Emotion detection method based on synchronous characteristics of ECG and respiratory signals

Abstract

The invention relates to an electrocardiograph and respiratory signal synchronous characteristic based emotion detection method. The method includes: acquiring ECG (electrocardiograph) signals and respiratory signals; preprocessing; extracting characteristics including 1) time domain synchronous characteristics, 2) frequency domain synchronous characteristics and 3) phase lock valves; assessing separability of each characteristic according to a Fisher discriminant ratio, calculating to finally obtain FDR (Fisher discriminant ratio) values of five characteristics, and distributing corresponding weight value wi to each characteristic according to the FDR values; adopting a SVM (support vector machine) for establishing an emotion recognition model to recognize a current emotion state of a user. The electrocardiograph and respiratory signal synchronous characteristic based emotion detection method has the advantage that accurate real-time emotion monitoring can be realized.

Description

technical field [0001] An emotion detection method based on the synchronization characteristics of ECG and respiratory signals is proposed. The invention relates to an emotional state identification method that can be used in the diagnosis and curative effect evaluation of clinical emotional disorders, and emotional neurofeedback regulation. Background technique [0002] Emotion is a comprehensive state produced by people whether objective things meet their own needs. As a high-level function of the human brain, it ensures the survival and adaptation of organisms, and affects human learning, memory and decision-making to varying degrees. In people's daily work and life, the role of emotion is everywhere. Negative emotions can affect our physical and mental health, reduce work quality and efficiency, and in severe cases can cause mental illness (such as depression, autism, etc.), and also cause serious work mistakes. Studies have shown that the long-term accumulation of ne...

AI Technical Summary

Problems solved by technology

At present, the commonly used physiological signals include electroencephalogram signal (EEG), electrocardiogram signal (ECG), electrodermal signal, respiratory signal (RSP), electromyographic signal (EMG), pulse signal, etc., but the current research is based on the above Physiological signals alone or a simple fusion of two or more signals for emotion recognition, the accuracy rate still needs to be improved

Although the EEG signal is rich in information, it is extremely vulnerable to external interference

Images