Random dimension arrangement of double branch combined fNIRS emotion classification recognition method

Abstract

The random dimension arrangement double-branch combined fNIRS emotion classification and recognition method comprises the following steps: S110: collecting functional near-infrared spectroscopy signal data of different emotions; S120: signal preprocessing; S130: constructing a functional near-infrared spectroscopy signal emotion classification and recognition model of RDBNet; S140: training the constructed RDBNet network; and S150: giving a classification result according to the functional near-infrared spectroscopy signal emotion classification and recognition method. The present application proposes an emotion recognition model combining random dimension arrangement of multivariate time series processing and fNIRS channel level features and spatial level features, as shown in FIG. 2. This method combines the multivariate time series processing method and the two features (delayed hemodynamic response and different brain region activation modes) of the functional near-infrared spectroscopy signal to design the architecture of the deep learning model. The deep learning algorithm model designed by combining the two characteristics can effectively solve the above problems.

Images