Emotion perception in stroke patients is affected since there is abnormality in the brain. Here, this thesis focused on the impact of left brain damage and right brain damage towards emotion recognition. Due to the impaired emotion recognition, it is a challenge for stroke patients to express themselves in daily communication. Hence, it is inspiring to see the possibility to predict patient’s emotional state so as to prevent recurrent stroke. In this work, electroencephalograph (EEG) of 19 left brain damage patients (LBD), 19 right brain damage patients (RBD) and 19 normal control (NC) are collected as database. During data collection, six emotions (sad, disgust, fear, anger, happy and surprise) are induced by using audio visual stimuli. After normalization, EEG signals are filtered by using Butterworth 6th order band-pass filter at the cut-off frequencies of 0.5 Hz and 49 Hz. Then, wavelet packet transform (WPT) technique is implemented to localize five frequency bands: alpha (8 Hz–13 Hz), beta (13 Hz–30 Hz), gamma (30 Hz–49 Hz), alpha-to-gamma (8 Hz–49 Hz), beta-to-gamma (13 Hz–49 Hz). On the other hand, tuned Q-factor wavelet transform (TQWT) is also applied on five frequency bands to obtain 6 sub-bands. In WPT, four wavelet families are chosen: daubechies 4 (db4), daubechies 6 (db6), coiflet 5 (coif5) and symmlet 8 (sym8). Hurst exponents (HE), detrended fluctuation analysis (DFA), recurrence quantification analysis (RQA) are used to extract hurst correlation exponent, DFA correlation exponent, and 11 different measures out of recurrence plot from each band and wavelet family and are classified by using K-nearest Neighbour (KNN), Probabilistic Neural Network (PNN) and random forest (RF). Classification stage is done on comparison between three groups and also between six emotions. 290 combinations of feature are done but the most significant feature in emotion recognition and group recognition is mean of diagonal line length (<L>) and Recurrence Probability Density Entropy (RPDE) respectively. RQA measures are found to be most contributing in giving high classification accuracy. Meanwhile, correlation exponents such as Hurst and DFA are found inefficient in emotion classification. Emotion recognition accuracy is improved tremendously after the features are transformed into score by using Principle Component Analysis (PCA) algorithm. The maximum emotion classification accuracy is found in LBD (85.29 %). In group classification, RPDE has given the highest accuracy (99.41 %) through RF classifier. LBD group is found to have the higher average accuracy compared to RBD group which supports the ‘right hemisphere hypothesis’. The result shows that the proposed emotional recognition system is reliable with an acceptable classification accuracy which is helpful in emotion monitoring among stroke patients as well as healthy person. Meanwhile, group recognition (LBD, RBD and NC) can be used to identify the presence of LBD or RBD stroke.
