中国电化教育

手势识别技术归属于人工智能范畴中的模式识别,指在自然的人机交互状态下,计算机识别人类手势的技术。手势识别技术在教育领域内已得到一定程度的应用,但其在真实课堂中的应用研究却相对缺乏。未来课堂项目一直将建立真实环境下的富信息技术课堂为主要研究目标,因此,我们尝试将手势识别技术融入未来课堂,从人工智能角度提升未来课堂的有效教学能力。该文以设计未来课堂智能教学系统为研究出发点,构建基于卷积神经网络原理的智能算法,提出以未来课堂为环境支撑、以手势识别为技术支持的智能诊断学生差异行为的"FCIT"教学模式,并给出实施细节。最后从优化智能算法与扩展应用范围两个方面对后续研究提出了建议。

Gesture recognition technology belongs to the category of artificial intelligence pattern recognition, refers to the natural human-computer interaction state, computer recognition of human gesture technology. Gesture recognition technology has been applied to a certain extent in the field of education, but its application research in the real classroom is very lack. The future classroom project has always taken the establishment of an information-rich classroom in the real environment as the main research objective.Therefore, we try to integrate gesture recognition technology into the future classroom, so as to improve the effective teaching of the future classroom from the perspective of artificial intelligence. Based on the design of future classroom intelligent teaching system as the starting point, this paper constructs an intelligent algorithm based on the principle of convolutional neural network, proposes a "FCIT" teaching model of intelligent diagnosis of students' different behaviors based on future classroom environment and gesture recognition technology, and gives implementation details. Finally, the future research is prospected from the perspective of optimizing intelligent algorithm and extending application range.