• Journal of the Institute of Electronics and Information Engineers
• /
• v.53 no.4
• /
• pp.142-148
• /
• 2016

Recently, human identification using various biological signals has been studied and human identification based on the gait has been actively studied. In this paper, we propose a human identification based on the EMG(Electromyography) signal of the thigh muscles that are used when walking. Various features such as RMS, MAV, VAR, WAMP, ZC, SSC, IEMG, MMAV1, MMAV2, MAVSLP, SSI, WL are extracted from EMG signal data and ANN(Artificial Neural Network) classifier is used for human identification. When we evaluated the recognition ratio per channel and features to select approptiate channels and features for human identification. The experimental results show that the rectus femoris, semitendinous, vastus lateralis are appropriate muscles for human identification and MAV, ZC, IEMG, MMAV1, MAVSLP are adaptable features for human identification. Experimental results also show that the average recognition ratio of method of using all channels and features is 99.7% and that of using selected 3 channels and 5 features is 96%. Therefore, we confirm that the EMG signal can be applied to gait based human identification and EMG signal based human identification using small number of adaptive muscles and features shows good performance.

• Biomedical Engineering Letters
• /
• v.8 no.4
• /
• pp.345-353
• /
• 2018

This study suggested a new EMG-signal-based evaluation method for knee rehabilitation that provides not only fragmentary information like muscle power but also in-depth information like muscle fatigue in the field of rehabilitation which it has not been applied to. In our experiment, nine healthy subjects performed straight leg raise exercises which are widely performed for knee rehabilitation. During the exercises, we recorded the joint angle of the leg and EMG signals from four prime movers of the leg: rectus femoris (RFM), vastus lateralis, vastus medialis, and biceps femoris (BFLH). We extracted two parameters to estimate muscle fatigue from the EMG signals, the zero-crossing rate (ZCR) and amplitude of muscle tension (AMT) that can quantitatively assess muscle fatigue from EMG signals. We found a decrease in the ZCR for the RFM and the BFLH in the muscle fatigue condition for most of the subjects. Also, we found increases in the AMT for the RFM and the BFLH. Based on the results, we quantitatively confirmed that in the state of muscle fatigue, the ZCR shows a decreasing trend whereas the AMT shows an increasing trend. Our results show that both the ZCR and AMT are useful parameters for characterizing the EMG signals in the muscle fatigue condition. In addition, our proposed methods are expected to be useful for developing a navigation system for knee rehabilitation exercises by evaluating the two parameters in two-dimensional parameter space.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.57 no.3
• /
• pp.527-535
• /
• 2008

Many signal processing techniques have been described in the literature for estimating amplitude, frequency and duration variables of the surface EMG signal detected during constant voluntary contractions. They have been used in different application areas for the non-invasive assessment of muscle function. The main purpose of our research is to compare the most frequently used algorithms for information extraction from surface EMG signals under varying conditions in terms of the different window lengths, muscle contraction levels, muscles and subjects. In particular we focus on the issue of estimating the slope and intercept to resolve an linear regression line with utilizing real SEMG signals which represents voluntary contractions during thirty seconds.

• Proceedings of the KIEE Conference
• /
• 1992.07a
• /
• pp.468-472
• /
• 1992

In this paper, we classified electromyographic(EMG) signal for prothesis control using neural network. For this study fast Fourier transform(FFT) with ensemble averaged spectrum is applied to two-channeI EMG signal for biceps and triceps. We used the three layer network. And a cumulative back-propagation algorithm is used for classification of six arm functions, flexion and extension of elbow and pronation and supination of the forearm and abduction and adduction of wrist.

