Journal of the HCI Society of Korea (한국HCI학회논문지)

The HCI Society of Korea (한국HCI학회)
권호 표지

Human-Computer Interface using sEMG according to the Number of Electrodes

전극 개수에 따른 근전도 기반 휴먼-컴퓨터 인터페이스의 정확도에 대한 연구

  • 이슬비 (울산대학교 전기전자컴퓨터공학과 의용생체공학전공) ;
  • 지영준 (울산대학교 전기공학부)
  • Received : 2015.03.17
  • Accepted : 2015.06.02
  • Published : 2015.11.30
Download PDF
⟨ Previous Next ⟩

Abstract

NUI (Natural User Interface) system interprets the user's natural movement or the signals from human body to the machine. sEMG (surface electromyogram) can be observed when there is any effort in muscle even without actual movement, which is impossible with camera and accelerometer based NUI system. In sEMG based movement recognition system, the minimal number of electrodes is preferred to minimize the inconvenience. We analyzed the decrease in recognition accuracy as decreasing the number of electrodes. For the four kinds of movement intention without movement, extension (up), flexion (down), abduction (right), and adduction (left), the multilayer perceptron classifier was used with the features of RMS (Root Mean Square) from sEMG. The classification accuracy was 91.9% in four channels, 87.0% in three channels, and 78.9% in two channels. To increase the accuracy in two channels of sEMG, RMSs from previous time epoch (50-200 ms) were used in addition. With the RMSs from 150 ms, the accuracy was increased from 78.9% to 83.6%. The decrease in accuracy with minimal number of electrodes could be compensated partly by utilizing more features in previous RMSs.

NUI(Natural User Interface)는 사용자의 자연스러운 동작이나 동작 시 발생하는 생체 신호를 해석하여 기계에 명령을 내리는 것을 말한다. 물리적인 변화가 있어야 사용이 가능한 가속도 센서나 영상 기반의 NUI와는 달리 특정 동작과 관련된 근육의 표면 근전도(surface Electromyogram, sEMG)를 측정하면 실제 움직임이 발생하지 않아도(isometric contraction) 동작 의도를 예측할 수 있다. 본 연구에서는 근전도 기반으로 손목 동작 의도를 분류할 때 전극 개수에 따른 정확도를 확인하고, 키보드 등에 적용 가능한 인터페이스 기술을 제안한다. 손목의 동작 중 신전(extension, up), 굴곡(flexion, down), 외전(abduction, right), 내전(adduction, left)의 네 가지 동작 의도를 분류하는 실험을 진행하였다. 50ms 간격으로 계산된 제곱평균제곱근(Root Mean Square, RMS)을 특징으로 사용하였고, 동작 의도 인식을 위해 역전파 알고리즘으로 학습한 다층 퍼셉트론 분류기를 사용하였다. 전극 쌍의 개수를 네 개(91.9%), 세 개(87.0%), 두 개(78.9%)로 줄여가며 정확도를 확인했다. 전극 쌍의 개수가 네 개에서 두 개로 줄었을 때 정확도는 약 13% 감소하였다. 두 쌍의 전극만 사용하는 경우의 분류 정확도를 높이기 위하여 직전의 RMS를 특징에 추가하였다. 150 ms 이전까지의 정보를 사용하였을 때, 분류 정확도가 78.9%에서 83.6%로 4.6% 증가하였다. 전극 쌍의 개수가 감소함에 따라 정확도는 감소하였지만, 이전 데이터를 함께 사용한 경우 부분적으로 증가 시킬 수 있음을 확인하였다.

Keywords

References

  1. Kim, J., Park, C., Jeong. J., Baek, N. and Yoo, K. A Gesture Based Camera Controlling Method in the 3D Virtual Space. International Journal of Smart Home. 6(4). Science & Engineering Research Support Society. pp. 117-126. 2012.
  2. Dardas, N. H. and Alhaj, M. Hand Gesture Interaction with a 3D Virtual Environment. The Research Bulletin of Jordan ACM. 2(3). Jordan ACM Chaper. pp. 86-94. 2011.
  3. Shenoy, P., Miller, K., Crawford, B. and Rao, R. Online Electromyographic Control of a Robotic Prosthesis. IEEE Transactions on Biomedical Engineering. 55(3). IEEE. pp. 1128-1135. 2008. https://doi.org/10.1109/TBME.2007.909536
  4. Fougner, A., Scheme, E., Chan, A., Englehart, K. and Stavdahl, O. Resolving the Limb Position Effect in Myoelectric Pattern Recognition. IEEE Transactions on Neural System and Rehabilitation Engineering. 19(6). IEEE. pp. 644-651. 2011. https://doi.org/10.1109/TNSRE.2011.2163529
  5. Choi, C., Micera, S., Carpaneto, J. and Kim, J. Development and Quantitative Performance Evaluation of a Noninvasive EMG Computer Interface. IEEE Transactions on Biomedical Engineering. 56(1). IEEE. pp. 188-191. 2009 https://doi.org/10.1109/TBME.2008.2005950
  6. Khokhar, Z., Xiao, Z. and Menon, C. Surface EMG Pattern Recognition for Real-time Control of a Wrist Exoskeleton. Biomedical Engineering OnLine. 9(41). BioMed Central. pp. 1-17. 2010. https://doi.org/10.1186/1475-925X-9-1
  7. Li, G., Schultz, S. and Kuiken, T. Quantifying Pattern Recognition-Based Myoelectric Control of Multifunctional Transradial Prostheses. IEEE Transactions on Neural System and Rehabilitation Engineering. 18(2). IEEE. pp. 185-192. 2010. https://doi.org/10.1109/TNSRE.2009.2039619
  8. Lee, S. and Chee, Y. A human-computer interface using sEMG by wrist movement. 48th The Korean Society of Medical & Biological Engineering. The Korean Society of Medical & Biological Engineering. pp. 30-32. 2013.
  9. Kim, K., Han, Y., Jung, W., Lee, Y., Kang, J., Choi, H. and Mun, C. Technical Development of Interactive Game Interface Using Multi-Channel EMG Signal. Journal of Korea Game Society. 10(5). Korea Game Society. pp. 65-73. 2010
  10. Jeong, E., Kim, S., Song, Y. and Lee, S. Artificial Neural Network based Motion Classification Algorithm using Surface Electromyogram. Journal of Rehabilitation Engineering and Assistive Technology Society of Korea. 6(1). Rehabilitation Engineering and Assistive Technology Society of Korea. pp. 67-73. 2012
  11. Cui, H., Kim, Y., Shim, H., Yoon, K. and Lee, S. Pattern Classification Algorithm for Wrist Movements Based on EMG. Journal of Rehabilitation Engineering and Assistive Technology Society of Korea. 7(2). Rehabilitation Engineering and Assistive Technology Society of Korea. pp. 67-74. 2013.