基于Fuzzy的隐马尔可夫模型主动轮廓线模型

A New Active Contour Model Based on Fuzzy and HMM

提出了一种基于Fuzzy和隐马尔可夫模型(HMM)的新型主动轮廓线模型,该模型将HMM理论应用于主动轮廓线模型中,利用HMM提供多种测量信息的能力,优化了模型参数,克服了传统主动轮廓线模型的缺点.同时利用模糊集理论建立了HMM中的状态与观测向量之间的模糊隶属关系,从而使蛇点准确收敛到图像边缘.对动态图像的仿真实验表明,改进后的主动轮廓线模型能很好地收敛到物体的凹陷边界,且具有良好的抗噪性.

A new active contour model based on Fuzzy and HMM is provided. Being of the capability of measuring many kinds of information, HMM is used to optimize the parameters of the model and overcome the disadvantages of the traditional active contour model. In addition, the Fuzzy theory is used to establish the relationship between the Fuzzy membership function of the states and the observed vectors in the HMM to make the snake points converge upon the edge of image. The simulative experiment of dynamic image shows that the boundary of a hollow object can be well reached in the imoroved active contour model, and the model has a good anti-noise property.