|
|
- 2018
气溶胶光学厚度估测中的LASSO特征选择方法
|
Abstract:
气溶胶光学厚度估测中通常利用遥感信息构造的多种特征属性作为输入，然而，这些属性中常常存在数据噪音、相互关联性和缺失值，从而降低了估测精度和估测强健性。针对这个问题，基于最小绝对收缩和选择算子（least absolute shrinkage selection operator，LASSO）方法和气溶胶光学厚度反演的先验知识，提出了一种针对遥感卫星观测的高维数据进行特征选择的方法，利用2009年4月2日至2011年4月1日2 a间与全球197个气溶胶地基自动观测网站点时空同步的MODIS（moderate-resolution imaging spectroradiometer）遥感数据，采用常用的人工神经网络作为估测模型进行实验分析，表明该方法能结合反演先验知识对多种异质遥感属性进行分组，通过组间迭代保留关键特征，去除冗余属性，有效进行特征选择，从而显著提高气溶胶光学厚度的估测精度
| [1] | Luo Y, Zheng X, Zhao T, et al.A Climatology of Aerosol Optical Depth over China from Recent 10 Years of MODIS Remote Sensing Data[J]. <em>International Journal of Climatology,</em> 2014, 34(3):863-870 |
| [2] | Kaufman Y J, Gobron N, Pinty B. Relationship Between Surface Reflectance in the Visible and Mid-IR Used in MODIS Aerosol Algorithm-Theory[J]. <em>Geophysical Research Letters,</em> 2002, 29(23):311-314 |
| [3] | Jun G, Ghosh J, Radosavljevic V. Predicting Ground-Based Aerosol Optical Depth with Satellite Images via Gaussian Process[C]. International Conference on Knowledge Discovery and Information Retrieval (KDIR), Valencia, Spain, 2010 |
| [4] | Tibshirani R. Regression Shrinkage and Selection via the Lasso[J].<em>Journal of the Royal Statistical Society-Methodological,</em> 1996, 58(1):267-288 |
| [5] | Breider T, Mickley L, Jacob D, et al.Annual Distributions and Sources of Arctic Aerosol Components, Aerosol Optical Depth, and Aerosol Absorption[J]. <em>Journal of Geophysical Research,</em> 2014, 119(7):4107-4124 |
| [6] | Kondratev K I, Buznikov A A, Vasilev O B, et al. Influence of the Atmosphere on the Spectral Brightnesses and Contrasts of Natural Formations in Spectrophotometry of the Earth from Space[J].<em>Izvestiya Atmospheric and Oceanic Physics,</em> 1975(11):348-361 |
| [7] | Liang S L, Zhong B, Fang H L. Improved Estimation of Aerosol Optical Depth from MODIS Imagery over Land Surfaces[J].<em>Remote Sensing of Environment,</em> 2006(104):416-425 |
| [8] | Vucetic S, Han B, Mi W, et al. A Data Mining Approach for the Validation of Aerosol Retrievals[J]. <em>IEEE Geoscience and Remote Sensing Letters,</em> 2008, 5(1):113-117 |
| [9] | Chu D A, Kaufman Y J, Ichoku C, et al. Validation of MODIS Aerosol Optical Depth Retrieval over Land[J]. <em>Geophysical Research Letters,</em> 2002, 29(12):MOD2-1-MOD2-4 |
| [10] | Remer L A, Kaufman Y J, Tanré D, et al. The MODIS Aerosol Algorithm, Products, and Validation[J]. <em>Journal of the Atmospheric Sciences,</em> 2005, 62(4):947-973 |
| [11] | Levy R C, Remer L A, Dubovik O. Global Aerosol Optical Properties and Application to Moderate Resolution Imaging Spectroradiometer Aerosol Retrieval over Land[J]. <em>Journal of Geophysical Research:Atmospheres,</em> 2007(112):3710-3711 |
| [12] | Levy R C, Remer L A, Kleidman R G, et al. Global Evaluation of the Collection 5 MODIS Dark-Target Aerosol Products over Land[J]. <em>Atmospheric Chemistry and Physics,</em> 2010, 10(21):10399-10420 |
| [13] | Han B, Gao X, Cui X. Model-Based Sensitivity Analysis on Aerosol Optical Thickness Prediction[J]. <em>International Journal of Distributed Sensor Networks,</em>2015(1):1-9 |
| [14] | Belloni A, Chernozhukov V, Wang L. Pivotal Estimation via Square-Root LASSO in Nonparametric Regression[J].<em>The Annals of Statistics,</em> 2014, 42(2):757-788 |
| [15] | Rumelhart D E, Hinton G E, Williams R J. Learning Representations by Back-Propagating Errors[J]. <em>Cognitive Modeling,</em> 1986, 5(3):533-536 |
| [16] | Ichoku C, Chu D A, Mattoo S, et al. A Spatio-Temporal Approach for Global Validation and Analysis of MODIS Aerosol Products[J]. <em>Geophysical Research Letters,</em> 2002, 29(12):MOD1-1-MOD1-4 |
