基于相干集的PolInSAR森林地区参数反演方法研究
|
摘要
PolInSAR(Polarimetric SAR interferometry,极化干涉合成孔径雷达)将极化SAR和干涉SAR技术结合在一起,将极化SAR获取目标散射体形状及散射特性的能力与干涉SAR获取目标散射体空间分布的能力集于一身。极化干涉技术能提高传统干涉SAR的测量精度,同时能够更好地获取目标散射机理和散射过程。因此,极化干涉技术在军事和民用领域都有着巨大的应用价值。
极化干涉SAR森林地区参数反演的研究是极化干涉SAR领域的一个重要分支。因为L和P波段SAR能够穿透森林表面直达地表面,因此能够获取其他遥感手段难以获取的森林参数,特别是森林植被高度参数和森林地表参数,这些参数为林地监测和环境监控提供了依据。
论文主要研究了基于相干集的PolInSAR森林地区参数反演算法。主要内容有:极化干涉技术分析,RVoG等相干散射模型,基于RVoG模型的森林参数反演法及基于相干集的森林参数反演方法。
论文首先对极化干涉系统进行了阐述,从极化SAR和干涉SAR的原理出发引出极化干涉的概念。然后,对三种森林中典型的相干散射模型进行介绍,重点分析了RVoG模型的原理和特点。
接下来,论文对PolInSAR森林参数反演算法进行了分析研究,研究了几种基于RVoG模型的PolInSAR森林区域参数反演法:三阶段法、ESPRIT法和相干幅度相位结合法。重点引入了相干集的概念,给出了相干集区域求解方法并总结了相干集的性质。通过结合相干集性质和RVoG模型,建立了一种联合森林参数反演算法。
最后,论文通过仿真数据和SIR-C真实数据提取森林地区参数并比较了几种反演算法的优劣,并分析了产生反演能力差异的原因。通过两组数据的反演精度分析对各算法进行分析,实验结果表明,使用相干集性质结合RVoG模型的方法具有一定的有效性和准确性。
Polarimetric SAR interferometry (PolInSAR) integrates Polarimetric SAR technology (PolSAR) and interferometric SAR (InSAR) technology together. It has the capability of acquiring target scattering structure and target spatial distribution. PolInSAR can help InSAR to improve height measurements and help PolSAR to improve interpretation of scattering mechanism. Therefore, PolInSAR has been used in military use and civilian use.
PolInSAR has been widely used in inverting parameters in forest area on account of the ability that L- or P-band SAR can penetrate vegetation crown. So it can be used to invert vegetation parameters in forest area, especially for vegetation height and terrain parameters. These parameters are the bases of forest monitor.
Parameters inversion in forest area using PolInSAR by coherence set is presented in this dissertation. The key contents include basic theory of PolInSAR technology, different coherence scattering models; parameters inversion in forest area by RVoG (Random Volume over Ground) model and coherence set.
Firstly, this dissertation recapitulates on the theory of PolSAR and InSAR,then the theory of PolInSAR is introduced. Furthermore, the representing of vector coherence scattering models are discussed, especially the theory and property of RVoG model.
Secondly, basic theory of coherence set used in the inversion procedure is discussed. After analyzing the parameters inversion in forest area by RVoG model, three detailed inversion algorithms (three-stage method, ESPRIT method and combination of coherence amplitude and phase method) are presented. Then principle of coherence set is introduced in forest parameters inversion. By combining the property of coherence set and the RVoG model, a novel parameters retrieve algorithm is presented.
Finally, simulated image and SIR-C (Spaceborne Imaging Radar-C/X-band Synthetic Aperture Radar) image of forestry area are used to validate the proposed algorithm. Performances of every inversion algorithm are compared. Results show that the inversion method used by coherence set presented is valid to a certain extent.
