• Journal of the Institute of Electronics and Information Engineers
• /
• v.53 no.3
• /
• pp.143-150
• /
• 2016

In this paper, we propose Korean monophthong recognition method optimizing muscle mixing based on facial surface EMG signals. We observed that EMG signal patterns and muscle activity may vary according to Korean monophthong pronunciation. We use RMS, VAR, MMAV1, MMAV2 which were shown high recognition accuracy in previous study and Cepstral Coefficients as feature extraction algorithm. And we classify Korean monophthong by QDA(Quadratic Discriminant Analysis) and HMM(Hidden Markov Model). Muscle mixing optimized using input data in training phase, optimized result is applied in recognition phase. Then New data are input, finally Korean monophthong are recognized. Experimental results show that the average recognition accuracy is 85.7% in QDA, 75.1% in HMM.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.34 no.10
• /
• pp.1393-1401
• /
• 2010

In this experimental study, we have examined the correlation between a human carpal motion and a surface electromyogram. The carpal motion patterns have been identified and the main muscles involved in the carpal motion have been determined by investigating the anatomical structure of a carpal. The torque acting against the carpal motion has been applied by using a device for carpal rehabilitation training, and the surface electromyogram signal corresponding to the torque at the main muscles has been measured. The root-mean-square (RMS) magnitude of the surface electromyogram signal has been calculated and used to analyze the correlation between the surface electromyogram signal and carpal motion. The experimental results have proved that for carpal torque values below $0.1\;N{\cdot}m$, the RMS magnitude of the surface electromyogram signal is linearly proportional to the carpal torque magnitude and that the carpal torque magnitude is linearly proportional to the cross-sectional area of the carpal muscles. Further, the analysis of the contribution of each muscle to the carpal motion has shown that the contribution of the most dominant muscle is consistently 60%. These three results can be applied to develop more sophisticated devices or robots for carpal rehabilitation training.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2018.10a
• /
• pp.866-868
• /
• 2018

본 연구는 표면 근전도를 이용하여 서로 다른 손가락 움직임을 분류하여 일상 생활 속 다양한 사물(e.g, TV, 에어컨 등)을 제어하는 시스템을 개발을 목표로 한다. 손등에 총 4 개의 양극성 전극을 사용하여 피험자 5명으로부터 표면 근전도를 측정하였다. 각 피험자는 검지, 중지, 약지, 소지를 구부리는 동작 및 휴직 상태에 다섯 가지 다른 과제를 각각 3초씩 50회 수행하였으며, 이 때 표면 근전도를 피험자의 손등에서 측정하였다. 측정한 근전도 신호의 분산을 특징으로 추출하여 선형 판별 분석을 적용한 결과 평균 $81.3{\pm}6.3%$의 분류 정확도를 얻을 수 있었다. 추후 분류 정확도 향상을 위한 추가 연구를 통해 시스템의 신뢰도를 더욱 향상시키고 실제 사물을 제어하는 시스템을 개발하고자 한다.

• Journal of Biomedical Engineering Research
• /
• v.20 no.3
• /
• pp.315-321
• /
• 1999

The purpose of this study is to develop a mathematical model for system identification in order to predIct muscle force based on eledromyographic signal. Therefore, a finding of the relalionship between characteristics of electromyographic signal and the corre spondng muscle force should be necessiiry through dynamic, joint model. To develop the dynamic joint model, the upper limb mcludmg the wrist and elbow joint has been considered. The kinematic and dynamic data, such as joint angular displacement, velocity, deceleration along with the moment of inertla, required to establish the dynamic model has been obtained by electrical flexible goniometer which has two degree-of-frcedoms. ln this model, muscle force can be predicted only electromyographs through the relationship between the integrated lorce and the mtegrated electromyographic signal over the duration of muscle contraclion in this study.

• 제어로봇시스템학회:학술대회논문집
• /
• 1986.10a
• /
• pp.641-642
• /
• 1986

We present a method for the estimation of intramuscular electromyographic(EMG) signals from the given surface EMG signals. This method is based on representing the surface EMG signal as an autoregressive(AR) time model with a delayed intramuscular EMG signal as an input. The parameters of the time series model that transforms the intramuscular signal to the surface signal are identified. The identified model is then used in estimating the intramuscular signal from the surface signal.

