基于脑电信号的思维任务分类

脑电信号（EEG）是研究脑活动的一种重要的信息来源，基于脑电信号的人与计算机的通信已成为一种新的人机接口方式。文中主要对不同心理作业的思维脑电信号运用独立分量分析进行预处理，然后采用AR模型提取特征，最后应用BP神经网络对AR系数特征进行训练和分类。实验表明，此方法可以达到很好的分类效果。     

基于脑电信号的思维任务分类

脑电信号(EEG)是研究脑活动的一种重要的信息来源,基于脑电信号的人与计算机的通信已成为一种新的人机接口方式。文中主要对不同心理作业的思维脑电信号运用独立分量分析进行预处理,然后采用AR模型提取特征,最后应用BP神经网络对AR系数特征进行训练和分类。实验表明,此方法可以达到很好的分类效果。     

基于稳态视觉诱发脑电信号的频率特征提取方法研究

脑电信号分析与处理是脑-机接口技术的关键环节,视觉诱发电位是脑-机接口技术较为常用的一种方法。采用功率谱估计中的自相关法、Welch法和AR模型法对稳态视觉诱发脑电信号进行频率特征提取,根据Fisher线性分类对3种方法提取的特征量进行分类判别。结果表明,AR模型法提取频率特征量的准确率最高。     

基于AR模型的脑－机接口问题研究

在脑一机接口的研究中分类识别技术占有重要地位。将脑电信号中事件去同步化/相同步化现象作为特征信息,深入讨论基于AR模型的自适应算法（AAR）和多变量参数AAR模型算法（MVAAR）在脑电信号特征提取中的应用。结合三种分类器,对这两种算法进行了比较,实验证明两种方法的实验效果都很好,但是MVAAR算法比AAR算法能够达到更高的分类正确率,其阶次一般选取也比较低,数据仿真吻合度高,具有更强的通用性。     

基于MODWT的运动想象脑电信号识别

对运动想象脑电信号进行分类识别,是脑机接口研究中的重要问题。为此,提出一种基于极大重叠小波变换和AR模型的脑电信号分类方法。将脑电信号波形进行极大重叠小波分解,抽取变换系数的统计特征,利用Burg算法提取其3层光滑的8阶AR模型系数以及3层光滑部分的能量曲线特征,将这3类特征进行组合后,使用神经网络、支持向量机及线性判别进行分类和比较。与BCI2003竞赛数据分类精度结果相比,该方法的识别率更高。将模型移植入自行研制的嵌入式脑电信号控制电机转向系统中,该模式识别方法的平均准确度达到了91.3%,可用于嵌入式脑机接口的系统设计。     

ICA在思维脑电特征提取中的应用

简要介绍了独立分量分析（ICA）的基本思想及算法，并将其应用在基于多导思维脑电（mental EEG)的特征提取方面。实验结果表明：ICA可以将脑电信号中包含的心电（ECG）、眼电（EOG）等多种干扰信号成功地分离出来，较好地完成了脑电消噪预处理工作。同时，通过使用ICA方法对不同心理作业的脑电信号进行分析处理，发现了与心理作业相对应的脑电独立分量特征，这些稳定的独立分量特征为心理作业分类和脑一机接口技术提供了新的实现方法。     

一种新的BP学习算法及在运动想象脑电信号分类识别中的应用

对运动想象(MI)脑电信号的正确分类是决定基于运动想象脑电的脑－机接口(BCI)性能的关键因素。为有效地提取MI脑电信号特征、提高分类正确率,提出一种基于单形进化的BP神经网络优化算法(BPSSSE)并运用于MI脑电信号的识别,提取自相关(AR)模型参数和希尔伯特边际谱作为特征输入,通过单形进化算法优化BP神经网络学习性能,实现对MI脑电信号的分类。测试实验中,对BCI竞赛数据进行左右手分类。结果表明在4s~ 8s时间段内平均分类正确率为80.17%,最高分类正确率为87.14%,证明了本文算法在基于MI脑电的脑机交互控制系统中应用研究的有效性和可行性。     

基于小波变换和AR参数模型的脑电信号识别方法

为了实现脑-计算机接口(Brain-computer interface,BCI)系统,对运动脑电信号的特征进行了提取和分类。将多路脑电信号进行CAR(Common average reference)滤波后,利用小波变换和AR参数模型提取特征并使用基于马氏距离的线性判别分析对运动脑电信号进行分类。结果表明,该方法提取的特征向量较好地反应了脑电信号的事件相关去同步(Event-related desynchronization,ERD)和事件相关同步(Event-related synchronization,ERS)的变化时程,为BCI研究中脑电信号的模式识别提供了有效的手段。     

