In this paper, an algorithm for automatic brain activity class identification of EEG (electroencephalographic) signals is presented. EEG signals are gathered from seventeen subjects performing one of the three tasks: resting, watching a music video and playing a simple logic game. The methodology applied consists of several steps, namely: signal acquisition, signal processing utilizing z-score normalization, parametrization and activity classification. The EEG signal is acquired from a headset containing 14 electrodes. For the parametrization two methods are used, namely, DiscreteWavelet Transform (DWT) employed as a reference parametrization technique and autoencoder neural network. Parameters obtained with those methods are fed to the input of classifiers which assigned them to one of three activity classes. Then, the effectiveness of the assignment of the frames of EEG data into appropriate classes is observed and compared. Results obtained using both methods show differences in accuracy with regard to the task detected depending on factors such as type of parametrization or complexity of the classifier employed for EEG activity classification.
Authors
Additional information
- DOI
- Digital Object Identifier link open in new tab 10.3233/jifs-179360
- Category
- Publikacja w czasopiśmie
- Type
- artykuły w czasopismach
- Language
- angielski
- Publication year
- 2019
