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.     

Inertial sensors-based torso motion mode recognition for an active postural support brace*

ABSTRACT

In order to achieve high-level control for an active postural support brace, it is important for a wearable robot to be capable of recognizing human motion intentions. An inertial sensors-based torso motion mode recognition method is proposed in this study. The experiments are conducted to define range of torso motion, capture human motion signals by using four inertial sensors on seven healthy subjects, and utilize a classification method to achieve torso motion recognition based on human intent. Up to sixteen modes for torso motion recognition are investigated, and cascaded classification methods combining a quadratic discriminant analysis (QDA) classifier and a support vector machine (SVM) classifier are deployed. With selected cascaded classification strategies, cross-validation yielded classification accuracies of 95.18% (QDA) and 96.24% (SVM). The obtained results of the study show that inertial sensors based motion recognition is viable to achieve in high recognition accuracy which is promising for future robotic applications.     

基于多源信息融合的手势智能交互系统

利用表面肌电信号（surface ElectroMyoGraphy， sEMG）设计了一个提高分类准确性和快速性的识别系统，用于捕获手势动作并进行人机交互.首先，基于无线肌电测量系统和飞行器自主搭建了智能交互平台；接着，采用滑动时间窗的方法对原始sEMG信号设计短时能量阈值进行信号活动段始末点的确定，从而抑制了动作刚执行时趋势段对识别结果的影响；然后，利用时域统计分析对sEMG信号进行特征分析，并提出了一种融合加速度特征信息和sEMG信号的方法来建立5种手势的分类模型.与仅使用sEMG信息源的方式相比，此方法提高了识别准确率.最后，手势控制飞行器运动的实验证明了本方法的可行性和有效性.     

针对上肢高位截肢者的肌电假肢设计

给出了一种基于人体表面肌电信号控制的上肢假肢系统,服务于基于靶向肌肉神经移植术的高位截肢者.通过解析人体肌电信号包含的动作意图控制假肢的运动.系统包括假肢的机械结构、控制系统的硬件和算法.假肢的机械结构包括肘关节、腕关节和手部.控制系统的硬件包括主控板、驱动板和肌电采集模块.模式识别算法采用基于贝叶斯原理的线性分类器,通过在线实时训练,得到需要的线性分类器参数.根据得到的分类器模型,对采集到的肌电信号进行在线实时识别,并根据识别的结果实时控制相应假肢关节的运动.系统样机在健康受试者身上进行了测试,效果良好.     

基于表面肌电信号的肌肉疲劳状态分类系统

为了实现肌肉疲劳状态的准确检测分类,提出一个完整的基于人体表面肌电（sEMG）信号的肌肉疲劳分类与检测系统。首先,通过AgCl表面贴片电极和高精度模拟前端ADS1299采集人体sEMG信号,进行小波消噪等预处理之后,提取可反映人体肌肉疲劳状态的sEMG信号时域和频域特征。然后,在常用特征如积分肌电图（IEMG）、均方根（RMS）、中值频率（MF）以及平均功率频率（MPF）基础上,为更加精细地刻画人体肌肉疲劳状态,引入sEMG信号的频域特征带谱熵（BSE）;为弥补傅里叶变换分析非平稳信号的不足,引入sEMG信号时频特征——基于经验模态分解-希尔伯特变换（EEMD-HT）的平均瞬时频率。最后,为提高肌肉非疲劳和疲劳状态分类的准确度,利用含突变的粒子群优化算法优化支持向量机（PSO-SVM）并对sEMG进行分类,实现人体肌肉疲劳状态检测。征集15名健康男青年进行sEMG信号采集实验,建立sEMG信号库,提取特征进行分类实验。实验结果表明,所提的系统能够进行高精度sEMG信号采集和肌肉疲劳程度的高准确度分类,分类准确率大于90%。     

上肢经肱骨截肢神经功能重建研究

表面肌电信号是一种安全、非侵入的电生理信息,作为实现直觉控制多功能肌电假肢系统的信息源而被广泛应用。由于经肱骨截肢者截肢的程度较高,残留的肢体肌肉少,缺乏足够的肌电信息源,无法实现多功能肌电假肢的直觉控制。目前现有技术是通过采用靶向肌肉神经功能重建的方法重建缺失肌电信息源。但目前国内尚未有关于截肢者残端神经功能重建方法的相关研究。因此,文章提出一种新型的神经吻合技术——目标神经功能替代术：采用靶向肌肉神经功能重建术与目标神经功能替代术相结合的方法,首次在国内对经肱骨截肢者成功实施了神经功能重建手术,成功建立了经肱骨截肢者神经功能重建模型,重建了因截肢而丧失的肌电信息。并采用高密度肌电技术对术前和术后的手-腕-肘部动作进行肌电信号采集,通过动作分类识别的准确率验证了该手术后肌电信息源重建的可靠性。这些结果初步验证了该方法可以为经肱骨截肢者残肢重建缺失肢体神经功能,并为直觉控制多功能肌电假肢提供潜在的信息源。     

