基于小波变换的肌电信号识别方法研究

针对肌电信号的非平稳特性,采用小波变换方法对表面肌电信号进行分析,提取小波系数最大值构造特征矢量输入神经网络分类器进行模式识别,经过训练能够成功地从掌长肌和肱桡肌采集的两道表面肌电信号中识别展拳、握拳、前臂内旋、前臂外旋四种运动模式。实验表明,基于小波变换的神经网络分类方法所需的数据短、运算快,对于肌电假肢的控制具有良好的应用前景。     

基于WPKPCA和SVM的SEMG动作识别方法

为了有效提取表面肌电信号SEMG(Surface Electromyographic)的特征,更好的识别人体上肢运动模式,提出了一种小波包核主元分析(WPKPCA)和支持向量机(SVM)相结合的新方法。通过虚拟仪器采集桡侧腕屈肌和肱桡肌两路表面肌电信号,应用小波包核主元分析法对表面肌电信号进行特征提取,采用支持向量机对表面肌电信号特征数据进行分类识别。实验结果表明,采用此方法能够从表面肌电信号中识别出握拳、展拳、手腕内翻和手腕外翻4种动作,更能有效提取表面肌电信号信息,动作识别率高达98%。     

基于PCA与核LDA的表面肌电信号特征识别

针对表面肌电信号的非线性和非平稳性等特点,提出了一种主元分析与核LDA判别分析相结合的表面肌电信号特征识别新方法;通过虚拟仪器采集桡侧腕屈肌和肱桡肌两路表面肌电信号,并提取其特征参数平均绝对值和均方根,采用主元分析法对表面肌电信号特征参数进行压缩降维,应用核LDA判别分析法对降维后的数据进行分类识别;经过实验表明,该方法将表面肌电信号的特征参数由4维降到2维,减小了数据的冗余度,能够成功的从表面肌电信号中识别握拳、展拳、手腕内翻和手腕外翻四种动作,识别率高达96%。     

基于磁场刺激的肌电信号模式识别的研究

对于人体表面肌电（SEMG）信号提出一种新的研究方法，即在磁场刺激下，采用小波变换的方法，对从掌长肌、肱桡肌、尺侧腕屈肌和肱二头肌上采集的4路表面肌电信号进行分析，并提取其6级小波分解系数绝对值累加和的平均值作为信号的特征．构建特征矢量．输入神经网络分类器进行模式识别，经过训练能够成功地识别出握举、展拳、腕内旋、腕外旋、屈腕、伸腕、前臂内旋、前臂外旋8种运动模式．实验结果表明，该方法识别率高，所需数据量少．运算速度快，实时性好，为肌电等生物电信号的研究提供了一种新方法．     

非线性PCA在表面肌电信号特征提取中的应用

针对表面肌电信号的特点,提出了一种应用非线性主分量分析(PCA)提取表面肌电信号特征的新方法.该方法在表面肌电信号滤波的基础上,采用非线性PCA方法完成数据压缩,将多路表面肌电信号转换为一维的特征数据主元,并以主元曲线的形式输出特征提取结果.本文采用基于自组织神经网络的非线性PCA对手臂尺侧腕伸肌和尺侧腕屈肌的两路表面肌电信号进行主元提取,试验结果表明,四种手部运动模式(握拳、展拳、腕外旋、腕内旋)对应的表面肌电信号利用该方法处理后,得到的主元曲线具有很好的类区分性,依据所得主元曲线的形状特征可以有效地进行手部动作类别的识别.     

多重分形分析在肌电信号模式识别中的应用

为提高肢体运动模式识别率,论文提出了一种经验模态分解与多重分形分析相结合的模式识别方法。先用经验模态分解得到代表肌电信号细节的多层内在模态函数,然后在内在模态函数上进行多重分形分析,提取其广义维数谱,作为肌电信号多模式识别的特征向量。最后以改进的K最近邻分类方法-KNN模型增量学习算法,实现对动作模式的识别。在对张开、合拢及腕内旋、腕外旋4个动作的识别实验中,正确识别率达到了93.0%。结果表明,方法具备一定的实用性,可用于遥操作机器人系统中操作者手腕运动模式识别。     

基于有监督Kohonen神经网络的步态识别

表面肌电信号随着时间的变化而改变,这将影响运动模式的分类精度.传统人体下肢假肢运动模式的识别算法不能保证在整个肌电控制时间内达到对运动模式的有效识别.为了解决这些问题,本文提取步态初期200ms的信号的特征值,将无监督和有监督的Kohonen神经网络算法应用到大腿截肢者残肢侧的步态识别中,并与传统BP神经网络进行了对比.结果表明,有监督的Kohonen神经网络算法将五种路况下步态的平均识别率提高到88.4%,优于无监督的Kohonen神经网络算法和BP神经网络.     

