|
|||||
|
|
| 基于分裂EM算法的GMM参数估计 | |
| 引用本文： | 钟金琴,辜丽川,檀结庆,李莹莹.基于分裂EM算法的GMM参数估计[J].计算机工程与应用,2012,48(34):28-32,59. |
| 作者姓名： | 钟金琴 辜丽川 檀结庆 李莹莹 |
| 作者单位： | 1. 安徽大学电子与信息系,合肥230031;合肥工业大学计算机与信息学院,合肥230009 2. 安徽农业大学计算机信息学院,合肥,230036 3. 合肥工业大学计算机与信息学院,合肥,230009 |
| 摘 要： | 期望最大化(Expectation Maximization,EM)算法是一种求参数极大似然估计的迭代算法,常用来估计混合密度分布模型的参数。EM算法的主要问题是参数初始化依赖于先验知识且在迭代过程中容易收敛到局部极大值。提出一种新的基于分裂EM算法的GMM参数估计算法,该方法从一个确定的单高斯分布开始,在EM优化过程中逐渐分裂并估计混合分布的参数,解决了参数迭代收敛到局部极值问题。大量的实验表明,与现有的其他参数估计算法相比,算法具有较好的运算效率和估算准确性。 |
| 关 键 词： | 高斯混合模型 期望最大化 参数估计 模式分类 |
Estimating parameters of GMM based on split EM |
|
| ZHONG Jinqin , GU Lichuan , TAN Jieqing , LI Yingying.Estimating parameters of GMM based on split EM[J].Computer Engineering and Applications,2012,48(34):28-32,59. | |
| Authors: | ZHONG Jinqin GU Lichuan TAN Jieqing LI Yingying |
| Affiliation: | 1.Department of Electronic and Information,Anhui University,Hefei 230031,China 2.School of Information and Computer,Anhui Agriculture University,Hefei 230036,China 3.School of Computer and Information,Hefei University of Technology,Hefei 230009,China |
| Abstract: | The expectation maximization algorithm has been classically used to find the maximum likelihood estimates of parameters in mixture probabilistic models. Problems of the EM algorithm are that parameters initialization depends on some prior knowledge, and it is easy to converge to a local maximum in the iteration process. In this paper, a new method of estimating the parameter of GMM based on split EM is proposed, it starts from a single mixture component, sequentially split and estimates the parameter of the mixture components during expectation maximization steps. Extensive experiments show the advantages and efficiency of the proposed method. |
| Keywords: | Gaussian Mixture Model （GMM） Expectation Maximization （EM） parameters estimation pattern classification |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ | |