摘要
提出一种新的基于马尔可夫随机场(Markov Random Field,MRF)的图像分割算法。根据Gibbs分布与MRF的等价性,图像分割问题转换为后验能量函数最小化所对应的标号问题。该文采用图割技术的-αexpansion算法进行后验能量函数的局部最优化,并通过近似于最大期望(EM)算法的迭代过程估算数据模型中的参数。对合成图像和遥感图像的分割实验表明,该方法的运算时间和分割精度都能达到满意的效果。
Image segmentation has been one of the hot fields of computer vision. In this paper, a novel Markov random fields image segmentation algorithm is proposed. According to Gibbs distribution and MRF equivalence, image segmentation problem is transformed to minimize the posterior energy function corresponding to the labeling problem. The energy function can be efficiently minimized using the-expansion move algorithm which is one of the most effective algorithms in graph cuts, and using an iterative process similar to the EM algorithm to estimate the data term parameter. Experimental results are provided to illustrate the satisfactory performance of this method on both synthetic and remote sensing images.
出处
《地理与地理信息科学》
CSSCI
CSCD
北大核心
2011年第2期38-40,共3页
Geography and Geo-Information Science
关键词
分割
马尔可夫随机场
图割
能量最小
segmentation
Markov random fields
graph cuts
energy minimization
