Recognition of Emotions Using Multichannel EEG Data and DBN-GC-Based Ensemble Deep Learning Framework
Table 1
Detailed description of raw EEG features.
Index
Type
Domain
Notations of the extracted features
No. 1–128
Statistical measures
Time domain
Mean, variance, zero-crossing rate, and approximate entropy of 32 EEG channels (4 features × 32 channels)
No. 129–288
Power features
Frequency domain
Average PSD in theta (4–8 Hz), slow-alpha (8–10 Hz), alpha (8–12 Hz), beta (12–30 Hz), and gamma (30–45 Hz) bands for all EEG channels (5 power × 32 channels)
No. 289–344
Power differences
Frequency domain
Difference of average PSD in theta, alpha, beta, and gamma bands for 14 EEG channel pairs between right and left scalp (4 power differences × 14 channel pairs)
No. 345–664
HHS features
Time-frequency domain
Average values of squared amplitude and instantaneous frequency of HHS-based time–frequency representation in delta (1–4 Hz), theta (4–8 Hz), alpha (8–12 Hz), beta (12–30 Hz), and gamma (30–45 Hz) bands for all EEG channels (2 features × 5 bands × 32 channels)
