Psychological analysis of autonomous learning basedon facial expression recognition

Abstract

With the proliferation of online learning platform, autonomous learning is increasingly convenient for students. However, the emotions of students may change in online learning environment, bringing uncertainties to the learning effect. This paper carries out a psychological analysis of autonomous learning in online environment based on facial expression recognition. First, a facial expression recognition framework was established, including modules like facial expression acquisition, image pre-processing, feature extraction, andfeature classification and discrimination. Then, the independent component analysis (ICA) was adopted to interpret the Gabor feature vector, and an improved ICA facial expression recognition algorithm was designed based on Gabor wavelet transform, with the aim to recognize images accurately while eliminating high-order statistical redundancy. To verify its performance, the proposed algorithm was applied to match expression features and analyse the psychological changes of randomly selected images from the online teaching library. The results show that our algorithm con recognize the facial expressions accurately in online autonomous learning environment. The research findings help students cope with emotional changes in autonomous learning.