zyl, Measuring Soil Moisture with Imaging Radars, IEEE, Transactions on Geoscience and Remote Sensing, 1995, pp915-925
[26] 郭华东等，雷达对地观测理论与应用，科学出版社，2000
[27] 郭华东等，中国雷达遥感图像分析，科学出版社，1999
[28] 周成虎等，遥感影像地学理解与分析，科学出版社，1999
[29] 容观澳等，计算机图像处理，清华大学出版社，1998
[30] 舒宁，微波遥感原理，武汉大学出版社，2003
[31] 章孝灿，黄智才，遥感数字图像处理，浙江大学出版社，1997
[32] [美]S．Weisberg，应用线性回归，中国统计出版社，1998
[33] 吴喜之，现代回归模型诊断，中国统计出版社，2003
[34] 李小文，先验知识在遥感反演中的作用，中国科学，D辑，1998，28(1)67—72
[35] 杨劲松，合成孔径雷达图像的近海面风场的反演，遥感学报，VOL 5，NO 1，JAN，2000．1，pp13—16
[36] 陆桂华，遗传算法在马斯京根模型参数估计中的应用，何海大学学报，VOL29，NO 4，JUL，2001．7，pp9—12
[37] 张佳华，利用遥感反演的页面指数研究中国东部的生态系统对东亚季风的影响，自然科学进展，2002．10，pp1098—1100
[38] 李国胜，东海真成光深度的遥感反演与影响机理研究，自然科学进展，2003．1，pp90—94
[39] 王耀南，小波神经网络的遥感图像分类，中国图像图形学报，1999．5，pp368—371
[40] 孙丹峰，自组织神经网络在遥感土地覆盖分类中的应用，遥感学报，1999．5，pp139—143
[41] 陈华，陈书海，K-means算法在遥感分类中的应用，红外与激光工程，2000．4，pp26—30
[42] 边肇祺，张学工，模式识别，清华大学出版社，1999