基于线性四叉树提出一种面向影像金字塔的空间索引算法。在分析线性四叉树拓扑关系的基础上,设计一个具有方向一致、层次递进特性的编码方式,建立影像金字塔与线性四叉树的映射方案,给出一个按照经度纬度自然增长的邻域查找算法,并构建一个全球多分辨率虚拟地形环境对编码和算法进行测试。实验结果表明,该算法能够明显地缩小空间影像的检索时间,具有较高的编码效率和查找效率。
Based on linear quadtree,a spatial image retrieval algorithm of image pyramid is proposed.According to the topological relations of linear quadtree,this paper designs an encoding method with the characteristics of direction coherence,progressive hierarchy;establishes a mapping between image pyramid and linear quadtree;gives a neighbors-searching algorithm in accordance with the natural growth of longitude-latitude;and constructs a global multi-resolution virtual terrain environment to test the encoding method and algorithm.Experimental results show that the algorithm can significantly reduce the time cost of spatial image retrieval,and has a high encoding efficiency and neighbors searching efficiency.