针对传统EM算法存在估计参数不具有最优性，以及在参数估计中需要人工参与等问题，提出一种基于高斯混合模型的改进EM算法。采用无人工参与的无监督思想，获取高斯混合模型对直方图拟合的最优参数组合。实验表明，该算法不仅能够快速地估计模型参量，而且能够给出最优参数，并在图像增强中使细节更明显，对比度更适中。
Abstract
In order to solve the disadvantages of traditional expectation maximization (EM) algorithm which lacks parameters optimization and needs human operation when estimating parameters, an improved EM algorithm based on Gaussian mixture model was proposed. The unsupervised theory was used to calculate optimal Gaussian mixture model parameters. The subjective and objective indices of experiments show that the algorithm can not only estimate parameters quickly but also figure out the optimal parameters, making the detail more obvious and the contrast more moderate in image enhancement application.
关键词
高斯混合模型 /
图像增强 /
EM算法
{{custom_keyword}} /
Key words
EM algorithm /
Gaussian mixture model /
image enhancement
{{custom_keyword}} /
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
[1]胡庆林,叶念渝,朱明富. 数据挖掘中聚类算法的综述[J]. 计算机与数字工程，2007, 35(2):17-20.
HU Qing--lin, YE Nian-yu, ZHU Ming-fu. Summarize of clustering algorithm in data mining[J]. Computer and Digital Engineering, 2007, 35(2):17-20. (in Chinese with an English abstract)
[2]王熙照,王亚东,湛燕,等. 学习特征权值对K-均值聚类算法的优化[J]. 计算机研究与发展, 2003, 40: 869-873.
WANG Xizhao, WANG Ye-dong, KAN Yan. Studying optimize of the feature weights about K-means clustering algorithm[J]. Research and Development of Computer,2003, 40: 869-873. (in Chinese with an English abstract)
[3]GAO Dan-yang, YANG Bing-ru. An impronved K-medoids clustering algorithm[C] //Proc of the 2nd International Conference on Computer and Autonmation Engineering (ICCAE). USA:IEEE,2010.
[4]赵国富,曲国庆. 聚类分析中CLARA算法的分析与实现[J]. 山东理工大学学报:自然科学版,2006,20(6):2-4.
ZHAO Guo-fu, QU Guo-qing
{{custom_fnGroup.title_cn}}
脚注
{{custom_fn.content}}