极化干涉SAR森林地区参数反演的研究是极化干涉SAR领域的一个重要分支。因为L和P波段SAR能够穿透森林表面直达地表面,因此能够获取其他遥感手段难以获取的森林参数,特别是森林植被高度参数和森林地表参数,这些参数为林地监测和环境监控提供了依据。
论文主要研究了基于相干集的PolInSAR森林地区参数反演算法。主要内容有:极化干涉技术分析,RVoG等相干散射模型,基于RVoG模型的森林参数反演法及基于相干集的森林参数反演方法。
论文首先对极化干涉系统进行了阐述,从极化SAR和干涉SAR的原理出发引出极化干涉的概念。然后,对三种森林中典型的相干散射模型进行介绍,重点分析了RVoG模型的原理和特点。
接下来,论文对PolInSAR森林参数反演算法进行了分析研究,研究了几种基于RVoG模型的PolInSAR森林区域参数反演法:三阶段法、ESPRIT法和相干幅度相位结合法。重点引入了相干集的概念,给出了相干集区域求解方法并总结了相干集的性质。通过结合相干集性质和RVoG模型,建立了一种联合森林参数反演算法。
最后,论文通过仿真数据和SIR-C真实数据提取森林地区参数并比较了几种反演算法的优劣,并分析了产生反演能力差异的原因。通过两组数据的反演精度分析对各算法进行分析,实验结果表明,使用相干集性质结合RVoG模型的方法具有一定的有效性和准确性。
Polarimetric SAR interferometry (PolInSAR) integrates Polarimetric SAR technology (PolSAR) and interferometric SAR (InSAR) technology together. It has the capability of acquiring target scattering structure and target spatial distribution. PolInSAR can help InSAR to improve height measurements and help PolSAR to improve interpretation of scattering mechanism. Therefore, PolInSAR has been used in military use and civilian use.
PolInSAR has been widely used in inverting parameters in forest area on account of the ability that L- or P-band SAR can penetrate vegetation crown. So it can be used to invert vegetation parameters in forest area, especially for vegetation height and terrain parameters. These parameters are the bases of forest monitor.
Parameters inversion in forest area using PolInSAR by coherence set is presented in this dissertation. The key contents include basic theory of PolInSAR technology, different coherence scattering models; parameters inversion in forest area by RVoG (Random Volume over Ground) model and coherence set.
Firstly, this dissertation recapitulates on the theory of PolSAR and InSAR,then the theory of PolInSAR is introduced. Furthermore, the representing of vector coherence scattering models are discussed, especially the theory and property of RVoG model.
Secondly, basic theory of coherence set used in the inversion procedure is discussed. After analyzing the parameters inversion in forest area by RVoG model, three detailed inversion algorithms (three-stage method, ESPRIT method and combination of coherence amplitude and phase method) are presented. Then principle of coherence set is introduced in forest parameters inversion. By combining the property of coherence set and the RVoG model, a novel parameters retrieve algorithm is presented.
Finally, simulated image and SIR-C (Spaceborne Imaging Radar-C/X-band Synthetic Aperture Radar) image of forestry area are used to validate the proposed algorithm. Performances of every inversion algorithm are compared. Results show that the inversion method used by coherence set presented is valid to a certain extent.
引文
[1] Cloude S R, Papathanassiou K P. Polarimetric SAR Interferometry[J]. IEEE Transactions on Geoscience and Remote Sensing, 1998, 36(5):1551-1565.
[2] Papathanassiou K P, Reiger A, Cloude S R. Vegetation and Ground Parameter Estimation using Polarimetric Interferometry Part I: The Role of Polarisation. Part II: Parameter Inversion and Optimal Polarisations[C]. Proc. of the CEOS SAR Workshop, Toulouse, 26-29 Oct. 1999:347~358.
[3] Papathanassiou K P, Cloude S R. Single-baseline Polarimetric SAR Interferometry[J]. IEEE Transactions on Geoscience and Remote Sensing, 2001, 39(11):2352-2363.
