Change Detection Algorithm Using Multitemporal SPOT Images

Abstract

Change detection algorithm using multitemporal SPOT images is proposed. Firstly, multitemporal images are registrated with Harris-Laplace feature points. Then, based on increment and decrement of pixel's value in four directions, candidate change region is estimated by improved increment sign correlation method. Finally, change area is output according to relationship between global difference rate and local difference rate. The experimental results show that this algorithm is simple and efficient, which can obtain accurate and integrated change detection result.