内容的图像检索系统中的关键技术

Study on key technology of content-based image retrieval system

介绍了基于内容的图像检索技术的研究现状和应用现状。阐述了基于内容的图像检索系统的通用框架中的查询模块、描述模块、匹配模块、提取模块、验证模块的作用。并对图像检索系统所涉及关键技术特征提取中的颜色、纹理、形状和物体间方位关系,以及颜色检索中的图像间的相似度,纹理是与物体表面材质有关的图像特征,高维索引中的维度压缩方法,KL变换和列向聚类,多维索引技术中的K-d树,R树及其变种R+树、R*树,四叉树等等,以及聚类法、神经网络法进行了分析和研究。给以后实现基于内容的图像检索系统提供了理论基础。

The research and current application on content-based image retrieval is introduced in the paper. This paper expatiate the content-based image retrieval system general frame,including the function of inquiry module, the description module, the match module, the extraction module. The retrieval of the color similarity between images, and surface texture is the image of the characteristics of materials, high-dimensional cited in the compression method, KL transform and out to the cluster, multi-dimensional indexing technology Kd trees, R tree and its variants R ＋ tree, R＊ tree, etc., as well as the clustering method, neural network analysis and research. Key technology and retrieval characteristics of content-based image retrieval are also analyze and study ,including the color, the texture, the shape and the object the position relations.