针对传统遥感影像检索方法大多存在精度不高、效率低下等不足的问题,该文提出了一种基于归一化转动惯性特征的遥感影像检索算法。该算法对经过脉冲耦合神经网络处理的二值影像序列进行计算,提取影像序列的归一化转动惯量特征;同时,利用马氏距离结合Pearson积矩相关法来度量各特征矢量之间的相似性,提高检索结果的正确率。实验结果证明,该算法有效地兼顾了影像的内容结构信息,不仅可以快速地进行检索计算,还能提高检索精度。 Aiming at the problems of low accuracy and low efficiency, most traditional remote sensing image retrieval methods have a problem of remote sensing image retrieval based on normalized rotational inertia. The algorithm calculates the normalized moment of inertia feature of the image sequence by using the binary neural network which is processed by the pulse-coupled neural network. At the same time, the Mahalanobis distance and Pearson’s method are used to measure the similarity between the feature vectors , Improve the accuracy of the search results. Experimental results show that this algorithm effectively balances the content structure of the image, which not only can quickly search and calculate, but also improve the retrieval accuracy.