Abstract
The emotion recognition of learners in the classroom plays an important role in improving classroom efficiency. At present, the recognition methods based on traditional image processing generally have problems such as low recognition accuracy, difficulty in feature extraction. In order to effectively solve the above problems, a deep learning-based method is proposed to perform emotion recognition for learners. This method first replaces the DarkNet-53 of YOLO v3 by introducing the convolution structure of MobileNet, which makes the model lighter and reduces the parameters. Then improve the loss function of the model by using GIoU loss. Finally, K-means clustering algorithm is used to obtain an anchor suitable for the emotion recognition scene on the self-made dataset. Thus, a high-accuracy and lightweight learner emotion recognition model ER_YOLO is obtained. Experimental results show that the improved model mAP increased by 4%, F1 score increased by 3.2%, detection time reduced by 1/3, and parameters reduced by 1/10.
Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
