Emotion recognition method and system based on electrocardiosignal

Abstract

The invention discloses an emotion recognition method and system based on electrocardiosignals, and the method comprises the steps: receiving a to-be-detected single-channel electrocardiosignal, removing noise, and dividing the to-be-detected single-channel electrocardiosignal into a plurality of single-cycle heart beats based on R wave peak points; carrying out Gramer angle field imaging processing on the single-cycle heart beats and inputting the processed data into a trained deep learning network PCANet, and extracting first emotional depth feature vectors; screening the first emotional depth features based on a Pearson correlation coefficient to obtain second emotional depth features; and inputting the second emotion depth feature into a trained vector classifier for classification to obtain a classification result of the single-cycle cardiac beat, and performing decision voting on classification results of a plurality of single-cycle heart beats corresponding to the to-be-detected single-channel electrocardiosignals to obtain a final emotion recognition result. According to the method, on the premise that the recognition accuracy is guaranteed, the calculation amount is reduced, redundant features are removed, and the problem of dimension disasters is avoided, so that the classification speed is increased, and the calculation cost is saved.

Description

technical field [0001] The present invention relates to the technical field of emotion recognition, in particular to an emotion recognition method and system based on electrocardiographic signals. Background technique [0002] In recent years, the emotion recognition technology for physiological signals has attracted much attention. The research on emotion recognition based on EEG signals has made good progress. The research on emotion recognition of peripheral physiological signals has attracted more and more attention. It has become an important task to mine the deep relationship between peripheral physiological signals and emotions. An emerging craze in the field of emotion recognition. [0003] Feature extraction and classification are the key technologies of emotion recognition. Since the emotional features are not yet clear, it is impossible to directly and accurately extract the features rich in emotional information, which has become a major difficulty in emotional ...

AI Technical Summary

Problems solved by technology

However, the feature extraction based on the deep learning network takes a long time to calculate, and due to the high dimensionality of the obtained deep features, it is easy to cause the disaster of dimensionality.

Images