Analysis of nonlinear and non-stationary signal to extract the features using Hilbert Huang transform

• Published in: 2014 IEEE International Conference on Computational Intelligence and Computing Research pp. 1 - 4
• Main Authors: Paithane, A. N., Bormane, D. S.
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.12.2014
• Subjects: Data mining; Electrocardiogram (ECG); Electrocardiography; Emotion recognition; Empirical mode decomposition; Feature extraction; Fission; Fusion; Mathematical model; Mean Frequency (MNF) Emotion Detection
• ISBN: 1479939749; 9781479939749
• DOI: 10.1109/ICCIC.2014.7238497

Human emotion recognition is a hot research topic in medical and engineering field to provide interface between users and computers. A novel technique for Feature Extraction (FE) has been presented here. This method is feasible for analyzing the nonlinear and non-stationary signals like electrocardiogram signal (ECG), Electromyogram (EMG) etc. We have used electrocardiogram signal as an input, each signal has important functions, which has been extracted by applying empirical decomposition method. These functions are used to extract the features using fission and fusion process. The features extracted from every IMF are used to compose feature vector. The extracted features are useful to recognize human emotions from ECG signal. The decomposition technique which we adopt is a new technique for adaptively decomposing signals. In this perspective, we have reported here potential usefulness of EMD based techniques. We evaluated the algorithm on Augsburg University Database; the manually annotated database.