Improved Deep Feature Learning by Synchronization Measurements for Multi-Channel EEG Emotion Recognition

Figure 6

The recognition accuracy for four types of EEG features in two emotion dimensions under different classifiers. (a) Global MIC features in the arousal dimension. (b) Fusion features in the arousal dimension. (c) Frequency-domain features in the arousal dimension. (d) Time-domain features in the arousal dimension. (e) Global MIC features in the valence dimension. (f) Fusion features in the valence dimension. (g) Frequency-domain features in the valence dimension. (h) Time-domain features in the valence dimension.

(a)

(b)

(c)

(d)

(e)

(f)

(g)

(h)