Method for calculating emotional electrocardiosignal scaling exponent

Abstract

The invention discloses a method for testing the emotional electrocardiosignal scaling exponent. According to the method, movie video clips are adopted for arousing target emotion, and the electrocardio physiological data of four emotions including joy, anger, sadness and fear of a testee is obtained. In the process of watching the video for a second time, the evaluation on the condition that whether the emotion is aroused or not is carried out on the watched emotion arousing video by the testee through press keys, so an emotion experience re-evaluation press key file is obtained. Emotional electrocardiosignals being 60s are intercepted, after preprocessing work of denoising and normalization, a detrending fluctuation analysis algorithm is adopted, indexes reflecting the emotional electrocardiosignal long-range power law relevancy are calculated, then, the scaling exponent is used as emotional features, a Fisher classifier is adopted for carrying out two-class emotional identification, the correct identification rates of the four emotions of joy, anger, sadness and fear are respectively 89.74 percent, 90.1 percent, 70.43 percent and 83.18 percent, and a better identification effect is achieved.

Description

technical field [0001] The invention belongs to the technical field of electronic information, in particular to a method for testing the scale index of an emotional electrocardiogram signal. Background technique [0002] Emotion recognition is one of the key issues in affective computing, and it is the basis for establishing a harmonious human-computer emotional interaction environment. The emotional state of human beings is an internal psychological and physiological change, which is manifested in the form of explicit behavior, expression, and physiological electrical signals. Therefore, emotion recognition is to indirectly deduce the emotional state of a person by analyzing these observable signals. Compared with facial expressions, speech, and body postures, physiological signals can more honestly reflect the working mode of the autonomic nerves of users in emotional states. Therefore, emotion recognition based on physiological signals has attracted the attention of a lar...

AI Technical Summary

Problems solved by technology

[0004] The purpose of the present invention is to provide a method for testing the scale index of emotional electrocardiographic signals, aiming to solve the fact that the existing research on emotional signals ignores the fact that electrocardiographic signals have typical nonlinear signal characteristics, and applies nonlinear methods A little-studied problem with emotional ECG signals

Images