EEG Feature Extraction in Brain-Mobile Phone Interfaces

Poyil, Azeemsha Thacham and Hema, C. R. (2015) EEG Feature Extraction in Brain-Mobile Phone Interfaces. British Journal of Applied Science & Technology, 8 (5). pp. 506-520. ISSN 22310843

[thumbnail of Poyil852014BJAST14781.pdf] Text
Poyil852014BJAST14781.pdf - Published Version

Download (2MB)

Abstract

Aims: The paper describes four methods for extracting features of brain signals in frequency and time domain that can be used as parameters for identifying the face images of people.
Place and Duration of Study: Human Computer Interface Lab, Karpagam University, India (July 2012-June 2014)
Methodology: The subject is asked to remember different known face images like father, mother and so on and the corresponding Electroencephalogram (EEG) signals are captured. Wigner Ville Distribution and other spectral methods are used for studying the features. The data was collected from 15 subjects having good mental health condition. Real EEG records from different subjects are taken for duration of 10 seconds in each trial. 10 such trials are taken from each test subject.
Results: It is noticed that the band of frequencies in the range 0-40 Hz shows higher spectral variations, due to “remembering” or retrieving memories possibly due to the presence of Alpha or Beta waves. Using Fourier spectrum analysis it is found that, the EEG signals corresponding to the face image of one person (e.g: Mother) was always giving a different range of values for the number of spectral crossings, in comparison to the second face image (eg: Father). During the Wigner Ville analysis the peak value of instantaneous power in one case was seen in the range of 2x105 to 3x105 for father’s face whereas for mother’s face in the range of 0.5x106 to 1x106. In the Power Spectral Density based analysis, the frequency range of 10-20 Hz showed a higher average value in case of mother’s face than in father’s. When the mean signal power was calculated from PSD for different trials, it is noticed that the signal power is significantly different in cases of father and mother and gavea 70-90% of correct classification result.
Conclusion: After identifying the features that are unique for a face image, the same is proposed to be used for the address book dialing in a smart phone which can be further used for helping physically disabled as well as normal people to interact with external world.

Item Type: Article
Subjects: East India Archive > Multidisciplinary
Depositing User: Unnamed user with email support@eastindiaarchive.com
Date Deposited: 03 Jul 2023 04:47
Last Modified: 21 Oct 2024 04:02
URI: http://ebooks.keeplibrary.com/id/eprint/1365

Actions (login required)

View Item
View Item