利用随机森林回归进行极化SAR土壤水分反演
|
摘要
全极化合成孔径雷达影像能够提供地物丰富的极化信息,挖掘这些信息在地表参数反演中的作用是目前相关领域的研究趋势之一。针对冬小麦区域的不同植被覆盖情况,利用随机森林回归对常用极化特征在土壤水分反演中的重要性进行评估,并在此基础上进行特征选择,挑选优化的极化特征组合,构建了高精度的土壤水分反演模型。实验结果显示,由重要性评分较高的极化特征所组成的反演模型能得到均方根误差(root mean square error, RMSE)小于6%的反演精度,比只输入传统线极化后向散射系数的模型在不同时相、不同数据集的精度都有所提高。与支持向量回归和人工神经网络模型进行比较,利用随机森林回归进行重要性评分与土壤水分反演的效果更好。
Soil moisture has great significance in the researches of hydrology, meteorology and agriculture yield estimation. The quad-polarimetric SAR images can provide a lot of polarimetric features, the significance of the features in surface parameter retrieval have attracted attentions in previous researches with no final conclusions because of the complexity of terrain scattering. In this paper, random forest regression(RFR) is used for both soil moisture retrieval and the importance evaluation of polarimetric features of Radarsat-2 images in winter wheat fields. According to the score of importance, feature selection and combination are done for modelling. We evaluate the retrieval accuracy of models with different feature combinations. The results show that models of important features selected by RFR have RMSE(root mean square error) less than 6% which are better results compared to traditional models; when compared with support vector regression and artifical neural networks, the RFR also shows best retrieval accuracies, which proves that RFR is suitable for soil moisture retrieval and feature selection. The high retrieval accuracies of LBC-CPD(linear backscatter coefficients-Cloude-Pottier decomposition) and LBC-CPR(linear backscatter coefficients-circular polarimetric ra-tio) indicates these features can improve the retrieval accuracy of soil moisture.
Soil moisture has great significance in the researches of hydrology, meteorology and agriculture yield estimation. The quad-polarimetric SAR images can provide a lot of polarimetric features, the significance of the features in surface parameter retrieval have attracted attentions in previous researches with no final conclusions because of the complexity of terrain scattering. In this paper, random forest regression(RFR) is used for both soil moisture retrieval and the importance evaluation of polarimetric features of Radarsat-2 images in winter wheat fields. According to the score of importance, feature selection and combination are done for modelling. We evaluate the retrieval accuracy of models with different feature combinations. The results show that models of important features selected by RFR have RMSE(root mean square error) less than 6% which are better results compared to traditional models; when compared with support vector regression and artifical neural networks, the RFR also shows best retrieval accuracies, which proves that RFR is suitable for soil moisture retrieval and feature selection. The high retrieval accuracies of LBC-CPD(linear backscatter coefficients-Cloude-Pottier decomposition) and LBC-CPR(linear backscatter coefficients-circular polarimetric ra-tio) indicates these features can improve the retrieval accuracy of soil moisture.
引文
[1] Ren Xin. A Surface Moisture Inversion Teclmique Using Multi-Polarization and Multi-Angle Radar Images[D]. Beijing:Institute of Remote Sensing Application, Chinese Academy of Sciences, 2003 (任鑫. 多极化多角度SAR土壤水分反演算法研究[D]. 北京:中国科学院遥感应用研究所, 2003)
[2] Wei Xiaolan, Li Zhen, Chen Quan. The Simulation Analysis and Validation of Soil Moisture Retrieval Using S-band Radar[J]. Geo-information Science, 2008, 10(1):97-101 (魏小兰, 李震, 陈权. S波段雷达数据反演土壤水分的模拟分析和验证[J]. 地球信息科学学报, 2008, 10(1):97-101)
