[1]郑加明,陈昭炯.局部颜色模型的交互式Graph-Cut分割算法[J].智能系统学报,2011,6(4):318-323.
ZHENG Jiaming,CHEN Zhaojiong.Interactive Graph-Cut for image segmentation using local color models[J].CAAI Transactions on Intelligent Systems,2011,6(4):318-323.
点击复制
《智能系统学报》[ISSN 1673-4785/CN 23-1538/TP] 卷:
6
期数:
2011年第4期
页码:
318-323
栏目:
学术论文—人工智能基础
出版日期:
2011-08-25
- Title:
-
Interactive Graph-Cut for image segmentation using local color models
- 文章编号:
-
1673-4785(2011)04-0318-06
- 作者:
-
郑加明,陈昭炯
-
福州大学 数学与计算机科学学院,福建 福州 350108
- Author(s):
-
ZHENG Jiaming, CHEN Zhaojiong
-
College of Mathematics and Computer Science, Fuzhou University, Fuzhou 350108, China
-
- 关键词:
-
GraphCut; 交互式图像分割; MeanShift; 实时交互性
- Keywords:
-
Graph-Cut; interactive image segmentation; MeanShift; real time interactivity
- 分类号:
-
TP391
- 文献标志码:
-
A
- 摘要:
-
由于目前大多数交互式GraphCut分割算法很难达到精确分割且实时交互的效果.对此,提出一种基于局部颜色模型的改进算法.该算法利用MeanShift预分割,建立基于局部颜色模型的交互式分割框架,并将像素级的GraphCut算法转化为基于区域的算法进行快速求解.预分割之后的区域保持了原有图像的结构,不仅提高了采用局部颜色模型估计分布的准确性,而且基于区域GraphCut的算法明显降低了计算的复杂度.实验结果表明,改进后的算法不仅保证了分割的精确性,而且还达到了实时交互.
- Abstract:
-
Most interactive segmentation algorithms based on GraphCut do not usually lend themselves to real time applications with accurate segmentation. In this paper, an improved algorithm using local color models was proposed to deal with the problem. Using MeanShift technology, the proposed algorithm built an interactive image segmentation framework based on local color models. A GraphCut algorithm was then applied to the presegmented regions instead of image pixels. The presegmented regions preserve the image structure, which improves the estimation accuracy of distribution based on local color models and dramatically reduces the computational complexity. Experimental results show that the proposed algorithm has good real time interactivity with accurate segmentation.
更新日期/Last Update:
2011-09-30