[4] Cloude S R, Papathanassiou K P. A 3-Stage Inversion Process for Polarimetric SAR Interferometry[C]. Proc. of EUSAR’02, Cologne, Germany, 4-6 June 2002:297-282.
[5] Cloude S R, Papathanassiou K P. Three-stage Inversion Process for Polarimetric SAR Interferometry[J]. IEE Proc. Radar Sonar Navig, 2003, 150(3):125-134.
[6] Lopez-Sanchez J M, Ballester-Berman J D, Marquez-Moreno Y. Model limitations and parameter estimationmethods for agricultural applications of polarimetric SAR interferometry[J], IEEE Transactions on Geoscience and Remote Sensing, 2007, 45(11),3481-3493.
[7] López-Martínez C, Alonso, A, Fàbregas X, Papathannassiou K P. Ground Topography Estimation over Forests Considering Polarimetric SAR Interferometry[C]. Proc. of IGARSS’10, Honolulu, 25-30 July 2010, 3612-3615.
[8] Cloude S R. Polarization Coherence Tomography[J]. Radio Science, 2006, 41:RS4017.
[9] Cloude S R. Polarisation Applications in Remote Sensing[M]. Oxford University Press, 2009.
[10] Flynn T, Tabb M, Carande R. Direction Estimation of Vegetation Parameters from Covariance Data in Polarimetric SAR Interferometry[C].Proc. of IGARSS’02, Toronto, Canada, 24-28 June 2002, 3:1908-1910.
[11] Tabb M, Flynn T, Carande R. Full Maximum Likelihood Inversion of PolInSAR Scattering Models[C]. Proc. of IGARSS’04, Anchorage, Alaska, 20-24 Sept. 2004, 2:1232-1235.
[12] Tabb M, Orrey J, Flynn T, et al. Phase Diversity: A Decomposition for Vegetation Parameter Estimation Using Polarimetric SAR Interferometry[C], Proc. EUSAR 2002:721-724.
[13] Flynn T, Tabb M, Richard Carande. Coherence Region Shape Extraction for Vegetation Parameter Estimation in Polarimetric SAR Interferometry[C]. Proc. of IGARSS’02, Toronto, Canada, 24-28 June 2002, 5:2596-2598.
[14] Neumann M, Reigber A, Ferro-Famil L. PolInSAR Coherence Set Theory and Application[C], EUSAR’2006, Dresden, May 2006.
[15] Neumann M, Ferro-Famil L, Reigber A. Estimation of Forest Structure, Ground, and Canopy Layer Characteristics from Multibaseline Polarimetric Interferometric SAR Data[J], IEEE Transactions on Geoscience and Remote Sensing, 2010, 48(3):1086-1104.
[16] Yamada H, Yamaguchi Y, Rodriguez E, Kim Y, et al. Polarimetric SAR Interferometry for Forest Canopy Analysis by Using The Super-resolution Method[C]. Proc. of IGARSS’01, Sydney, Australia, 9-13 July 2001, 3:1101-1103.
[17] Yamada H, Yamaguchi Y. Polarimetric SAR Interferometry for Forest Analysis Based on the ESPRIT Algorithm[J]. IEICE Trans. Electron, 2001, E84-C(12):1917~1924.
[18] Guillaso S, Reigber A, Ferro-Famil L. Evaluation of the ESPRIT Approach in Polarimetric Interferometric SAR[C]. Proc. of IGARSS’05, Seoul, Korea, 25-29 July 2005, 1:1~4.
[19]杨震,杨汝良.极化合成孔径雷达干涉技术[J].遥感技术与应用. 2001, 16(3):139~143
[20]邹斌,张腊梅,孙德明,王伟.极化干涉合成孔径雷达图像信息提取技术的进展及未来[J].电子与信息学报. 2006, 28(10):1979-1984
[21]李廷伟,梁甸农,朱炬波.极化干涉SAR森林高度反演综述[J],遥感信息2009, 3:85-91.
