A revision for gaussian mixture density decomposition algorithm

Authors:
Xiaobing Yang;Fansheng Kong;Bihong Liu
Affiliations:
Artificial Intelligence Institute, Zhejiang University, Hangzhou, China;Artificial Intelligence Institute, Zhejiang University, Hangzhou, China;Artificial Intelligence Institute, Zhejiang University, Hangzhou, China
Venue:
CIS'04 Proceedings of the First international conference on Computational and Information Science
Year:
2004

Citing 3
Cited 0

A Highly Robust Estimator Through Partially Likelihood Function Modeling and Its Application in Computer Vision

IEEE Transactions on Pattern Analysis and Machine Intelligence
Data mining: concepts and techniques

Data mining: concepts and techniques
Gaussian mixture density modeling, decomposition, and applications

IEEE Transactions on Image Processing

Quantified Score

Hi-index	0.02

Visualization

Abstract

Gaussian mixture density decomposition (GMDD) algorithm is an approach to the modeling and decomposition of Gaussian mixtures, and it performs well with the least prior knowledge in most case. However, there are still some special cases in which the GMDD algorithm is difficult to converge or can not gain a valid Gaussian component. In this article, a k-means method for Gaussian mixture density modeling and decomposition is studied. Then based on the GMDD algorithm andk-means method, a new algorithm, called k-GMDD algorithm is proposed. It solves the problems of GMDD caused by the symmetry excellent, and consequently makes the applications of GMDD algorithm more extensive.