Application of Multitemporal Composition and Classification to Land Use Change Detection

This paper discusses the situation of digital remote sensing techniques which has been applied to monitor the land use dynamic change, and introduces the notion of land use dynamic monitoring by remote sensing. The land use change in Guangxi province in 2002 and 2003 is detected by using direct classification of composed multi-temporal MODIS images. The results show PCA bands can well inherit the original images to achieve better accuracy.