基于图割优化的Markov随机场图像分割方法综述

Overview of Markov Random Fields for Image Segmentation Based on Graph Cut Optimization Method

Markov随机场能有效地刻画图像像素间的关系,具有完善的理论基础,因而在图像分割等领域得到研究者的关注。图割作为一种基于图论的组合优化方法,已经较好地用在基于MRF的图像分割等计算机视觉方面。简要介绍了Markov随机场模型和图割优化方法,给出了基于图割优化的Markov随机场图像分割框架,分析了Markov随机场方法和图割优化在图像分割中的研究现状,并指出其将来的发展方向。

Markov random field（MRF） can effectively describe the relationship between image pixels. With a perfect theoretical basis , MRF in image segmentation fields are the concern of researchers. Graph cut as a combination optimization method based on graph theory,it has a good use of the image segmentation based on MRF in the field such as computer vision. The MRF model and graph cut optimization method is introduced in this paper, image segmentation framework is given based on graph cut optimization of MRF model, analysis of MRF method and graph cut optimization in the research status of image segmentation , points out the developing direction in the future.