左右手运动想象脑—机接口系统的分析与研究

发布时间：2018-10-08 21:54

【摘要】：脑—机接口（Brain-Computer Interface，BCI）是指在人脑和计算机或其他电子设备之间建立的直接的交流和控制通道。作为一种新颖的人机接口方式，脑—机接口使瘫痪的患者看到了与外界自主交互的新希望。无创式脑—机接口由于其成本低、对大脑无损伤等特点，是当前脑—机接口研究的主要方向。有研究证实，脑—机接口对有运动障碍患者的恢复性训练有着潜在的临床应用背景，因此研发出具有实时性的脑—机接口在线系统具有十分重要的意义。本课题来源于重庆市科技攻关计划项目(CSTC,2009AC5023)。本文结合国内外研究背景及现状，以左右手运动想象脑电信号作为切入点，对大脑感觉运动皮层的事件相关去同步/同步现象（ERD/ERS）进行了离线分析，寻求适合于脑—机接口在线系统的特征提取及模式识别算法，同时设计并实现了基于左右手运动想象的脑—机接口在线系统。本文主要研究了基于左右手运动想象脑—机接口在线系统的实现方案，其中运动想象脑电特征信号的实时特征提取及分类识别是脑—机接口的关键技术部分。脑-机接口实验中结合生理信号测量系统Active One和一台PC机实现了基于左右手运动想象的脑—机接口在线系统。实验采用无创性的头皮电极采集运动想象脑电信号，，具有较好的应用前景。实验过程中受试者注视视觉刺激器上播放的左右手运动图像，左右手图像代表不同肢体的运动想象刺激，通过Active One系统采集受试者运动想象的脑电信号，对信号进行特征提取及分类识别，分析并判断受试者想象的肢体。采用基于滑动窗的时域二阶距估计法提取运动想象脑电特征并从能量的角度进行分析，结合基于Fisher准则函数的线性分类器对提取的特征进行降维与识别。实验结果表明，该脑—机接口系统能有效地提取出左右手运动想象特征并分类识别，能够判断出受试者所想象的肢体，将结果返回到人机交互界面。
[Abstract]:Brain-computer interface (Brain-Computer Interface,BCI) is a direct communication and control channel between human brain and computer or other electronic devices. As a novel man-machine interface, brain-computer interface (BCI) offers paralyzed patients a new hope for autonomous interaction with the outside world. Non-invasive brain-computer interface (BCI) is the main research direction of brain-computer interface (BCI) because of its low cost and no damage to the brain. Some studies have confirmed that BCI has a potential clinical application background for patients with motor disorders, so it is of great significance to develop a real time BCI online system. This topic comes from Chongqing Science and Technology key Project (CSTC,2009AC5023). In this paper, based on the research background and present situation at home and abroad, the event related desynchronization / synchronization (ERD/ERS) in the sensorimotor cortex is analyzed offline by using the left and right hand motion imagination EEG as the starting point. The feature extraction and pattern recognition algorithms suitable for the online brain-computer interface system are sought. At the same time, an on-line brain-computer interface system based on left-right motion imagination is designed and implemented. In this paper, we mainly study the realization scheme of the online system based on the left and right hand motion imagining brain-computer interface, in which the real-time feature extraction and classification recognition of the motion imagination EEG feature signal is the key technology part of the brain-computer interface. In the brain-computer interface experiment, an online brain-computer interface system based on left-right hand motion imagination is realized by combining the physiological signal measurement system (Active One) and a PC machine. The experiment uses non-invasive scalp electrode to collect motion imagination EEG signal, which has a good application prospect. In the course of the experiment, the subjects gazed at the left and right hand motion images played on the visual stimulator, and the left and right hand images represented the motion imagination stimuli of different limbs. The EEG signals of the subjects' motion imagination were collected through the Active One system. The feature extraction and classification of the signals were carried out, and the imaginary limbs of the subjects were analyzed and judged. The second order distance estimation method based on sliding window is used to extract the motion imaginary EEG features and analyze them from the point of view of energy. The extracted features are reduced and recognized by a linear classifier based on Fisher criterion function. The experimental results show that the brain-computer interface system can effectively extract the left and right hand motion imagination features and classify and recognize, can judge the subjects' imaginary limbs, and return the results to the human-computer interface.
【学位授予单位】：重庆大学
【学位级别】：硕士
【学位授予年份】：2012
【分类号】：TP334.7

【参考文献】