摘要
基于内容的图像检索技术是当前多媒体数据库系统和数字图书馆研究中的一个热点。文章基于二值图像的NMI不变特征(归一化转动惯量)提出了一种全新的基于内容的图像检索方法,其特点是计算简单、精确度高、具有抗几何畸变性,对滤波、平滑、压缩等图像操作有一定的鲁棒性。试验结果表明,该方法具有一定使用价值,可用于图像检索。
Content-based image retrieval is a hotspot in the research of multimedia database system and digital library. A new image retrieval method based on NMI invariable feature (normalized moment of inertia) of 1-bit image is presented in this paper, which is simple with high efficiency, and has the ability of anti-geometrical distortion while robust to some image processing such as filtering and smoothing etc. to some extent. Experimental results show that it can be used in image retrieval.
出处
《计算机工程》
CAS
CSCD
北大核心
2004年第16期14-15,38,共3页
Computer Engineering
关键词
图像检索
归一化转动惯量特征
特征提取
Image retrieval
Normalized moment of inertia feature
Feature extraction