[22]陈曦,张红,王超.极化干涉SAR反演植被垂直结构剖面研究[J],国土资源遥感, 2009, 82:49-52.
[23]陈兵,张平.基于散射体模型的PolInSAR数据模拟反演研究[J],系统仿真学报, 2008, 20(8):2200-2202.
[24]韦顺军,张晓玲.基于分类统计的PolInSAR植被高度最大似然估计[J],现代雷达, 2009, 31(11):60-63.
[25]李新武,郭华东,李震,王长林.用SIR-C航天飞机双频极化干涉雷达估计植被高度的方法研究[J],高技术通讯, 2005, 15(7):79-84.
[26]于大洋,董贵威,杨健,彭应宁,王超,张红.基于干涉极化SAR数据的森林树高反演[J],清华大学学报(自然科学版), 2005, 45(3):334-336.
[27]张晓玲,韦顺军,韩迪.结合相干系数的极化干涉SAR植被高度估计方法研究[J],电子科技大学学报, 2009, 38(3):321-324.
[28]李哲,陈尔学,王建.几种极化干涉SAR森林平均高反演算法的比较评价[J],遥感技术与应用, 2009, 24(5):611-616
[29]罗环敏,陈尔学,程建,李小文.极化干涉SAR森林高度反演方法研究[J],遥感学报, 2010, 14(4):814-821.
[30]李廷伟,梁甸农,黄海风,朱炬波.一种基于BP神经网络的极化干涉SAR植被高度反演方法[J],国防科技大学学报, 2010, 32(3):60-64.
[31]陈兵,徐绍剑,张平.单基线PolInSAR反演算法研究[J],电子与信息学报, 2008, 30(7):1744-1746.
[32]谈璐璐,陈兵,杨汝良.利用POLInSAR数据反演植被高度的改进三阶段算法[J],系统仿真学报, 2010, 22(4):996-999.
[33]白璐,曹芳,洪文.相干区域长轴的快速估计方法及其应用[J],电子与信息学报, 2010, 32(3):548-553.
[34] Cloude S R. A Review of Target Decomposition Theorem in Radar Polarimetry[J], IEEE Transactions on Geoscience and Remote Sensing, 1996, 34(2):498-518
[35] Freeman A, Durden S L. A Three-Component Scattering Model for Polarimetric SAR Data[J], IEEE Transactions on Geoscience and Remote Sensing, 1998, 36(3):963-973
[36] Bamler R. Synthetic Aperture Radar Interferometry[J]. Inverse Problem 1998, 14(4):1~54.
[37] Treuhaft R N, Madsen S N, Moghaddam M, et al. VegetationCharacteristics and Surface Topography from Interferometric Radar[J]. Radio Science, 1996, 31:1449~1485.
[38] Treuhaft R N, Cloude S R. The Structure of Oriented Vegetation from Polarimetric Interferometry[J]. IEEE Transactions on Geoscience and Remote Sensing, 1999, 37(5):2620-2624.
[39] Treuhaft R N. Vertical Structure of Vegetated Land Surfaces from Interferometric and Polarimetric Radar[J]. Radio Sciences, 2000, 35(1):141-177.
[40] Williams M L. Forest Coherent SAR Simulator within PolSARPro: an Educational Toolbox for PolSAR and PolInSAR Data Processing[C]. Asian Conference on Remote Sensing, Kuala Lumpur, Malaysia, 2007.
[41] Yamada H, Okada H, Yamaguchi Y. Accuracy Improvement of ESPRIT based Polarimetric SAR Interferometry for Forest Height Estimation[C]. Proc. of IGARSS’05, Seoul, Korea, 25-29 July 2005, 6:4077-4080.
[2] Papathanassiou K P, Reiger A, Cloude S R. Vegetation and Ground Parameter Estimation using Polarimetric Interferometry Part I: The Role of Polarisation. Part II: Parameter Inversion and Optimal Polarisations[C]. Proc. of the CEOS SAR Workshop, Toulouse, 26-29 Oct. 1999:347~358.
