Research Of Emotion Recognition Based On Human Physiological Signals

For a long time,the feature extraction of human physiological signals based on conventional statistics is widely used.However,the method based on the conventional statistical features is not ideal in classifying and distinguishing effect.In order to solve this kind of problem,a method based on recursive graph and recursive quantitative analysis is proposed.Recurrence rate,the determination rate and the diagonal structure length of the physiological signal and so on can be extracted from recursive graph by recursive quantitative analysis.Neural Network,K Nearest Neighbor,Naive Bias,Decision Tree algorithm are applied to emotion recognition.The experimental results show that the feature in recursive graphs is a very effective set of characteristics.Compared with traditional statistical feature extraction,nonlinear feature extraction has less features,but it is better than the method of statistical feature extraction in the effect of classification.The method improves the problem of the large number of traditional feature extraction and unsatisfactory effect.In order to classify effectively,this thesis presents a fusion ant colony algorithm and particle swarm optimization algorithm.Although the performance of emotion recognition is improved after feature selection using PSO and ant colony algorithm,the hybrid feature selection algorithm proposed in this thesis is more effective than PSO and ant colony algorithm.It effectively solves the emotional recognition of human physiological signals.