期刊名称：ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences
印刷版ISSN：2194-9042
电子版ISSN：2194-9050
出版年度：2008
卷号：XXXVII Part B7
页码：797-802
出版社：Copernicus Publications
摘要：Urban change detection of high resolution images is very difficult due to the complex nature of the urban environments. A robust approach is proposed in this paper for urban change detection of high resolution images, which is based on the integration of object- specific features. To model the contextual information and improve the interclass variability between different classes, homogeneous regions are first extracted. Based on the robust object-specific difference measure, the complex changes of urban scene are represented by the fusion of object-specific spectral and textural features. Experimental results obtained on QuickBird images confirm the effectiveness and advantage of the proposed approach