[3] Papathanassiou K P, Cloude S R. Single-baseline Polarimetric SAR Interferometry[J]. IEEE Transactions on Geoscience and Remote Sensing, 2001, 39(11):2352-2363.
[4] Cloude S R, Papathanassiou K P. A 3-Stage Inversion Process for Polarimetric SAR Interferometry[C]. Proc. of EUSAR’02, Cologne, Germany, 4-6 June 2002:297-282.
[5] Cloude S R, Papathanassiou K P. Three-stage Inversion Process for Polarimetric SAR Interferometry[J]. IEE Proc. Radar Sonar Navig, 2003, 150(3):125-134.
[6] Lopez-Sanchez J M, Ballester-Berman J D, Marquez-Moreno Y. Model limitations and parameter estimationmethods for agricultural applications of polarimetric SAR interferometry[J], IEEE Transactions on Geoscience and Remote Sensing, 2007, 45(11),3481-3493.
[7] López-Martínez C, Alonso, A, Fàbregas X, Papathannassiou K P. Ground Topography Estimation over Forests Considering Polarimetric SAR Interferometry[C]. Proc. of IGARSS’10, Honolulu, 25-30 July 2010, 3612-3615.
[8] Cloude S R. Polarization Coherence Tomography[J]. Radio Science, 2006, 41:RS4017.
[9] Cloude S R. Polarisation Applications in Remote Sensing[M]. Oxford University Press, 2009.
[10] Flynn T, Tabb M, Carande R. Direction Estimation of Vegetation Parameters from Covariance Data in Polarimetric SAR Interferometry[C].Proc. of IGARSS’02, Toronto, Canada, 24-28 June 2002, 3:1908-1910.
[11] Tabb M, Flynn T, Carande R. Full Maximum Likelihood Inversion of PolInSAR Scattering Models[C]. Proc. of IGARSS’04, Anchorage, Alaska, 20-24 Sept. 2004, 2:1232-1235.
[12] Tabb M, Orrey J, Flynn T, et al. Phase Diversity: A Decomposition for Vegetation Parameter Estimation Using Polarimetric SAR Interferometry[C], Proc. EUSAR 2002:721-724.
[13] Flynn T, Tabb M, Richard Carande. Coherence Region Shape Extraction for Vegetation Parameter Estimation in Polarimetric SAR Interferometry[C]. Proc. of IGARSS’02, Toronto, Canada, 24-28 June 2002, 5:2596-2598.
[14] Neumann M, Reigber A, Ferro-Famil L. PolInSAR Coherence Set Theory and Application[C], EUSAR’2006, Dresden, May 2006.
[15] Neumann M, Ferro-Famil L, Reigber A. Estimation of Forest Structure, Ground, and Canopy Layer Characteristics from Multibaseline Polarimetric Interferometric SAR Data[J], IEEE Transactions on Geoscience and Remote Sensing, 2010, 48(3):1086-1104.
[16] Yamada H, Yamaguchi Y, Rodriguez E, Kim Y, et al. Polarimetric SAR Interferometry for Forest Canopy Analysis by Using The Super-resolution Method[C]. Proc. of IGARSS’01, Sydney, Australia, 9-13 July 2001, 3:1101-1103.
[17] Yamada H, Yamaguchi Y. Polarimetric SAR Interferometry for Forest Analysis Based on the ESPRIT Algorithm[J]. IEICE Trans. Electron, 2001, E84-C(12):1917~1924.
[18] Guillaso S, Reigber A, Ferro-Famil L. Evaluation of the ESPRIT Approach in Polarimetric Interferometric SAR[C]. Proc. of IGARSS’05, Seoul, Korea, 25-29 July 2005, 1:1~4.
