摘要
基于Hastie T.和Tibshirani R.等提出的"Gap statist"的思想,通过分析样本灰度数据分布的差别,提出了函数特征、间隙、总间隙及全间隙等概念,建立了图像分割的Gap统计模型,并给出了较为详细的算法。分析了图像分割Gap统计模型中正则部分和奇异部分的特点,导出了区域特征自相似函数的分割结果与模型调节参数的关系。比较了用于图像分割的Gap统计模型与 Mumford-Shah模型,结果表明图像分割Gap统计模型的复杂度明显低于Mumford-Shah模型。
Several new conceptions including function feature, total Gap and full Gap are proposed, and the Gap statistical model for image segmentation is established and the detail algorithm is developed. Also, we take an analysis of regular part and singular part in the Gap statistic model for image segmentation. And the relationship between parameters in the model and the self-similar function of area feature is derived. Compared with the famous model of Mumford- Shah functional, our model has lower complexity.
出处
《计算机科学》
CSCD
北大核心
2005年第12期223-226,共4页
Computer Science
基金
高等学校博士学科点专项科研基金(20020288024)资助
关键词
图像分割
函数特征
自相似
GAP统计
模型
Image segmentation, Function feature, Self-similar, Gap statistic, Model