Feature extraction of the first difference of EMG time series for EMG pattern recognition

This paper demonstrates the utility of a differencing technique to transform surface EMG signals measured during both static and dynamic contractions such that they become more stationary. The technique was evaluated by three stationarity tests consisting of the variation of two statistical properties, i.e., mean and standard deviation, and the reverse arrangements test. As a result of the proposed technique, the first difference of EMG time series became more stationary compared to the original measured signal. Based on this finding, the performance of time-domain features extracted from raw and transformed EMG was investigated via an EMG classification problem (i.e., eight dynamic motions and four EMG channels) on data from 18 subjects. The results show that the classification accuracies of all features extracted from the transformed signals were higher than features extracted from the original signals for six different classifiers including quadratic discriminant analysis. On average, the proposed differencing technique improved classification accuracies by 2–8%.     

安徽省利辛县平原区人工林树种分类研究

快速准确地绘制平原区人工林树种分布对研究人工林的生态水文和社会经济效益具有重要的意义。将资源3号（ZY-3）全色波段分别同ZY-3多光谱、哨兵2号多光谱进行融合,在图像分割基础上提取变量,采用分层优化变量组合的随机森林分类方法对安徽省利辛县人工林树种进行分类,并与分类回归树和随机森林相比较。结果表明：利用分层分类方法,平原区的人工林树种分类精度可以达到92%以上;基于哨兵2号和ZY-3融合的光谱特征变量分类精度比ZY-3数据本身的融合提高了2.49%～2.91%;而分别加入纹理变量后,分层分类方法大幅度提高了树种分类精度。因此,基于高分辨率遥感数据的光谱和纹理特征,采用分层分类方法,可以有效提高平原区人工林树种的分类精度。     

Identification of EMG signals using discriminant analysis and SVM classifier

The electromyography (EMG) signal is a bioelectrical signal variation, generated in muscles during voluntary or involuntary muscle activities. The muscle activities such as contraction or relaxation are always controlled by the nervous system. The EMG signal is a complicated biomedical signal due to anatomical/physiological properties of the muscles and its noisy environment. In this paper, a classification technique is proposed to classify signals required for a prosperous arm prosthesis control by using surface EMG signals. This work uses recorded EMG signals generated by biceps and triceps muscles for four different movements. Each signal has one single pattern and it is essential to separate and classify these patterns properly. Discriminant analysis and support vector machine (SVM) classifier have been used to classify four different arm movement signals. Prior to classification, proper feature vectors are derived from the signal. The feature vectors are generated by using mean absolute value (MAV). These feature vectors are provided as inputs to the identification/classification system. Discriminant analysis using five different approaches, classification accuracy rates achieved from very good (98%) to poor (96%) by using 10-fold cross validation. SVM classifier gives a very good average accuracy rate (99%) for four movements with the classification error rate 1%. Correct classification rates of the applied techniques are very high which can be used to classify EMG signals for prosperous arm prosthesis control studies.     

A heterogeneous framework for real-time decoding of bioacoustic signals: Applications to assistive interfaces and prosthesis control

Tongue-movement ear pressure (TMEP) signals provide an unobtrusive, completely non-invasive, wearable and assistive human–machine interface (HMI). The HMI concept is based on monitoring volitional bioacoustic activity generated through prescribed tongue motions. In this paper, a heterogeneous decoding framework is presented, enabling effective real-time polychotomous classification between the various tongue actions and dichotomous discrimination of unintended bioacoustic activity from the volitional signals. Using this customised framework and developed software, the real-time performance was evaluated for both three (six subjects) and four action (four subjects) discrimination, using healthy subjects. Ignoring false negative rejections, the system achieved sensitivities of >90% for three-action discrimination and >80% for four action discrimination, across all tested subjects. The interference rejection (IR) capabilities of the framework were also fully demonstrated, using challenging offline data sets. This included a subset of low frequency interference signals with similar temporal characteristics and frequency distributions as the volitional tongue activity. The IR subsystem achieved an average specificity of 76.2% during three-action discrimination and 79.9% during four-action discrimination. To highlight the potential of the system for substituting or augmenting existing assistive interfaces, a case study is presented demonstrating the utility of TMEP signals for hand prosthesis control. Full tongue control was evaluated against three alternative control strategies, namely natural human-hand manipulation, proportional-based control and a hybrid strategy, when performing an everyday object manipulation task. In all cases, the task was completed with the hybrid strategy performing comparably and even outperforming the proportional-based control strategy.     

