基于多神经网络的电子信息类人机交互式远程教学系统设计

The design of human-computer interactive distance teaching system for electronic information based on multiple neural networks

为进一步提高我国的远程教学的整体质量,针对电子信息类课程,提出一种基于多神经网络的人机交互远程教学系统。通过动态的指尖识别对指尖轨迹进行了跟踪,并通过静态的手势识别实现了指尖轨迹的规划和运行,总体上提高了远程教学系统的精度和效率。实验结果表明,提出的手势识别系统能够实现精确的手势类型识别,平均准确率达到了98%以上;与其他手势识别算法相比,提出的手势识别算法精度更高,识别速度更高,分别为0.013 s和99.52%;在完成各种轨迹任务时,与传统的人机交互教学系统相比,使用的远程教学系统具有更高的效率和精度,平均耗时仅为7.235 s。以上实验结果表明,多神经网络进行人机交互式远程教学系统的设计能够有效提高电子信息类课程的远程教学质量,对于实际的开发设计有一定的工程价值。

In order to further improve the overall quality of distance education in China, a human-computer interactive distance education system based on multiple neural networks is proposed for electronic information courses. The dynamic fingertip recognition realizes the tracking of moving fingertip trajectory, while the static gesture recognition based on convolutional neural network realizes the planning and operation of fingertip trajectory, which generally improves the accuracy and efficiency of the distance learning system. The experimental results show that the gesture recognition system based on convolutional neural network proposed in this study can achieve accurate gesture type recognition, with an average accuracy of more than 98%;Compared with other gesture recognition algorithms, the gesture recognition algorithm proposed in this study has higher accuracy and recognition speed, which are 0.013 s and 99.52% respectively;Compared with the traditional human-computer interactive teaching system, the distance teaching system used in this study has higher efficiency and accuracy when completing various trajectory tasks, and the average time consumption is only 7.235 s. The above experimental results show that the design of human-computer interactive distance education system based on finger tip detection and gesture recognition can effectively improve the quality of distance education, and has certain engineering value for the actual development and design.