Open AccessResearch on the recognition of students’ classroom learning state based on facial expressions
Author(s) -
Qiuchen Lin,
Xinmin Lai
Publication year - 2021
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1914/1/012052
Subject(s) - computer science , mathematics education , state (computer science) , quality (philosophy) , affect (linguistics) , teaching method , mechanism (biology) , artificial intelligence , psychology , communication , philosophy , epistemology , algorithm
The learning state of students in the classroom can not only affect the learning effect of students, but also reflect the teaching quality of teachers. In this paper, a new image database of classroom learning state is established according to students’ learning state in real classroom environment, and an improved ResNeSt network model based on spatial attention mechanism is proposed. In this method, spatial attention mechanism is introduced into the ResNeSt network, which is complementary to the channel attention thought of the original network, so as to improve the recognition accuracy of learning state images. Using this method to identify students’ classroom learning state can provide a basis for teaching administrators to evaluate teachers’ teaching quality, as well as provide guidance for teachers to better adjust teaching models and teaching methods, and improve teaching efficiency.