Abstract: Aiming at the problem that conventional classification methods for change detection of medium and low resolution images can not achieve ideal results, this paper proposes a method of urban change detection combining radar data and optical images based on ratio method.The method integrates the advantages of radar data and optical image, takes Dingnan county of Ganzhou as the research object,uses ratio method respectively to detect the urban change of two kinds of data sources. Through the analysis of urban ground object target scattering characteristics of the multiple phase change it composite false color. Finally, the city construction planning is analyzed and verified to complete the real-time dynamic change detection of urban.The results show that the ratio method using multi-source and multi-temporal data can successfully extract the change information of the experimental area, and the texture information is clearer and more accurate in the change detection of urban buildings.

Key words: urban change detection, optical image, radar data, ratio method, quantity anaysis