Figure 1
The flow diagram is arranged horizontally into four vertical columns of stacked rectangles, with downward arrows connecting each rectangle within a column. Starting from the leftmost column at the top, a rectangle reads “Proposed Unimodal E R Method for Arousal Dimension”. A downward arrow points to the next rectangle, reading “E E G Signal (32 Channels)”. A downward arrow points to “Pre-processing”. Another downward arrow points to “Time and Frequency Domain Feature Extraction”. A downward arrow points to “Feature Selection using Best First Strategy”. A downward arrow points to “Stacking”. A final downward arrow points to the bottom rectangle, reading “Output Class Labels - L A, H A”. In the second column, the top rectangle reads “Proposed Multimodal E R Method for Valence Dimension”. A downward arrow points to a rectangle reading “Physiological Signals”, containing the texts “z E M G” and “E O G”. A downward arrow points to “Pre-processing”. A downward arrow points to “Time and Frequency Domain Feature Extraction”. A downward arrow points to “Feature Selection using Best First Strategy”. A downward arrow points to “Logit Boost”. A final downward arrow points to “Output Class Labels - L V, H V”. In the third column, the top rectangle reads “Proposed E R Method from ergonomic perspective for Valence Dimension”. A downward arrow points to “E O G Signal”. A downward arrow points to “Pre-processing”. A downward arrow points to “Time and Frequency Domain Feature Extraction”. A downward arrow points to “Linear Discriminant Analysis”. A final downward arrow points to “Output Class Labels - L V, H V”. In the fourth column, the top rectangle reads “Proposed E R Method from ergonomic perspective for Arousal Dimension”. A downward arrow points to “Plethysmograph Signal”. A downward arrow points to “Pre-processing”. A downward arrow points to “Time and Frequency Domain Features”. A downward arrow points to “Linear Discriminant Analysis”. A final downward arrow points to “Output Class Labels - L A, H A”.Proposed multimodal and unimodal emotion recognition methods for arousal dimension and valence dimension