Abstract: It is one of key techniques to retrieve the object region efficaciously from massive remote sensing image database. In this paper, we give a kind of model of content-based retrieval in remote sensing image according to the characters of remote sensing image application firstly, and then we analyze and discuss the algorithm for feature extraction adopted and multidimensional indexing and relevance feedback by this model in detail. Finally, we propose the intending contents to be research in this model.