Department of Education, Graduate School of Sehan University, Yeongam-gun, Jeollanam-do 58447, Republic of Korea.
Guangdong Medical University, Dongguan 523808, China.
Comput Intell Neurosci. 2022 Apr 12;2022:9257827. doi: 10.1155/2022/9257827. eCollection 2022.
With the rapid development of the Internet of Things and to improve the teaching efficiency of the art classroom, a smart art classroom system based on the Internet of Things is proposed, which can effectively assist in teaching. First, we give the general design of the smart art classroom, including the composition of the hardware and software, and the construction method of the application system. Based on existing technologies such as RFID, smart camera, smart voice, smart terminal, and smart screen interaction, an all-around smart art classroom is constructed. Further, we present the design of an intelligent camera-based classroom assistance system based on face detection and facial expression recognition, which can effectively determine the status of students in class and can be used to assist in reminding teachers of their teaching tasks. Among them, face detection and facial expression recognition algorithms are designed based on different convolutional neural network architectures. Finally, experimental data sets are constructed to verify the accuracy of the used algorithms. The experimental results show that the detection accuracy of classroom faces is better than 95% and the accuracy of expression recognition is 88%, which can meet the application needs of intelligent art classrooms.
随着物联网的快速发展，为了提高艺术课堂的教学效率，提出了一种基于物联网的智能艺术课堂系统，可以有效辅助教学。首先，给出了智能艺术课堂的总体设计，包括软硬件组成和应用系统的构建方法。基于 RFID、智能相机、智能语音、智能终端和智能屏幕交互等现有技术，构建了一个全方位的智能艺术课堂。进一步提出了基于人脸检测和面部表情识别的智能相机课堂辅助系统设计，能够有效判断课堂中学生的状态，辅助教师提醒教学任务。其中，人脸检测和面部表情识别算法是基于不同的卷积神经网络架构设计的。最后，构建实验数据集验证所用算法的准确性。实验结果表明，课堂人脸的检测准确率优于 95%，表情识别的准确率为 88%，能够满足智能艺术课堂的应用需求。
