基于Landsat TM影像不同地表温度反演算法的比较

Comparison of Different Land Surface Temperature Algorithms Based on Landsat TM Images

地表温度是反映地表环境的一个重要参数,精确获取地表温度的方法对研究城市热岛效应、进行生态环境监测评价是必不可少的。利用遥感手段进行地表温度的反演是一种较新颖的方法,相对成熟的反演算法主要有单通道算法、辐射传输方程法及单窗算法。本文以山东省威海市Landsat TM数据为例,分别利用辐射传输方程法和单窗算法进行地表温度反演,通过遥感目视解译的方法提取研究区各类典型地物对应的地表温度,进行统计分析,结果表明,两者反演得到的地表温度具有一致的变化趋势,其中单窗算法中不同典型地物的地表温度整体标准差较小,如建设用地为0.530,起伏波动小,算法精度略优于辐射传输方程法。

The land surface temperature is an important parameter to reflect the land surface environment. It is essential for studying the urban heat island effect and making the monitoring and evaluation of the ecological environment to get accurately the surface temperature. It is a new method to make the inversion of the land surface temperature by using the remote sensing technology. Relatively mature inversion algorithms mainly include the single-channel algorithm, the radiative transfer equation algorithm and the mono-window algorithm. Taking the Landsat TM data of Weihai city, Shandong province as an example, this paper respectively uses the radiative transfer equation algorithm and the mono-window algorithm for the inversion of the land surface temperature, extracts all kinds of typical objects in the study area to correspond to the land surface temperature by means of the visual interpretation, and makes statistical analysis. The results show that the land surface temperatures derived from these two inversions of the land surface temperature followed the same trends, and the mono-window algorithm inversion has a lower standard deviation of different typical object's land surface temperature. For example, the standard deviation of the construction land is 0.530. With less fluctuations, the algorithm precision is slightly better than the radiative transfer equation algorithm.