SUN Jun-ding, CHEN Su-xia, GUO Qi-qiang. Automatic extraction of regions of interest based on rough sets[J]. 2009, 28(3): 101-103+107. DOI: 10.13873/j.1000-97872009.03.014.
基于粗糙集的图像感兴趣区域自动提取算法
摘要
在分析常规区域生长算法的基础上
提出一种基于粗糙集理论的感兴趣区域自动提取算法.在显著度图和相对位置指示图的基础上
以像素点的显著度、相对位置及颜色信息构成知识表达系统
通过属性约简和值约简
导出了有用的区域生长决策规则。实验结果表明:该方法与现有算法相比
它能够更好地提取图像感兴趣区域。
Abstract
By the analysis of traditional region growth algorithm
an automatic extraction method of regions of interest(ROI)based on rough sets is presented.Combination of salience and relative position image
knowledge expression system is introduced by the salience
relative position and the color information of each image pixel.The region growth decision rules are educed by attribute and value reduction.The experiment shows that this method can extract the ROI more precisely than the traditional region growth method.