论文部分内容阅读
利用遥感数据可以对地球资源环境进行大面积连续监测,得到更为精确的研究结果,MODIS LST数据因其优化的时空分辨率成为较理想和常用的数据源。同时,遥感数据由于受到云、气溶胶以及传感器角度等影响均存在不同程度的噪声污染、数据缺失等问题。针对该现象,以河南省为研究区域,以MODIS LST数据为研究对象,利用谐波分析方法对河南省2011年全年每天四个时刻的MODIS LST时间序列数据进行重构。结果表明,利用该方法重构的数据可对MODIS缺值70%以上的影像进行弥补,并且60%以上影像误差可控制在3℃以内,能得到较好的重构结果;同时重构LST数据与相应气温数据相关性大部分在0.8左右,能够较好拟合LST的变化趋势。
Remote sensing data can be used to monitor the earth’s resources and environment on a large area continuously, and get more accurate research results. MODIS LST data has become an ideal and commonly used data source due to its optimized spatial and temporal resolution. At the same time, there are some problems such as noise pollution and data missing in the remote sensing data due to the cloud, aerosol and sensor angle. In view of this phenomenon, taking Henan Province as the research area and MODIS LST data as the research object, the MODIS LST time series data of Henan Province at four times a day in 2011 are reconstructed by the method of harmonic analysis. The results show that the data reconstructed by this method can make up more than 70% of the missing images in MODIS, and more than 60% of the image errors can be controlled within 3 ° C, and the better reconstructed results can be obtained; meanwhile reconstructing LST data Correlation with the corresponding temperature data most of the 0.8 or so, can better fit the trend of LST.