Earticle

초록

영어

The decomposition methods of surface electromyogram (SEMG) signals are mainly based on independent component analysis, blind source separation and the neural network. Because actual signals are the decomposition faced to single-guide signals, so the neural network decomposition method has more advantage. In this paper, we improve the composition of neural network based on the generation principle and decomposition significance of SEMG, and use this network to decompose the signal and to obtain a higher accuracy through the experimental data above. Beside, under medium-low shrinkage level the decomposition algorithm can successfully extract the dissemination information of motor unit action potentials in SEMG.

목차

Abstract
 1. Introduction
 2. EMG Generation Principle
 3. EMG Signal Processing
  3.1. Pretreatment of EMG Signal
  3.2. Activities Segment Extraction
  3.3. Feature Extraction
  3.4. The Training of Neural Network
  3.5. The Combination of Template and the Classification of Active Segments
 4. The Results and the Analyses
 5. Conclusion
 References

키워드

저자

• Bo You [ College of Automation, Harbin University of Science and Technology, Harbin, China ]
• Shoutong Tao [ College of Automation, Harbin University of Science and Technology, Harbin, China ]
• Yi Liu [ College of Automation, Harbin University of Science and Technology, Harbin, China ]
• Hanqing Zhao [ College of Measure-Control Technology and Communication Engineering, Harbin University of Science and Technology, Harbin, China, College of Mechanical Engineering Heilongjiang University of Science and Technology, Harbin, China ]

참고문헌

발행기관

간행물

표지보기 관심저널 등록

• 간행물명

  International Journal of Signal Processing, Image Processing and Pattern Recognition

• 간기

  격월간

• pISSN

  2005-4254

• 수록기간

  2008~2016

• 십진분류

  KDC 505 DDC 605