A comparative study of wavelet families for classification of wrist motions

The selection of most suitable mother wavelet function is still an open research problem in various signal and image processing applications. This paper presents a comparative study of different wavelet families (Daubechies, Symlets, Coiflets, and Biorthogonal) for analysis of wrist motions from electromyography (EMG) signals. EMG signals are decomposed into three levels using discrete wavelet packet transform. From the decomposed EMG signals, root mean square (RMS) value, autoregressive (AR) model coefficients (4th order) and waveform length (WL) are extracted. Two data projection methods such as principal component analysis (PCA) and linear disciminant analysis (LDA) are used to reduce the dimensionality of the extracted features. Probabilistic neural network (PNN) and general regression neural network (GRNN) are employed to classify the different types of wrist motions, which gives a promising accuracy of above 99%. From the analysis, we inferred that ‘Biorthogonal’ and ‘Coiflets’ wavelet families are more suitable for accurate classification of EMG signals of different wrist motions.     

基于人工神经网络对sEMG信号的手势动作识别

针对利用表面肌电信号(sEMG)对手势动作的肌电信号的研究较少和sEMG信号处理过于复杂的问题,提出了利用人工神经网络和sEMG信号对人的手势动作进行识别研究,引入了MYO硬件设备对新的手势动作sEMG信号采集.利用MYO从手臂上获取每一个手势动作的sEMG信号,提取信号特征值,作为算法的训练数据和测试数据.采用人工神经网络中的反向传递神经网络算法来进行对4种不同手势动作分类,对应目标手指识别率在90.35%.研究结果可以被用来做临床诊断和生物医学的应用以及用于现代硬件的发展和更现代化的人机交互的发展.     

Usage of class dependency based feature selection and fuzzy weighted pre-processing methods on classification of macular disease

In this paper, we propose a new feature selection method called class dependency based feature selection for dimensionality reduction of the macular disease dataset from pattern electroretinography (PERG) signals. In order to diagnosis of macular disease, we have used class dependency based feature selection as feature selection process, fuzzy weighted pre-processing as weighted process and decision tree classifier as decision making. The proposed system consists of three parts. First, we have reduced to 9 features number of features of macular disease dataset that has 63 features using class dependency based feature selection, which is first developed by ours. Second, the macular disease dataset that has 9 features is weighted by using fuzzy weighted pre-processing. And finally, decision tree classifier was applied to PERG signals to distinguish between healthy eye and diseased eye (macula diseases). The employed class dependency based feature selection, fuzzy weighted pre-processing and decision tree classifier have reached to 96.22%, 96.27% and 96.30% classification accuracies using 5–10–15-fold cross-validation, respectively. The results confirmed that the medical decision making system based on the class dependency based feature selection, fuzzy weighted pre-processing and decision tree classifier has potential in detecting the macular disease. The stated results show that the proposed method could point out the ability of design of a new intelligent assistance diagnosis system.     

EMG feature evaluation for improving myoelectric pattern recognition robustness

In pattern recognition-based myoelectric control, high accuracy for multiple discriminated motions is presented in most of related literature. However, there is a gap between the classification accuracy and the usability of practical applications of myoelectric control, especially the effect of long-term usage. This paper proposes and investigates the behavior of fifty time-domain and frequency-domain features to classify ten upper limb motions using electromyographic data recorded during 21 days. The most stable single feature and multiple feature sets are presented with the optimum configuration of myoelectric control, i.e. data segmentation and classifier. The result shows that sample entropy (SampEn) outperforms other features when compared using linear discriminant analysis (LDA), a robust classifier. The averaged test classification accuracy is 93.37%, when trained in only initial first day. It brings only 2.45% decrease compared with retraining schemes. Increasing number of features to four, which consists of SampEn, the fourth order cepstrum coefficients, root mean square and waveform length, increase the classification accuracy to 98.87%. The proposed techniques achieve to maintain the high accuracy without the retraining scheme. Additionally, this continuous classification allows the real-time operation.     