[3] Bourgeau-Chavez L L, Leblon B, Charbonneau F, et al. Evaluation of Polarimetric Radarsat-2 SAR Data for Development of Soil Moisture Retrieval Algorithms over a Chronosequence of Black Spruce Boreal Forests[J]. Remote Sensing of Environment, 2013, 132(1): 71-85
[4] Wiseman G, McNairn H, Homayouni S, et al. Radarsat-2 Polarimetric SAR Response to Crop Biomass for Agricultural Production Monitoring[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2014,7(11): 4 461-4 471
[5] Adams J R, Berg A A, McNairn H, et al. Sensitivity of C-band SAR Polarimetric Variables to Unvegetated Agricultural Fields[J]. Canadian Journal of Remote Sensing, 2013, 39(1): 1-16
[6] Baghdadi N, Dubois-Fernandez P, Dupuis X, et al. Sensitivity of Main Polarimetric Parameters of Multifrequency Polarimetric SAR Data to Soil Moisture and Surface Roughness over Bare Agricultural Soils[J]. IEEE Geoscience and Remote Sensing Letters, 2013, 10(4): 731-735
[7] Cloude S R, Pottier E. An Entropy Classification Scheme for Land Applications of Polarimetric SAR Data[J]. IEEE Transactions on Geoscience and Remote Sensing, 1997, 35: 68-78
[8] Notarnicola C, Angiulli M, Posa F. Soil Moisture Retrieval from Remotely Sensed Data: Neural Network Approach Versus Bayesian Method[J]. IEEE Transactions on Geoscience and Remote Sensing, 2008, 46(2): 547-557
[9] Ahmad S, Kalra A, Stephen H. Estimating Soil Moisture Using Remote Sensing Data: A Machine Learning Approach[J]. Advances in Water Resources, 2010, 33(1): 69-80
[10] Breiman L. Random Forest[J]. Machine Learning, 2001, 45(1): 5-32
[11] Baghdadi N, Cresson R, El-Hajj M, et al. Estimation of Soil Parameters over Bare Agriculture Areas from C-band Polarimetric SAR Data Using Neural Networks[J]. Hydrology and Earth System Scien-ces, 2012, 16: 1 607-1 621
[12] Pasolli L, Notarnicola C, Bruzzone L, et al. Polarimetric Radarsat-2 Imagery for Soil Moisture Rtrieval in Alpine Areas[J]. Canadian Journal of Remote Sensing, 2011, 37: 535-547
[13] Srivastava P K, Han D, Ramirez M R, et al. Machine Learning Techniques for Downscaling SMOS Satellite Soil Moisture Using MODIS Land Surface Temperature for Hydrological Application[J]. Water Resources Management, 2013, 27: 3 127-3 144
[14] Karjalainen M, Kankare V, Vastaranta M, et al. Prediction of Plot-Level Forest Variables Using TerraSAR-X Stereo SAR Data[J]. Remote Sensing of Environment, 2012, 117: 338-347
[15] Baghdadi N, Cerden O, Zribi M, et al. Operational Performance of Current Synthetic Aperture Radar Sensors in Mapping Soil Surface Characteristics in Agricultural Environments: Application to Hydrological and Erosion Modelling[J]. Hydrological Processes, 2008, 22: 9-20
[16] Yang Guijun, Shi Yuechan, Zhao Chunjiang, et al. Estimation of Soil Moisture from Multi-Polarized SAR Data over Wheat Coverage Areas[C]. The First International Conference on Agro-Geoinformatics, Shanghai, China, 2012
[17] 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
[18] Krogager E. A New Decomposition of the Radar Target Scattering Matrix[J]. Electronics Letter, 1990, 26(18): 1 525-1 526
[19] Srivastava P, Han D, Ramirez M R, et al. Machine Learning Techniques for Downscaling SMOS Satellite Soil Moisture Using MODIS Land Surface Temperature for Hydrological Application[J]. Water Resource Management, 2013, 27: 3 127-3 144
[20] Shataee S, Kalbi S, Fallah A, et al. Forest Attri-bute Imputation Using Machine-Learning Methods and ASTER Data: Comparison of K-NN, SVR and Random Forest Regression Algorithms[J]. International Journal of Remote Sensing, 2012, 33: 6 254-6 280
[2] Wei Xiaolan, Li Zhen, Chen Quan. The Simulation Analysis and Validation of Soil Moisture Retrieval Using S-band Radar[J]. Geo-information Science, 2008, 10(1):97-101 (魏小兰, 李震, 陈权. S波段雷达数据反演土壤水分的模拟分析和验证[J]. 地球信息科学学报, 2008, 10(1):97-101)