[19]杨震,杨汝良.极化合成孔径雷达干涉技术[J].遥感技术与应用. 2001, 16(3):139~143
[20]邹斌,张腊梅,孙德明,王伟.极化干涉合成孔径雷达图像信息提取技术的进展及未来[J].电子与信息学报. 2006, 28(10):1979-1984
[21]李廷伟,梁甸农,朱炬波.极化干涉SAR森林高度反演综述[J],遥感信息2009, 3:85-91.
[22]陈曦,张红,王超.极化干涉SAR反演植被垂直结构剖面研究[J],国土资源遥感, 2009, 82:49-52.
[23]陈兵,张平.基于散射体模型的PolInSAR数据模拟反演研究[J],系统仿真学报, 2008, 20(8):2200-2202.
[24]韦顺军,张晓玲.基于分类统计的PolInSAR植被高度最大似然估计[J],现代雷达, 2009, 31(11):60-63.
[25]李新武,郭华东,李震,王长林.用SIR-C航天飞机双频极化干涉雷达估计植被高度的方法研究[J],高技术通讯, 2005, 15(7):79-84.
[26]于大洋,董贵威,杨健,彭应宁,王超,张红.基于干涉极化SAR数据的森林树高反演[J],清华大学学报(自然科学版), 2005, 45(3):334-336.
[27]张晓玲,韦顺军,韩迪.结合相干系数的极化干涉SAR植被高度估计方法研究[J],电子科技大学学报, 2009, 38(3):321-324.
[28]李哲,陈尔学,王建.几种极化干涉SAR森林平均高反演算法的比较评价[J],遥感技术与应用, 2009, 24(5):611-616
[29]罗环敏,陈尔学,程建,李小文.极化干涉SAR森林高度反演方法研究[J],遥感学报, 2010, 14(4):814-821.
[30]李廷伟,梁甸农,黄海风,朱炬波.一种基于BP神经网络的极化干涉SAR植被高度反演方法[J],国防科技大学学报, 2010, 32(3):60-64.
[31]陈兵,徐绍剑,张平.单基线PolInSAR反演算法研究[J],电子与信息学报, 2008, 30(7):1744-1746.
[32]谈璐璐,陈兵,杨汝良.利用POLInSAR数据反演植被高度的改进三阶段算法[J],系统仿真学报, 2010, 22(4):996-999.
[33]白璐,曹芳,洪文.相干区域长轴的快速估计方法及其应用[J],电子与信息学报, 2010, 32(3):548-553.
[34] Cloude S R. A Review of Target Decomposition Theorem in Radar Polarimetry[J], IEEE Transactions on Geoscience and Remote Sensing, 1996, 34(2):498-518
[35] Freeman A, Durden S L. A Three-Component Scattering Model for Polarimetric SAR Data[J], IEEE Transactions on Geoscience and Remote Sensing, 1998, 36(3):963-973
[36] Bamler R. Synthetic Aperture Radar Interferometry[J]. Inverse Problem 1998, 14(4):1~54.
[37] Treuhaft R N, Madsen S N, Moghaddam M, et al. VegetationCharacteristics and Surface Topography from Interferometric Radar[J]. Radio Science, 1996, 31:1449~1485.
[38] Treuhaft R N, Cloude S R. The Structure of Oriented Vegetation from Polarimetric Interferometry[J]. IEEE Transactions on Geoscience and Remote Sensing, 1999, 37(5):2620-2624.
[39] Treuhaft R N. Vertical Structure of Vegetated Land Surfaces from Interferometric and Polarimetric Radar[J]. Radio Sciences, 2000, 35(1):141-177.
[40] Williams M L. Forest Coherent SAR Simulator within PolSARPro: an Educational Toolbox for PolSAR and PolInSAR Data Processing[C]. Asian Conference on Remote Sensing, Kuala Lumpur, Malaysia, 2007.
[41] Yamada H, Okada H, Yamaguchi Y. Accuracy Improvement of ESPRIT based Polarimetric SAR Interferometry for Forest Height Estimation[C]. Proc. of IGARSS’05, Seoul, Korea, 25-29 July 2005, 6:4077-4080.