结合EM/MPM算法和Voronoi划分的图像分割方法
Combining the EM/MPM and Voronoi Tessellation for Image Segmentation
-
摘要: 为了实现在模型参数先验分布知识未知情况下进行基于区域和统计方法的图像分割,同时获取更精确的模型参数估计结果,提出了一种结合Voronoi划分技术、最大期望值(Expectation Maximization, EM)和最大边缘概率(Maximization of the Posterior Marginal, MPM)算法的图像分割方法。该方法利用Voronoi划分技术将图像域划分为若干子区域,待分割图像中的同质区域可以由一组子区域拟合而成,并假定同一同质区域内像素强度服从同一独立的正态分布,从而建立图像模型,然后结合EM/MPM算法进行图像分割和模型参数估计,其中,MPM算法用于实现面向同质区域的图像分割,EM算法用于估计图像模型参数。为了验证本文图像分割方法,分别对合成图像和真实图像进行了分割实验,测试结果的定性和定量分析表明了该方法的有效性和准确性。Abstract: This paper presents a new approach for image segmentation, which combines Voronoi tessellation technique and expectation-maximization/maximization of the posterior marginal (EM/MPM) algorithm. By Voronoi tessellation, the domain of a give image is partitioned into sub-regions for constructing homogeneous regions. The EM/MPM algorithm is based on the MPM algorithm for segmentation and the EM algorithm for parameter estimation. This paper also presents the experimental results demonstrating the performance of the EM/MPM algorithm.