基于小波包变换的肌电信号特征提取

本文提出一种新的基于小波包变换的特征提取方法,提取表面肌电信号进行小波包变换后得到的信号的协方差矩阵的特征值的最大值作为特征值。利用该方法对表面肌电信号提取特征值构建特征矢量,送入Elman神经网络对手部6种动作模式进行识别,在Matlab平台上进行实验仿真。实验结果表明,该方法取得了很好的识别效果。     

结合距离分类器的神经网络手写体汉字识别

手写体汉字识别技术中如何解决复杂的大类别识别问题,是汉字识别中的一个难点。该文介绍了基于笔划的手写体汉字特征抽取方法,提出了一种基于预分类的神经网络汉字识别方法,该方法用一个传统的距离分类器先对汉字进行预分类,神经网络根据预分类结果进行有选择的训练和识别,能有效解决神经网络大类别模式识别中的训练和分类问题,学习时间很短,识别效果较理想。     

基于多特征融合和卷积神经网络的植物叶片识别

植物叶片识别是植物自动分类识别研究的重要分支和热点,利用卷积神经网络进行图像分类研究已成为主流.为了提高植物叶片识别准确率,提出了基于多特征融合和卷积神经网络的植物叶片图像识别方法.首先对植物叶片图像进行预处理,提取LBP特征和Gabor特征,将多特征相加融合输入网络进行训练,使用卷积神经网络(AlexNet)构架作为分类器,利用全连接层对植物叶片进行识别.为了避免过拟合现象,使用"dropout"方法训练卷积神经网络,通过调节学习率、dropout值、迭代次数优化模型.实验结果表明,基于多特征融合的卷积神经网络植物叶片识别方法对Flavia数据库32种叶片和MEW2014数据库189种叶片识别分类效果较好,平均正确识别率分别为93.25％和96.37％,相比一般的卷积神经网络识别方法,该方法可以提高植物叶片的识别准确率,鲁棒性更强.     

基于眼动辅助脑电信号的手部动作分类方法

提出一种利用小波包变换和支持向量机对手部动作的运动想象脑电信号进行分类的方法。在相关眼动辅助情况下采集想象手部动作时的C3、C4 、P3和P4通道脑电信号,用小波包变换的方法提取4种特征节律波,分别计算每种节律波能量占4种节律波能量之和的比值作为特征,然后将16维特征向量输入支持向量机分类器进行手部动作分类。对上翻、下翻、展拳、握拳4种手部动作的分类实验中平均识别率为82。3%,表明眼动辅助能有效提高运动想象脑电信号可分性。     

一种基于WPT和LVQ神经网络的手部动作识别方法

针对表面肌电信号(SEMG)的手部动作识别,提出一种采用小波包变换(WPT)和学习向量量化(LVQ)算法的神经网络分类器。对SEMG信号进行基于熵准则的最优小波包基分解得到各个节点分解系数,计算信号各个节点相应子频段的系数能量,归一化处理后的特征向量输入LVQ神经网络,实现基于SEMG的手部动作识别。实验结果表明,采取两路SEMG信号,该分类器能有效识别伸腕、屈腕、展拳和握拳4种动作模式,达到96%的识别率,能可靠应用于2个自由度肌电假手的控制。     

Hand shape classification in various pronation angles using a wearable wrist contour sensor

Hand gestures are potentially useful for communications between humans and between a human and a machine. However, existing methods entail several problems for practical use. We have proposed an approach to hand shape recognition based on wrist contour measurement. Especially in this paper, two assignments are addressed. First is the development of a new sensing device in which all elements are installed in a wrist-watch-type device. Second is the development of a new hand shape classifier that can accommodate pronation angle changes. The developed sensing device enables wrist contour data collection under conditions in which the pronation angle varies. The classifier recognizes the hand shape based on statistics produced through data forming and statistics conversion processes. The most important result is that no large difference exists between classification rates that include or those that exclude the independent (preliminary) pronation estimation process using inertia measurement units. This result shows two possible insights: (1) the wrist contour has some features that depend on the hand shape but which do not depend on the pronation angle, or (2) the wrist contour potentially includes pronation angle variation information. These insights indicate the possibility that hand shape can be recognized solely from the wrist contour, even while changing the pronation angle.     

