中文摘要：

通过构建物元模型将遥感影像的查询问题转化为多属性决策问题。设计了基于拓扑和度量关系相结合的关联函数,定量评价遥感影像各属性与用户需求的关联度;利用熵权法确定权重反映用户兴趣在遥感影像元数据项上的侧重,从而得到个性化的评价结果。实例评价结果证明了该方法的合理性与可行性。

英文摘要：

The growth in the remotely sensed image available has led to information overload,as navigating through and finding relevant information has become more and more challenging.In this paper,a multiple attribute decision model is proposed to assist the user in his/her quest for accurate data.By constructing a matter-element model,the solutions of the remotely sensed image query can be converted into the solutions of multiple attribute decision of an alternative scheme.In addition,correlation functions combining topology with metrization are proposed to evaluate the similarity degree between user′s requirement and data quantitatively.The entropy-weight method is used to calculate the weight from user early search record,which reflects user preference on different metadata of remotely sensed image.Therefore,the advantages and disadvantages of different schemes can be compared.Finally,an example is given to show the usability of the presented model.