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。     

Analysis and recognition of operations using SEMG from upper arm muscles

Accurate muscular force estimation (from upper arm muscles) based on surface electromyogram forms an important issue in upper limb prosthetic design applications. The whole system consists of surface electrodes, signal acquisition protocols, and signal conditioning at different levels. Labview soft scope was used to acquire the surface electromyogram signal from the designed hardware. The study is concerned with the estimation of characteristics of recorded signals, and for that, statistical techniques of PCA were exercised for verifying the effectiveness of the processed signal against different upper arm motions before its classification. Thereafter, artificial neural network classifier was implemented for the classification surface electromyogram signals with best classification rate of 89.30%. Finally, the processing technique was used to significantly (p < .05) improve classification rate, without much loss of information.     

基于密度分类及组合特征的人数估计算法*

为了克服不同人群密度及所采用特征对人数估计的影响,提出了一种基于人群密度分类及组合特征的人数统计算法。该算法包括离线特征组合选取和在线实时估计两个阶段。在离线阶段,选取密度阈值将图像样本分为高、低密度两类,然后通过实验方法选取最优的特征组合。在线估计阶段首先通过分类器将样本分为高、低密度两类,然后利用离线阶段选取的特征组合训练得到高斯模型,并分别对两类样本进行人数估计。实验结果表明,与不分高低密度相比,平均估计误差由10.6%降至8.1%;与目前主流的人数估计算法相比,本文算法的平均估计误差也更小。     

SAR applications in agriculture A comparison of steep and shallow mode (30° and 53° incidence angles) data

Abstract

The multi-date data set discussed in this paper was acquired over an agricultural site near Melfort, Saskatchewan, in 1983. The Synthetic Aperture Radar (SAR) data are C-band (5.26GHz) with vertical polarization and two incidence angles, 53° and 30°. A comparison between the data at the two incidence angles shows that the pixel statistics of the image data are different for the two angles. Examination of the statistical distribution shows that the 53° data are more influenced by vegetation than the 30° angle data. The classification accuracy of the 53° data was higher than that obtained with the 30° data. In addition, the classification accuracies obtained using multi-date combinations with the 53° data were superior to accuracies obtained using multi-angle combinations (53°+ 30°). An overall improvement in classification accuracy was obtained using a post segmentation field classifier.     

Combined influence of forearm orientation and muscular contraction on EMG pattern recognition

The performance of intelligent electromyogram (EMG)-driven prostheses, functioning as artificial alternatives to missing limbs, is influenced by several dynamic factors including: electrode position shift, varying muscle contraction level, forearm orientation, and limb position. The impact of these factors on EMG pattern recognition has been previously studied in isolation, with the combined effect of these factors being understudied. However, it is likely that a combination of these factors influences the accuracy. We investigated the combined effect of two dynamic factors, namely, forearm orientation and muscle contraction levels, on the generalizability of the EMG pattern recognition. A number of recent time- and frequency-domain EMG features were utilized to study the EMG classification accuracy. Twelve intact-limbed and one bilateral transradial (below-elbow) amputee subject were recruited. They performed six classes of wrist and hand movements at three muscular contraction levels with three forearm orientations (nine conditions). Results indicate that a classifier trained by features that quantify the angle, rather than amplitude, of the muscle activation patterns perform better than other feature sets across different contraction levels and forearm orientations. In addition, a classifier trained with the EMG signals collected at multiple forearm orientations with medium muscular contractions can generalize well and achieve classification accuracies of up to 91%. Furthermore, inclusion of an accelerometer to monitor wrist movement further improved the EMG classification accuracy. The results indicate that the proposed methodology has the potential to improve robustness of myoelectric pattern recognition.     

Extraction and analysis of multiple time window features associated with muscle fatigue conditions using sEMG signals

In this work, an attempt has been made to differentiate surface electromyography (sEMG) signals under muscle fatigue and non-fatigue conditions with multiple time window (MTW) features. sEMG signals are recorded from biceps brachii muscles of 50 volunteers. Eleven MTW features are extracted from the acquired signals using four window functions, namely rectangular windows, Hamming windows, trapezoidal windows, and Slepian windows. Prominent features are selected using genetic algorithm and information gain based ranking. Four different classification algorithms, namely naïve Bayes, support vector machines, k-nearest neighbour, and linear discriminant analysis, are used for the study. Classifier performances with the MTW features are compared with the currently used time- and frequency-domain features. The results show a reduction in mean and median frequencies of the signals under fatigue. Mean and variance of the features differ by an order of magnitude between the two cases considered. The number of features is reduced by 45% with the genetic algorithm and 36% with information gain based ranking. The k-nearest neighbour algorithm is found to be the most accurate in classifying the features, with a maximum accuracy of 93% with the features selected using information gain ranking.