基于多语义特征的彩色图像检索技术研究  

Content-based Color Image Retrieval Using Multi-semantics

作　　者：牛盼盼[1] 王向阳[1,2] 周璐[1] 杨红颖[1] 

机构地区：[1]辽宁师范大学计算机与信息技术学院,大连116029 [2]北京大学视觉与听觉信息处理国家重点实验室,北京100871

出　　处：《计算机科学》2009年第3期226-231,共6页Computer Science

基　　金：国家自然科学基金(60773031);计算机软件新技术国家重点实验室(南京大学)开放基金(A200702);视觉与听觉信息处理国家重点实验室(北京大学)开放基金(0503);大连市科技基金(2006J23JH020);“图像处理与图像通信”江苏省重点实验室(南京邮电大学)开放基金(ZK205014);江苏省计算机信息处理技术重点实验室(苏州大学)开放课题基金(KJS0602)资助

摘　　要：基于语义内容的图像检索已成为解决图像低层特征与人类高级语义之间"语义鸿沟"的关键。以性能优越的回归型支持向量机(SVR)理论为基础,结合重要的图像边缘信息及人眼视觉特性,提出了一种基于多语义特征的彩色图像检索新算法。该算法首先利用Canny检测算子提取原始图像的边缘信息,并得到低层纹理特征,同时利用SVR将低层特征映射到高级语义,以获得图像的高级纹理语义。然后结合人眼视觉系统感知特性,给出基于重要区域主要颜色的高级颜色语义。最后根据上述高级语义特征(纹理语义和颜色语义)进行图像检索。实验结果表明,该算法能够有效地对图像高级语义进行刻画,不仅图像匹配检索效果良好,而且具有稳定的检索性能,其对于缩小低层视觉特征与高级语义概念之间的"语义鸿沟"具有重要意义。The performance of content-based image retrieval （CBIR） systems is largely limited by the gap between the low-level feature and high-level semantic concept. A new content-based color image retrieval method using multi-semantics was proposed, which not only takes into consideration the important image edge information and human visual system, but also utilizes the support vector regression （SVR） theory. Firstly, the important image edge was extracted by using canny detection operator and the low-level texture feature was computed. Then, the high-level texture semantic was determined by mapping the low-level feature using SVR. Secondly, the high-level color semantic could be obtained by using the main color of the most important region. Finally, image retrieval was implemented by using both texture semantic and color semantic. Experimental results show that the proposed image retrieval is effective in characterizing image high-level semantic and can provide sound and robust image retrieval performance, which has strong significance for reducing the ＂semantic gap＂ between the visual feature and semantic concern.

关 键 词：图像检索 高级语义 回归型支持向量机 视觉感知特性 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]