基于CBLSTM算法的脑电信号特征分类

由于传统的脑电信号分类方法识别率较低,且识别率随着脑电信号类别的增加逐渐下降,针对脑电信号时空特征结合的特点,设计了一个多层的卷积双向LSTM型递归神经网络（CBLSTM）分类模型。此分类模型利用多层的卷积神经网络有效提取脑电序列的频域特征,采用双向LSTM提取脑电信号的时域特征,并将脑电信号序列逐帧输入到此分类模型中进行标记,最后输出分类结果。对比研究验证了所提出方法的可行性,实验表明此分类模型平均分类识别率得到了提高,且鲁棒性较好。     

基于HOG特征的脑电地形图分类算法研究

脑电地形图（Brain Electrical Activity Mapping,BEAM），是一种先进的研究脑功能和临床诊断的重要手段，既能进行病理诊断又可进行功能诊断，具有较高的敏感性。通过BEAM判断人在不同高低负荷下的疲劳情况并进行有效分类，能最大程度避免高危从业人员的危险发生。目前，大多数脑力负荷分类方法只是简单地利用脑电信号的四种频段特征进行分类，但分类效果并不理想。在此基础上，提出将脑电信号可视化分析，构建脑电地形图，并将方向梯度直方图（Histogram of Oriented Gradient,HOG）特征应用到BEAM分类中。BEAM是根据各频段功率谱密度值用不同颜色表示的球面头皮展成的平面图形，所以针对BEAM的分类研究是基于图像浅层特征的，而HOG正是图像处理中一种简单有效的浅层特征描述符。在BEAM中，HOG能有效地提取各脑功能区的边缘结构特征，并且能提取到地形图表象和形状的方向分布。首先将采集到的脑电信号进行预处理后，选择三频段脑电特征构建BEAM，进行HOG特征提取及分类任务，并与其他算法进行对比。得到的脑电地形图分类结果表明，提取HOG特征的BEAM分类精...     

Classification of mental tasks using stockwell transform

In recent years, various physiological signal based rehabilitation systems have been developed for the physically disabled in which electroencephalographic (EEG) signal is one among them. The efficiency of such a system depends upon the signal processing and classification algorithms. In order to develop an EEG based rehabilitation or assistive system, it is necessary to develop an effective EEG signal processing algorithm. This paper proposes Stockwell transform (ST) based analysis of EEG dynamics during different mental tasks. EEG signals from Keirn and Aunon database were used in this study. Three classifiers were employed such as k-means nearest neighborhood (kNN), linear discriminant analysis (LDA) and support vector machine (SVM) to test the strength of the proposed features. Ten-fold cross validation method was used to demonstrate the consistency of the classification results. Using the proposed method, an average accuracy ranging between 84.72% and 98.95% was achieved for multi-class problems (five mental tasks).     

Mental tasks discrimination by neural networks with wavelet transform

The study mainly focuses on the analysis of Electroencephalogram (EEG), to classify mental tasks by using features based on wavelet transform. We have used the daubechies family wavelets, level 6, to transform obtained signal from independent component analyzed EEG signal. As Fourier analysis consists of breaking up a signal into sine waves of various frequencies. Similarly, wavelet analysis is the breaking up of a signal into shifted and scaled versions of the original wavelet. Signals with sharp changes might be better analyzed with a Wavelet than with a Fourier transform. It also makes sense that local features can be described well with wavelets that have local extent. This offers improved features to the neural networks obtaining several classified mental tasks. Through several processes, it led us more developed variety mental tasks classification results. We find that the neural networks perform over 75% success resulting with small number of electrodes better than a previous 70% resulting.     

