摘要
针对大学数字博物馆数据库中文物图像的特点,提出了一种新的文物图像检索方法。特征提取采用基于分块图像建立模糊颜色直方图;模糊颜色直方图的建立不仅考虑了不同颜色索引像素之间的差异,也考虑了同一颜色索引像素间的差异;图像分块策略结合了文物图像的颜色特征与形状特征。一种新的图像相似性度量方法———交互信息距离(MID)用来进行相似性匹配。参数AVRR/IAVRR用来进行检索性能评价,评价结果表明,本文的方法在文物图像检索中具有较高的检索准确率。
Concerning about the characteristic of the cultural relic images, a new approach was proposed to cultural relic image retrieval. For feature extraction, fuzzy color histogram based on segmented image was used. Fuzzy color histogram (FCH )considers the color similarity of each pixel's color associated to all the histogram bins through fuzzy-set membership function. The strategy of using segmented image represented not only the color properties but also the shape properties. A new similarity measurement-mutual information distance was proposed. The proposed method was proved to be an effective one in terms of the evaluation parameter AVRR/IAVRR.
出处
《计算机应用》
CSCD
北大核心
2005年第8期1789-1791,共3页
journal of Computer Applications
基金
国家教育部资助项目(教技司[2001]234)
关键词
分块图像
模糊颜色直方图
文物图像
基于内容的图像检索
<Keyword>image subblock
FCH(Fuzzy Color Histogram)
cultural relic images
content-based image retrieval
