高级检索

基于主色匹配的图像检索系统

Content Based Image Retrieval (CBIR) System Based on Dominant Color Matching

  • 摘要: 基于内容的图像检索CBIR(Content Based Image Retrieval)是多媒体信息管理系统的重要组成部分.在简要介绍CBIR技术的基础上,提出了CBIR系统的结构模型,新的图像内容的特征提取的方法和基于该特征的图像匹配方法--基于主色的图像特征提取(DCE, Dominant Color Extracting)和匹配方法(DCM, Dominant Color Matching),并在此基础上设计实现了WWW发布方式的SPRING实验系统,取得了理想的实验结果.

     

    Abstract: Content Based Image Retrieval (CBIR) is an important component of Content Based Retrieval (CBR) system. Its function modules and the new approach of Dominant Color Block (DCB) for feature extraction and image matching is summarized. A 3-D clustering technology and statistics calculation based Dominant Color Extraction (DCE) and Dominant Color Matching (DCM) is also realized. On the basis of the detail study of CBIR technologies, a CBIR system named SPRING and supporting WWW distribution is designed and implemented, it shows a good performance.

     

/

返回文章
返回