长时间序列遥感数据回归分析支持下的植被变化分析

The Vegetation Cover Changes of Regress Analysis on Using Time-serial Images of Remote Sensing

本文利用1982～2000年19年序列时间的FVC数据,以年单位作为自变量,将每个像元的FVC值作为因变量,利用线性回归理论来分析中国西部地区的植被变化状况,同时.讨论了对回归方程的F检验和t检验的结果和意义.通过以上工作认为:将具有时间序列的植被指数信息对时间进行回归分析后,所得到的结果比简单的差值分析的结果稳定,而且回归方程充分利用了所有的图像信息,并且具有延伸预测的理论基础.

In this paper, we use FVC（Fractional Vegetation Cover） and images were derived from NOAA/AVHRR time series data from 1982 to 2000 as time（year） is independent variable, every pixel is attributive variable, using regress mothed to analyze the change of vegetaion cover in Westen China, meanwhile, we discuss the result and signification of F-test and t-test for this regress. Through above work：1）the result of regress method for time series data is much stablization than Phase Difference Image; 2）Regress method can be used to forecast the change of vegetation and Phase difference image can not;3）By regress image,we can find that the increase of vegetation cover is close relationship with old oasis, Human activities is obviously one of the most important factors contributing to the change of vegetation cover in the arid lands in Westen China. In Qaidam High Basin, there are few signs of human habitation, the vegetation cover decreased, indicating the environment has degraded in this area.