A Neural Network Method for Retrieval of Land Surface Temperature and Emissivity from ASTER1B Data

摘要：

　　The accuracy of RM (radiance transfer model)-NN (neural network) for separating land surface temperature (LST) and emissivity from AST09 (the ASTER Standard Data Product,Surface leaving radiance) is very high,but it is limited by the accuracy of atmospheric correction.This paper continues to use the neural network and radiance transfer model (MODTRAN4) to directly retrieve land surface temperature and emissivity from ASTER1B data,which overcomes the difficulty of atmospheric correction in previous methods.The retrieval average accuracy of land surface temperature is about 1.1 K,and the average accuracy of emissivity in bands 11~14 is under 0.016 for simulated data when the input nodes are the combination of brightness temperature in bands 11~14.The average accuracy of land surface temperature is under 0.8 K when the input nodes are the combination of water vapour content and brightness temperature in bands 11~14.Finally,the comparison of retrieval results with ground measurement data indicates that the RM-NN can be used to accurately retrieve land surface temperature and emissivity from ASTER1B data.

关键词： Land surface temperatue emissivity Neural network ASTER data

作者: MAO Kebiao LI Sanmei WANG Daolong ZHANG Lixin TANG Huajun WANG Xiufeng

作者单位: Key Laboratory of Resources Remote Sensing and Digital Agriculture,MOA,Key Laboratory of Dryland Far National Satellite Meteorological Center,CMA,Beijing,100081,China Key Laboratory of Resources Remote Sensing and Digital Agriculture,MOA,Key Laboratory of Dryland Far School of Geography,Beijing Normal Universtiy,Beijing,100875,China Graduate School of Agriculture,Hokkaido University,N-9,W-9,Kita-ku,Sapporo 060-8589,Japan

会议类型: 国际会议

会议名称: Woekshop on Agricultural Land Use and its Effect in APEC Member Economies(APEC地区农业土地利用及其影响国际研讨会)

会议地点: 北京

会议语种:英文

页码: 47-57

在线出版日期: 2009-10-19（万方平台首次上网日期，不代表论文的发表时间）