HHT方法在不同思维作业脑电信号分析中的应用

介绍了一种处理非线性、非平稳信号的新方法——HHT的原理及特点,并将其应用于不同思维作业脑电信号分析。实验结果表明,不同思维作业脑电信号经HHT后的HH谱和Hilbert边际谱都差异显著,证明HHT方法对脑电信号处理的可行性。     

Bio-signal-based geometric modeling application for physically disabled users

This paper discusses the challenges in achieving bio-signal-based design environments. While the main motivation of this paper was to provide a user interface for physically disabled people to express their artistic natures, a special emphasis is given on graphical user interface design where bio-signals are the single input source. Among three bio-signal sources investigated—electromyography, electrooculography and electroencephalography (EEG)—stimulus-based human–computer interaction design (EEG feature extraction method) is found to be the most promising for achieving design environments to perform complex tasks. In the proposed stimulus-based brain–computer-interaction application, the user communication with a computer is achieved by coupling intended functionalities with stimuli signals on the computer screen. Constant focus on the intended command stimulates the brain. In return, the brain releases a response signals (steady state visual evoked potential). In theory, brain’s response signals and the stimulus signals are identical. Once successfully identified, the presence of a signal pattern that is identical to the one of the alternative stimulus signals (paired with a command in a user interface) indicates the intention of a user. Since each option is associated with a unique signal pattern, multiple options can simultaneously be offered to users. The main challenge of working with stimulus signals is that the response signals are weak and they are buried inside of highly polluted EEG signals that include brain’s natural activities. In this paper, we introduce a signal processing algorithm based on Lorenz systems of differential equations for identifying the source of stimulus signals. Our experiments strongly suggest that bio-signal-based design environments to perform complex tasks, including geometric modeling can be achieved by utilizing stimulus-based signal processing methodology.     

A local neural classifier for the recognition of EEG patternsassociated to mental tasks

This paper proposes a novel and simple local neural classifier for the recognition of mental tasks from on-line spontaneous EEG signals. The proposed neural classifier recognizes three mental tasks from on-line spontaneous EEG signals. Correct recognition is around 70%. This modest rate is largely compensated by two properties, namely low percentage of wrong decisions (below 5%) and rapid responses (every 1/2 s). Interestingly, the neural classifier achieves this performance with a few units, normally just one per mental task. Also, since the subject and his/her personal interface learn simultaneously from each other, subjects master it rapidly (in a few days of moderate training). Finally, analysis of learned EEG patterns confirms that for a subject to operate satisfactorily a brain interface, the latter must fit the individual features of the former.     

基于多类运动想象任务的EEG信号分类研究

针对多类运动想象脑电信号个体差异性强和分类正确率比较低的问题,提出了一种时-空-频域相结合的脑电信号分析方法:首先利用小波包对EEG原始信号进行分解,根据EEG信号的频域分布提取出运动想象脑电节律,通过“一对多”共空间模式(CSP)算法对不同运动想象任务的脑电节律进行空间滤波提取特征;然后将特征向量输入到“一对多”模式下的支持向量机(SVM)中,并利用判断决策函数值的方法对SVM的输出结果进行融合;最后通过引入时间窗对脑电信号进行时域滤波,消除运动想象开始和结束时脑电的波动,进一步提高信号信噪比和算法的分类效果;实验结果显示:在时间窗为2 s时,平均最大Kappa系数达到了0.72,比脑机接口竞赛第一名提高了0.15,验证了该算法能够有效减小脑电信号个体差异性影响,提高多类识别正确率。     

Mental tasks-based brain–robot interface

This paper describes a Brain Computer Interface (BCI) based on electroencephalography (EEG) that allows control of a robot arm. This interface will enable people with severe disabilities to control a robot arm to assist them in a variety of tasks in their daily lives. The BCI system developed differentiates three cognitive processes, related to motor imagination, registering the brain rhythmic activity through 16 electrodes placed on the scalp. The features extraction algorithm is based on the Wavelet Transform (WT). A Linear Discriminant Analysis (LDA) based classifier has been developed in order to differentiate between the three mental tasks. The classifier combines through a score-based system four LDA-based models simultaneously. The experimental results with six volunteers performing several trajectories with a robot arm are shown in this paper.     

异步BCI系统中运动想象诱发脑电特征识别

针对异步运动想象脑机交互（Brain Computer Interface,BCI）系统中空闲状态检测和不同想象任务分类的问题,在小波变换提取脑电信号特征基础上,设计了阈值判别结合支持向量机的二级分类器。由于大脑想象单侧肢体运动时,会导致同侧和对侧运动皮层脑区EEG信号在μ节律上分别出现事件相关同步和去同步,而大脑处于空闲状态时则无此现象。基于大脑活动的这一特性,提出了小波能量阈值判别法,进行空闲状态检测,径向基核函数和交叉检验的支持向量机方法,进行左、右手运动想象任务分类。结果表明该分类器最佳分类正确率达到了80.7%,且整个时间消耗仅为3.0 s,可以较好地满足异步在线运动想象BCI系统的应用。