Brain wave emotion classification method, system and device, medium and terminal

Abstract

The invention belongs to the technical field of machine learning and intelligent man-machine interaction, and discloses a brain wave emotion classification method, system and device, a medium and a terminal.The brain wave emotion classification method comprises the steps that baseline removal preprocessing is conducted on brain wave data; converting an input signal by taking 5 seconds as a unit, calculating to obtain a first-order differential value of a time domain signal, and taking the first-order differential value as an electroencephalogram feature; using the Euclidean distance as a standard, and selecting a sample closer to the target domain from the source domain as an actual training sample; jDA and TCA are utilized to map training set and test set samples into more similar feature spaces, an integrated classifier is adopted to perform classification evaluation, and circulation is performed to achieve a better effect. According to the method, the Euclidean distance is adopted for sample screening, TCA feature space conversion is added in the first circulation, the difference between a source domain and a target domain is greatly reduced, the convergence time of an algorithm is shortened, and cross-subject brain wave emotion recognition is finally achieved.

Description

technical field [0001] The invention belongs to the technical field of machine learning and intelligent human-computer interaction, and in particular relates to a brain wave emotion classification method, system, equipment, medium and terminal. Background technique [0002] Currently, in the field of human-computer interaction, in order to achieve precise and natural interactions, computers and robots must have emotional processing capabilities. Emotion recognition methods vary from facial images, gestures, voice signals to other physiological signals. Among them, EEG signals are directly generated by brain neurons, which are spontaneous and not affected by the subject's subjective consciousness. They have special advantages in some application scenarios. Many scholars have done a lot of research on EEG emotion recognition. [0003] Psychological research shows that there are significant differences in the way individuals feel and express emotions, so different people have ...

AI Technical Summary

Problems solved by technology

[0007] (3) The existing brain wave emotion classification model is too dependent on the subjects, unable to adapt to new individuals, and a large amount of data from different sources is wasted

[0008] The difficulty of solving the above problems and defects is: there are individual differences among different people, how to shorten the differences between them or how to find more extensive EEG features in different source data to obtain more accurate classification results

Images