ABSTRACT
We are living in a changing world. In the Earth environment, natural and human induced environmental changes occur in time and space. Effective detection and modelling of such change in the context of geospatial information technology are typically termed change detection, which often includes the detection of changes of the objects on the ground or more general change of the environmental background (Richard et al. 2005, Bovolo & Bruzzone 2007). Timely and correct detection of changes of surface features is extremely important for better understanding of relationships and interactions between human and natural phenomena, and lays the foundation for better decision making (Lu et al. 2004). Remotely sensed data have been the most important data source for studying the human living environment in the past 40 years and the large collection of past and present remote sensing imagery makes it possible to analyse
spatio-temporal patterns of environmental elements and impact of human activities in past decades. Research has been widely reported on the methodology of remote sensing change detection and monitoring (e.g. Singh 1989, Lu et al. 2004).
