An Integrated Hierarchical Gaussian Mixture Model to Estimate Vigilance Level Based on EEG Recordings

Jing-Nan Gu, Hong-Jun Liu, Hongtao Lu, Bao-Liang Lu

Abstract

Effective vigilance level estimation can be used to prevent disastrous accident occurred frequently in high-risk tasks. Brain Computer Interface(BCI) based on ElectroEncephalo-Graph(EEG) is a relatively reliable and convenient mechanism to reflect one’s psychological phenomena. In this paper we propose a new integrated approach to predict human vigilance level, which incorporate an automatically artifact removing pre-process, a novel vigilance quantification method and finally a hierarchical Gaussian Mixed Model(hGMM) to discover the underlying pattern of EEG signals. A reasonable high classification performance (88.46% over 12 data sets) is obtained using this integrated approach. The hGMM is proved to be more powerful against Support Vector Machine(SVM) and Linear Discriminant Analysis(LDA) under complicated probability distributions.
BibTeX
@inproceedings{gu2011integrated,
  author = {Jing-Nan Gu and Hong-Jun Liu and Hongtao Lu and Bao-Liang Lu},
  title = {{An Integrated Hierarchical Gaussian Mixture Model to Estimate Vigilance Level Based on EEG Recordings}},
  booktitle = {ICONIP},
  year = {2011},
  doi = {10.1007/978-3-642-24955-6\_46},
}

Loading related content...

Figure