基于SEMG分析的交互意图感知方法

针对以往依赖鼠标、键盘等传统设备的交互方式,其易受到各种场景和使用环境的限制,已成为虚拟现实以及新型显示技术发展的屏障,因此提出了一种基于SEMG分析的交互意图感知方法。由于连续表面肌电信号的实时识别不能通过单独的动作产生的活动段进行信号的分割识别,采用一种连续表面肌电信号的上下文分割思想进行实时信号识别。最后对识别出的信号进行模糊决策的交互意图分类,将识别的信号数据对设备进行交互感知控制。通过实验分析可知,基于SEMG分析的人机交互能够较好地感知识别人的不同意图动作,交互识别正确率能够达到95%以上。

Interactive Intent-aware Method Based onAnalysis of the SEMG

The way relies on traditional interaction devices such as mouse, keyboard is easily subject to the constraints of operating environment and different scenes, it has become a barrier which prevent the development of VR and new display technology. Thus an interactive intent-aware method based on analysis of SEMG method is presented in this paper. Due to the real-time recognition of continuous surface EMG signals could not split or recognize signal by using the active segment result from singular activities. Use a context segmentation method of continuous surface EMG signal to do the real-time signal identification. At last, classified interactive intentions of fuzzy decision of the identified signal. Interact and control with devices by using data of the identified signals. Human-computer interaction which is based on SEMG analysis method could be used to identify different intentions well through analyzing experiments. The accuracy rate of interactive identification can be above 95%.