• 제어로봇시스템학회:학술대회논문집
• /
• 2001.10a
• /
• pp.32.2-32
• /
• 2001

We have developed the power assistive leg called HAL (Hybrid Assistive Leg) which provide the walking aid for walking disorder persons or aged persons without nursing person. We developed HAL-3 by considering some problems of HAL-1,2 which had developed previously. The mechanism of HAL-3 actuator could be simplified and sophisticated by using the harmonic drive. As the control signal of HAL-3 EMG signal was used. We proposed a calibration method to identify parameters which relates the EMG to joint torque by using HAL-3. We could obtain suitable torque estimated by EMG and realize power assist in walking according to the intention of the operator To the remove discomfort for quick motion power assist, the feedforward controller was installed at the beginning of motion ...

• Journal of IKEEE
• /
• v.24 no.1
• /
• pp.269-274
• /
• 2020

This paper is to study the exoskeleton robot for the walking of the elderly and the disabled. We developed and tested an Exoskeletal robot with two axes of freedom for joint motion. The EEG and EMG signals were used to move the joints of the Exoskeletal robot. By analyzing the EMG signal, the control signal was extracted and applied to the robot to facilitate the walking operation of the walking assistance robot. In addition, the brain-computer interface technology is applied to perform the operation of the robot using brain waves, spontaneous electrical activities recorded on the human scalp. These two signals were fused to study the walking recognition method of the supporting robot leg.

• Proceedings of the KIEE Conference
• /
• 1998.07g
• /
• pp.2448-2450
• /
• 1998

For the exact classification of the arm motion this paper proposes EMG pattern recognition method with neural network. For this autoregressive coefficient, linear cepstrum coefficient, and adaptive cepstrum coefficient are selected for the feature parameter of EMG signal, and they are extracted from time series EMG signal. For the function recognition of the feature parameter a radial basis function network, a field of neural network is designed. For the improvement of recognition rate, a number of radial basis function network are combined in parallel, comparing with a backpropagation neural network an existing method.

• The Transactions of the Korean Institute of Electrical Engineers D
• /
• v.52 no.2
• /
• pp.112-119
• /
• 2003

This paper proposes a model of SP(silent period) generation in masseter muscle by means of computer simulation. The model is based on the anatomical and physiological properties of trigeminal nervous system. In determining the SP generation pathway, evoked SPs of masseter muscle after mechanical stimulation to the chin are divided into normal and abnormal group. Normal SP is produced by the activation of mechanoreceptors in periodontal ligament. The activation of nociceptors contributes to the latter part of normal SP, abnormal extended SP is produced. As a result, the EMG signal generated by a proposed SP generation model is similar to both real EMG signal including normal SP and abnormal extended SP with TMJ patients. The result of this study have shown differences of SP generation mechanism between subjects both with and without TMJ dysfunction.

• The Journal of Korea Robotics Society
• /
• v.19 no.2
• /
• pp.213-220
• /
• 2024

Fall detection and prevention technologies play a pivotal role in ensuring the well-being of individuals, particularly those living independently, where falls can result in severe consequences. This paper addresses the challenge of accurate and quick fall detection by proposing a Bayesian probability-based measure applied to surface electromyography (sEMG) signals. The proposed algorithm based on a Bayesian filter that divides the sEMG signal into transient and steady states. The ratio of posterior probabilities, considering the inclusion or exclusion of the transient state, serves as a scale to gauge the dominance of the transient state in the current signal. Experimental results demonstrate that this approach enhances the accuracy and expedites the detection time compared to existing methods. The study suggests broader applications beyond fall detection, anticipating future research in diverse human-robot interface benefiting from the proposed methodology.

• 제어로봇시스템학회:학술대회논문집
• /
• 1990.10a
• /
• pp.559-563
• /
• 1990

A new speed improvement method for quantitative superimposed EMG signal analysis to diagnose the neuromuscular dysfunction is described. The improvement is achieved through the use of efficient software and hardware signal processing techniques. The software approch is composed of the MANDF filter and HRWA algorithm which provides the optimal set and time delays of-selected templates. The hardware employs a TMS32OC25 DSP chip to execute the intensive calculation part. The purposed method is verified through a simulation with real templates which are obtained from needle EMG. As a results, the proposed method provides an overall speed improvement of 32-40 times.