图学学报
• 图像处理与模式识别 • 上一篇 下一篇
出版日期:
发布日期:
Online:
Published:
摘要: :图切分(Graph Cuts,GC)是近年来兴起的基于图论框架的图像分割方法, 该理论的新颖之处在于它的全局最优性和结合多种知识的统一性。 但当图像较大时运算非 常耗时。该文提出了一种基于GC 的层次式图像分割方法。先在低分辨率中用GC 以较低的 分割代价获取粗尺度的初始分割,再将结果轮廓映射回高分辨率图像中并构造出窄带,进而 采用matting 思想,在窄带内获取精确分割。实验结果表明,本文方法在确保分割结果准确 性的同时,运算速度大幅度提高。
关键词: 信息处理技术, 图像分割, 图切分, 多层次分割, 抠图
Abstract: Graph Cuts (GC) is a novel image segmentation method based on graph theory framework. The innovations of this theory lie in its global optimization and the unity of knowledge. However, if the image is large, computation will be very time-consuming. This paper presents a GC-based Hierarchical image segmentation method. First the initial segmentation is obtained through GC in the low-resolution with a very low computational cost. Then the contour is projected back to the high-resolution image to construct a narrow band. At last accurate segmentation in the narrow-band is achieved by using of matting arithmetic. Experimental results show that this method can ensure the accuracy of segmentation results with a significant increasing in computing speed.
Key words: information processing technology, image segmentation, graph cuts, multilevel segmentation, matting
蒋建国, 张 婕, 詹 曙, 郭艳蓉. 层次式图切分快速分割算法[J]. 图学学报.
Jiang Jianguo, Zhang Jie, Zhan Shu, Guo Yanrong. Multilevel graph cuts for fast image segmentation[J]. Journal of Graphics.
0 / / 推荐
导出引用管理器 EndNote|Ris|BibTeX
链接本文: http://www.txxb.com.cn/CN/
http://www.txxb.com.cn/CN/Y2012/V33/I1/44