一种基于颜色和纹理的图像检索方法  

Image Retrieval Based on Color and Texture

作　　者：徐慧英[1] 袁杰[1] 赵建民[1] 朱信忠[1] 

机构地区：[1]浙江师范大学数理与信息工程学院,金华321004

出　　处：《计算机科学》2009年第5期282-286,共5页Computer Science

基　　金：国家自然科学基金(60773197);浙江省自然科学基金(Y107750)资助

摘　　要：提出了一种基于颜色和纹理的新的图像检索方法。该方法在HSV颜色空间量化过程中,采用一种基于线性隶属度的模糊量化方法,使量化结果更接近于人类感知;在提取分块主色的基础上引入一种颜色相邻矩阵,用于描述颜色的相对空间分布情况,以提高对图像变形的鲁棒性;还加入图像的纹理信息,最后结合这些颜色和纹理特征进行复合图像检索。实验表明,该方法能有效提高图像检索精度。A new picture retrieval algorithm based on color and texture was presented. In the process of scaling color in HSV color space, the new algorithm used a scaling method based on fuzzy scaling to better meet human＇s sensation. The algorithm also introduced a color adjacent matrix after extracting every picture block＇s main color to describe colors＇ relative spacious distribution and enhanced its robustness when the target picture had changed. It added pictures＇ texture information too, at last, it combines all those above to search involved pictures. Experiments indicate that this algorithm can improve the searching precision effectively.

关 键 词：基于内容图像检索 颜色空间 颜色量化 颜色直方图 纹理 

分 类 号：TP391.3[自动化与计算机技术—计算机应用技术] TP391.41[自动化与计算机技术—计算机科学与技术]