摘要
论文阐述了一种基于在线学习算法的彩色图像区域增长法,用于解决基于内容的图像检索系统。该算法采用贝叶斯估计法的变分描述每一个增长区域,由此得出的图像生成过程应用简单,初始参数具有鲁棒性,且具有线性复杂性。图像处理结果表明了该方法用于基于内容的图像检索系统的可行性。
This paper presents a color images region growing method based on online learning algorithm,which is used for content-based image retrieval systems.The algorithm adopts a variation of Bayesian estimation procedure to characterize each grown region.The resulting image growing procedure is simple to implement,robust to initial parameters and has a linear complexity.Results on several images show the feasibility of the proposed method,which is used in the content- based image retrieval systems.
出处
《计算机工程与应用》
CSCD
北大核心
2005年第9期35-36,58,共3页
Computer Engineering and Applications
基金
国家"十五"科技攻关项目(编号:2001BA706B-15)
关键词
彩色图像分割
区域增长
贝叶斯学习
在线学习
color image segmentation,region growing,Bayesian learning,online learning
