Improving the Accuracy and Training Speed of Motor Imagery Brain-Computer Interfaces Using Wavelet-Based Combined Feature Vectors and Gaussian Mixture Model-Supervectors.
David LeeSang-Hoon ParkSang-Goog LeePublished in: Sensors (2017)
Keyphrases
- gaussian mixture model
- brain computer interface
- feature vectors
- motor imagery
- training speed
- support vector machine
- healthy subjects
- mixture model
- spinal cord injury
- eeg signals
- training process
- learning rate
- signal processing
- feature space
- em algorithm
- subband
- feature extraction
- stochastic gradient descent
- brain activity
- wavelet coefficients
- multiresolution
- maximum likelihood
- feature set
- texture features
- speech recognition
- similarity measure
- multiscale
- wavelet transform
- face images
- eeg data
- classification accuracy
- gabor filters
- unsupervised learning
- decision trees
- learning algorithm