Analysis Of Human-Computer Interaction Based On EEG And Motor Imagination EEG

HCI(Human-Computer Interaction) technology offer an effective way to achieve a dialogue between man and machine through the input and output device, which is a key technology in the computer field. Research and development of human-computer interaction technology can effectively improve production efficiency, while improving the accessibility of human-computer interaction technology can overcome the difficulties people with disabilities to use computers and the natural exchange with the outside world. HCI also can provide the same using-channel level for a number of specific obstacles in the society, and realize the harmonious development of society. To improve the performance of traditional human-computer interaction, this paper analyzes the EOG and motor imagery EEG, and designed a combination of EOG and EEG Online HCI system, and the HCI system is implemented.This paper presents an adaptive normalized threshold classification method, which can effectively deal with EOG and identify eyeball glance actions. This paper using the LabVIEW platform for real-time online communication with Scan platform, and compiling the real-time processing system, established a real-time cursor control system based on EOG. Online analysis of motor imagery EEG is realized. Finally, combined with EOG and motor imagery EEG, multimodal human computer interaction system is designed.An efficient and accurate analytic technique of the EOG and EEG is the basis of establishing the multimodal human computer interaction system. Based on the EOG has the characteristics of the obvious time domain signal and rapid signal generation, an adaptive normalized threshold method is proposed first. The problem of changing the peak value of EOG to reduce the accuracy rate of the traditional threshold method is solved. Eye scanning action is efficient identified. Then choose the common space model and linear discriminant analysis to extract features and classify the motor imagery EEG. After off-line analysis, it is proved that the analysis method of the EOG and EEG is correct, which provides a theoretical basis for the follow-up study.Offline experiments demonstrate the feasibility of the algorithm of EOG and EEG analysis, and then this paper has carried on the online research. This paper establish online communication by the LabVIEW, and design the online eye cursor control system and the online motor imagery experiment. The accuracy rate of the proposed algorithm and the selected analysis method for online application was demonstrated through the experiment. The online experiment has provided experimental support for the design of human computer interaction system, and this paper designs a multimodal human-computer interaction system based on the EEG and motor imagery EEG. This system consists of three sub modules, namely the character input module, the game entertainment module and the demand monitoring module. The system uses the motor imagery EEG as the control signal to switch module, and the three sub module were controlled by EOG. This system can provide barrier-free human-computer interactive channels for people with disabilities.