Emotion recognition method and device based on wearable device and electronic device

Abstract

The application discloses an emotion recognition method and device based on a wearable device and electronic equipment, and relates to the fields of artificial intelligence and wearable technology. The method comprises the following steps: collecting initial electrocardio signals, initial respiratory signals and motion data of a to-be-recognized object through a wearable device; removing motion artifacts from the initial electrocardio signals and the initial respiratory signals respectively by using the motion data to obtain electrocardio signals and respiratory signals; extracting time-frequency features from the electrocardio signals and the respiratory signals respectively, and performing fusion processing on the extracted features to generate to-be-recognized features; and inputting the to-be-recognized features into a pre-trained emotion recognition model to determine an emotion recognition result by the emotion recognition model. The application uses the synchronously collected motion data as a reference signal, and fuses the features of the electrocardio and respiratory modalities, thereby improving the accuracy and system robustness of emotion recognition.

Images