Loading...

Classification of Different Mental Activities Based on Riemannian Geometry

Ghamchili, Mehdi | 2015

638 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 46972 (05)
  4. University: Sharif University of Technology
  5. Department: Electrical Engineering
  6. Advisor(s): Babaiezadeh, Massoud
  7. Abstract:
  8. Brain-Computer Interface (BCI) presents a way for brain’s direct connection with external world. BCI system is composed of three parts: 1) Signal acquisition, 2) Signal processing and 3) External device control. The main part of this system is signal processing which includes three subparts: 1) Feature extraction, 2) Dimension reduction and 3) Signal separation and classification. In this thesis, we focus on the signal processing section in BCI systems. One of the most successful works done in signal processing is the use of covariance matrices in feature extraction from brain signals. Since covariance matrices are positive semi-definite and symmetric, they belong to certain manifolds called Riemannian manifolds. Therefore, these matrices are transferred to Riemannian tangent space of a certain point Cref and then the matrix elements are used as spatial features. Finally, in this method brain signals are classified after dimension reduction by a linear classifier. In this thesis, we present an algorithm for detection and removing outlier in order to generalize the classification ability and increase the accuracy of brain signals classification. In addition, we suggest a combination of obtained spatial information with the information available in other domains such as time, frequency and time-frequency, to increase the accuracy of mental activity classification compared to the above-mentioned algorithm. Suggested algorithms and methods were implemented on multi-class dataset IIa from the BCI competition IV. The results show that better performance is achieved through combining the spatial information with the frequency information in which the accuracy of brain signal classification is increased by %2 compared to the above-mentioned algorithm
  9. Keywords:
  10. Covariance Matrix ; Brain-Computer Interface (BCI) ; Riemannian Manifold

 Digital Object List

 Bookmark

No TOC