[3] Bourgeau-Chavez L L, Leblon B, Charbonneau F, et al. Evaluation of Polarimetric Radarsat-2 SAR Data for Development of Soil Moisture Retrieval Algorithms over a Chronosequence of Black Spruce Boreal Forests[J]. Remote Sensing of Environment, 2013, 132(1): 71-85
[4] Wiseman G, McNairn H, Homayouni S, et al. Radarsat-2 Polarimetric SAR Response to Crop Biomass for Agricultural Production Monitoring[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2014,7(11): 4 461-4 471
[5] Adams J R, Berg A A, McNairn H, et al. Sensitivity of C-band SAR Polarimetric Variables to Unvegetated Agricultural Fields[J]. Canadian Journal of Remote Sensing, 2013, 39(1): 1-16
[6] Baghdadi N, Dubois-Fernandez P, Dupuis X, et al. Sensitivity of Main Polarimetric Parameters of Multifrequency Polarimetric SAR Data to Soil Moisture and Surface Roughness over Bare Agricultural Soils[J]. IEEE Geoscience and Remote Sensing Letters, 2013, 10(4): 731-735
[7] Cloude S R, Pottier E. An Entropy Classification Scheme for Land Applications of Polarimetric SAR Data[J]. IEEE Transactions on Geoscience and Remote Sensing, 1997, 35: 68-78
[8] Notarnicola C, Angiulli M, Posa F. Soil Moisture Retrieval from Remotely Sensed Data: Neural Network Approach Versus Bayesian Method[J]. IEEE Transactions on Geoscience and Remote Sensing, 2008, 46(2): 547-557
[9] Ahmad S, Kalra A, Stephen H. Estimating Soil Moisture Using Remote Sensing Data: A Machine Learning Approach[J]. Advances in Water Resources, 2010, 33(1): 69-80
[10] Breiman L. Random Forest[J]. Machine Learning, 2001, 45(1): 5-32
[11] Baghdadi N, Cresson R, El-Hajj M, et al. Estimation of Soil Parameters over Bare Agriculture Areas from C-band Polarimetric SAR Data Using Neural Networks[J]. Hydrology and Earth System Scien-ces, 2012, 16: 1 607-1 621
[12] Pasolli L, Notarnicola C, Bruzzone L, et al. Polarimetric Radarsat-2 Imagery for Soil Moisture Rtrieval in Alpine Areas[J]. Canadian Journal of Remote Sensing, 2011, 37: 535-547
[13] Srivastava P K, Han D, Ramirez M R, et al. Machine Learning Techniques for Downscaling SMOS Satellite Soil Moisture Using MODIS Land Surface Temperature for Hydrological Application[J]. Water Resources Management, 2013, 27: 3 127-3 144
[14] Karjalainen M, Kankare V, Vastaranta M, et al. Prediction of Plot-Level Forest Variables Using TerraSAR-X Stereo SAR Data[J]. Remote Sensing of Environment, 2012, 117: 338-347
[15] Baghdadi N, Cerden O, Zribi M, et al. Operational Performance of Current Synthetic Aperture Radar Sensors in Mapping Soil Surface Characteristics in Agricultural Environments: Application to Hydrological and Erosion Modelling[J]. Hydrological Processes, 2008, 22: 9-20
[16] Yang Guijun, Shi Yuechan, Zhao Chunjiang, et al. Estimation of Soil Moisture from Multi-Polarized SAR Data over Wheat Coverage Areas[C]. The First International Conference on Agro-Geoinformatics, Shanghai, China, 2012
[17] 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
[18] Krogager E. A New Decomposition of the Radar Target Scattering Matrix[J]. Electronics Letter, 1990, 26(18): 1 525-1 526
[19] Srivastava P, Han D, Ramirez M R, et al. Machine Learning Techniques for Downscaling SMOS Satellite Soil Moisture Using MODIS Land Surface Temperature for Hydrological Application[J]. Water Resource Management, 2013, 27: 3 127-3 144
[20] Shataee S, Kalbi S, Fallah A, et al. Forest Attri-bute Imputation Using Machine-Learning Methods and ASTER Data: Comparison of K-NN, SVR and Random Forest Regression Algorithms[J]. International Journal of Remote Sensing, 2012, 33: 6 254-6 280
© 2004-2018 中国地质图书馆版权所有 京ICP备05064691号 京公网安备11010802017129号 地址：北京市海淀区学院路29号 邮编：100083 电话：办公室：(+86 10)66554848；文献借阅、咨询服务、科技查新：66554700 |