The purpose of this research is to select the best features to have a high rate of motion classification for controlling an artificial hand. Here, 19 EMG signal features have been taken into account. Some of the features suggested in this study include combining wavelet transform with other signal processing techniques. An assessment is performed with respect to three points of view: (i) classification of motions, (ii) noise tolerance and (iii) calculation complexity. The energy of wavelet coefficients of EMG signals in nine scales, and the cepstrum coefficients were found to produce the best features in these views.
1 April 2009 in Computer Methods and Programs in Biomedicine
This paper proposed a multimodal fusion between brain and peripheral signals for emotion detection. The input signals were electroencephalogram, galvanic skin resistance, temperature, blood pressure and respiration, which can reflect the influence of emotion on the central nervous system and autonomic nervous system respectively. The acquisition protocol is based on a subset of pictures which correspond to three specific areas of valance-arousal emotional space (positively excited, negatively excited, and calm). The features extracted from input signals, and to improve the results, correlation dimension as a strong nonlinear feature is used for brain signals. The performance of the quadratic discriminant classifier has been evaluated on different feature sets: peripheral signals, EEG's, and both. In comparison among the results of different feature sets, EEG signals seem to perform better than other physiological signals, and the results confirm the interest of using brain signals as peripherals in emotion assessment. According to the improvement in EEG results compare in each row of the table, it seems that nonlinear features would lead to better understanding of how emotional activities work.