Hand Gesture Recognition Based Omnidirectional Wheelchair Control Using IMU and EMG Sensors

This paper presents a hand gesture based control of an omnidirectional wheelchair using inertial measurement unit (IMU) and myoelectric units as wearable sensors. Seven common gestures are recognized and classified using shape based feature extraction and Dendogram Support Vector Machine (DSVM) classifier. The dynamic gestures are mapped to the omnidirectional motion commands to navigate the wheelchair. A single IMU is used to measure the wrist tilt angle and acceleration in three axis. EMG signals are extracted from two forearm muscles namely Extensor Carpi Radialis and Flexor Carpi Radialis and processed to provide Root Mean Square (RMS) signal. Initiation and termination of dynamic activities are based on autonomous identification of static to dynamic or dynamic to static transition by setting static thresholds on processed IMU and myoelectric sensor data. Classification involves recognizing the activity pattern based on periodic shape of trajectories of the triaxial wrist tilt angle and EMG-RMS from the two selected muscles. Second order Polynomial coefficients extracted from the sensor trajectory templates during specific dynamic activity cycles are used as features to classify dynamic activities. Classification algorithm and real time navigation of the wheelchair using the proposed algorithm has been tested by five healthy subjects. Classification accuracy of 94% was achieved by DSVM classifier on ‘k’ fold cross validation data of 5 users. Classification accuracy while operating the wheelchair was 90.5%.     

基于多策略的异构数据库知识获取分析与实现*

为对异构数据库中的大量孤立、没有语义描述的数据进行自动归类及本体建模,实现异构数据库数据的知识获取,提出了一个基于本体与Web服务的异构数据库知识获取框架,给出了通过Web服务包装异构数据库的访问机制,设计出贝叶斯分类器并应用该分类器对获取的异构数据自动映射到相关的本体.该方法能够通过贝叶斯分类器自动对异构数据归类,实现了异构数据库的交互知识获取.     

Comprehensive features with randomized decision forests for hand segmentation from color images in uncontrolled indoor scenarios

Multimedia Tools and Applications - Hand segmentation is an integral part of many computer vision applications, especially gesture recognition. Training a classifier to classify pixels into hand or...     

基于人工鱼群算法优化SVM的手部动作sEMG识别

为了提高人体手部运动模式识别的准确性,提出了一种基于人工鱼群算法优化支持向量机( SVM)的模式识别方法.该方法对采集的表面肌电信号( sEMG)去噪后提取小波系数最大值作为特征样本,将提取后的特征输入到SVM进行动作模式识别,同时采用人工鱼群算法优化SVM( AFSVM)的惩罚参数和核函数参数,避免参数选择的盲目性,提高模型的识别精度.通过对内翻、外翻、握拳、展拳四种动作仿真结果表明:该方法与传统的SVM方法相比具有更高的识别率.     

基于张量分解的运动想象脑电分类算法

论文结合短时傅里叶变换(short time Fourier transform,STFT)和平行因子分解(parallel factor,PARAFAC)模型,对运动想象脑电(EEG)进行分类。首先,通过短时傅里叶变换获得左、右手运动想象脑电的时频分布,然后平行因子分解方法从构建的张量数据中提取时域特征,最后采用贝叶斯分类器对特征进行分类。在短时傅里叶变换中,选择合理的窗函数长度和相邻片段重叠程度很重要,对分类结果有较大的影响,通过调整这两个参数,论文提出的方法获得了较好的分类结果。     

Differentiating types of muscle movements using a wavelet based fuzzy clustering neural network

Abstract: The electromyographic signals observed at the surface of the skin are the sum of many small action potentials generated in the muscle fibres. After the signals are processed, they can be used as a control source of multifunction prostheses. The myoelectric signals are represented by wavelet transform model parameters. For this purpose, four different arm movements (elbow extension, elbow flexion, wrist supination and wrist pronation) are considered in studying muscle contraction. Wavelet parameters of myoelectric signals received from the muscles for these different movements were used as features to classify the electromyographic signals in a fuzzy clustering neural network classifier model. After 1000 iterations, the average recognition percentage of the test was found to be 97.67% with clustering into 10 features. The fuzzy clustering neural network programming language was developed using Pascal under Delphi.