• Proceedings of the ESK Conference
• /
• 1993.04a
• /
• pp.79-87
• /
• 1993

등척성 수축에서 국소근육피로의 연구에는 표면전극을 사용한 근전도신호의 스펙트럼분석 이 효과적이다. 많은 연구결과에서 자발적 수축동안 근전도신호의 주파수 성분이 근육의 피로에 의해 낮은 쪽으로 천이 된다는 사실이 밝혀졌으며, 이를 근육피로의 지수로 사용하려는 노력이 계속되어 왔다. 본 연구에서는 실제와 유사한 형태의 주기적 작업수행시에 얻어진 근전도신호의 중간주파수 천이양상을 지속적 수축에서 얻어진 중간주파수의 천이양상과 비교하여, 주기적 작업의 피로축적은 지속적 수축에서의 피로축적과 같은 경향성을 지님을 찾아냈으며, 이 결과를 근거로 주기적 수축의 수축지속시간으로 정규화된 개념의 피로지수를 개발하였다. 이러한 피로지수는 정량적으로 제시될 수 있음은 물론, 작업설계와 허용한계의 설정등에 응용될 수 있을 것이다.

• 한국HCI학회:학술대회논문집
• /
• 2009.02a
• /
• pp.418-421
• /
• 2009

This paper presents a human-machine interaction based on a realtime upper limb motion prediction method using surface electromyography (sEMG). The motions were predicted using an artificial neural network algorithm and sEMG signals which are acquired from five muscles, and then a manipulator was controlled to follow after the predicted motions. Upper limb motions were restricted to 2D vertical plane with the contact condition between a user and an end-effector of manipulator. In order to demonstrate the feasibility of the proposed method, experiments using developed method and using a goniometer were performed. The results showed that the proposed real-time motion prediction method can be implemented a human-machine interaction system.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2002.11a
• /
• pp.378-381
• /
• 2002

In this paper, we proposed an electromyography(EMG) based control method of a virtual robot arm as an adaptive human supporting system or remote control system, which consists of an shoulder control part, elbow control part, and wrist control part. The system uses four surface electrodes to acquire the EMG signal from operator. It is shown from the experiments that the EMG patterns during arm motions can be classified sufficiently by using SOM and LVQ. The interface system based on PC environment is constructed to 3-D graphic user interface(GUI) program. Experimental results show that proposed method obtains approximately 94 percent of success in classification.

• Journal of the Institute of Electronics Engineers of Korea SC
• /
• v.47 no.3
• /
• pp.28-39
• /
• 2010

In this paper, a new method to estimate MU (motor unit) location in the short head of BIC (biceps brachii) muscle using surface EMG (electromyogram) is proposed. The SMUAP (single motor unit action potential) is generated from a MU located at certain depth from the skin surface. The depth is referred as MU location. For estimating muscle force precisely, the information of the MU location is required. The reference SMUAPs are simulated based on anatomical structure of human muscle, and compared with acquired real EMG signals using 3-channel surface EMG electrode. The proposed method was compared with the results of previous researchers and verified its accuracy by computer simulation. From the simulation result in case of the MU located in 8[mm], the average estimation error of proposed method was 0.01[mm]. But the average estimation error of Roeleveld's method was 2.33[mm] and Akazawa's method was 1.70[mm]. Therefore the proposed method was more accurate than the methods of previous researchers.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.10 no.10
• /
• pp.1779-1785
• /
• 2006

This paper presents a method of virtual robot arm control by EMG pattern recognition using the proposed hybrid system. The proposed hybrid system is composed of the LVQ and the SOFM, and the SOFM is used for the preprocessing of the LVQ. The SOFM converts the high dimensional EMG signals to 2-dimensional data. The EMG measurement system uses three surface electrodes to acquire the EMG signal from operator. Six hand gestures can be classified sufficiently by the proposed hybrid system. Experimental results are presented that show the effectiveness of the virtual robot arm control by the proposed hybrid system based classifier for the recognition of hand gestures from EMG signal patterns.