In the classroom, student state changes are subtle and key features are difficult to capture. Therefore, a real-time collection and dynamic analysis method based on hybrid intelligence is proposed to detect abnormal learning states in a timely manner. By using image enhancement technology to preprocess and collect images, the enhanced images are input into a convolutional neural network. After convolutional processing, deep level feature maps and key feature maps are extracted, and further input into a long short-term memory network to achieve dynamic analysis and recognition of the learning status of classroom students. Through experimental verification, this method can provide real-time feedback on the learning status of students, with good recognition effect and high stability. With the help of this method, students' academic performance can be effectively improved based on the correction of relevant academic personnel, and it has certain application value.
JiangYu-chen, HeHe, LiuTao, et al. Research on student learning state monitoring technology for library based on artificial intelligence[J]. Electronic Devices, 2023, 46(4): 1070-1074.
LinMing-wei, XuJiang-song, LinJia-yin, et al. A review of emotion recognition of learners for online education[J]. Control and Decision, 2024, 39(4): 1057-1074.
WangTao-chun, QiuQing, WangCheng-tian, et al. LD-identify:network learning state recognition based on passive RFID[J]. Control and Decision,2024, 39(1): 219-226.
SuYue, YangChun-jin, WangJian-xia. Detection of students' classroom states based on expression recognition technology[J]. Hebei Industry Science and Technology,2022,39(5):396-403.
[9]
ChenZ, LiangM, XueZ,et al.STRAN: student expression recognition based on spatio-temporal residual attention network in classroom teaching videos[J].Applied Intelligence: The International Journal of Artificial Intelligence, Neural Networks, and Complex Problem-Solving Technologies,2023,53(21):25310-25329.
GuWen-juan, DingCan, WeiJin, et al. Low-light image enhancement based on the fusion of bilateral filter MSR and AutoMSRCR[J]. Optics and Precision Engineering, 2023, 31(24): 3606-3617.
RenYan-li, YuLing-zan, HeGang, et al. A scheme of privacy-preserving convolutional neural network prediction[J]. Chinese Journal of Computers,2023, 46(8): 1606-1619.
MaoXin-xin, WuSheng-xi, XianBo-long, et al. Adaptive graph convolution and LSTM action recognition based on skeleton[J]. Journal of East China University of Science and Technology(Natural Science Edition), 2022, 48(6): 816-825.
YuJin-shuo, LuXian-ling. Human action recognition algorithm of feature fusion CNN-Bi-LSTM based on split-attention[J]. Journal of Electronic Measurement and Instrumentation, 2022, 36(2): 89-95.