Neural Computing and Applications

, Volume 31, Issue 12, pp 8455–8462 | Cite as

Surface EMG signals and deep transfer learning-based physical action classification

  • Fatih Demir
  • Varun BajajEmail author
  • Melih C. Ince
  • Sachin Taran
  • Abdulkadir Şengür
Original Article


Human physical action classification is an emerging area of research for human-to-machine interaction, which can help to disable people to interact with real world, and robotics application. EMG signals measure the electrical activity muscular systems, which involved in physical action of human. EMG signals provide more information related to physical action. In this paper, we proposed deep transfer learning-based approach of human action classification using surface EMG signals. The surface EMG signals are represented by time–frequency image (TFI) by using short-time Fourier transform. TFI is used as input to pre-trained convolutional neural network models, namely AlexNet and VGG16, for deep feature extraction, and support vector machine (SVM) classifier is used for classification of physical action of EMG signals. Also, the fine-tuning of the pre-trained AlexNet model is also considered. The experimental results show that deep feature extraction and SVM classification method and fine-tuning have obviously improved the classification accuracy when compared with various results from the literature. The 99.04% accuracy score is obtained with AlexNet fc6 + AlexNet fc7 + VGG16 fc6 + VGG16 fc7 deep feature concatenation and SVM classification. 98.65% accuracy score is performed by fine-tuning of the AlexNet model. We also compare the obtained results with some of the existing methods. The comparisons show that the deep feature concatenation and SVM classification method provide better classification accuracy than the compared methods.


Time–frequency image (TFI) Convolutional neural networks (CNNs) Support vector machines (SVMs) Physical action classification 


Compliance with ethical standards

Conflict of interest

The authors declare that they have no conflict of interest.


  1. 1.
    Durandau G, Sartori M, Bortole M, Moreno JC, Pons JL, Farina D (2016) EMG-driven models of human–machine interaction in individuals wearing the H2 exoskeleton. IFAC-PapersOnLine 49(32):200–203CrossRefGoogle Scholar
  2. 2.
    Furui A, Hayashi H, Nakamura G, Chin T, Tsuji T (2017) An artificial EMG generation model based on signal-dependent noise and related application to motion classification. PLoS ONE 12(6):e0180112. CrossRefGoogle Scholar
  3. 3.
    Sezgin N (2012) Analysis of EMG signals in aggressive and normal activities by using higher-order spectra. Sci World J, Volume 2012, Article ID 478952, 5 pages. CrossRefGoogle Scholar
  4. 4.
    Wang L, Ge KD, Wu JY, Ye Y, Wei W (2018) A novel approach for the pattern recognition of hand movements based on EMG and VPMCD. J Mech Med Biol 18(01):1750115CrossRefGoogle Scholar
  5. 5.
    Podrug E, Subasi A (2015) Surface EMG pattern recognition by using DWT feature extraction and SVM classifier. In The 1st conference of medical and biological engineering in Bosnia and Herzegovina (CMBEBIH 2015), 13–15 March 2015, SarajevoGoogle Scholar
  6. 6.
    Li G (2011) Electromyography pattern-recognition-based control of powered multifunctional upper-limb prostheses. In: Mizrahi J (ed) Advances in applied electromyography. InTech, pp 99–116Google Scholar
  7. 7.
    Gokgoz E, Subasi A (2015) Comparison of decision tree algorithms for EMG signal classification using DWT. Biomed Signal Process Control 18:138–144CrossRefGoogle Scholar
  8. 8.
    Sahu GN, Chaurasia N, Suwalka PP, Bajaj V, Kumar A (2015) HHT based features for discrimination of EMG signals. Information systems design and intelligent applications. Springer, Berlin, pp 95–103CrossRefGoogle Scholar
  9. 9.
    Bajaj V, Kumar A (2015) Features based on intrinsic mode functions for classification of EMG signals. Int J Biomed Eng Technol 18(2):156–167CrossRefGoogle Scholar
  10. 10.
    Mishra VK, Bajaj V, Kumar A, Sharma D, Singh GK (2017) An efficient method for analysis of EMG signals using improved empirical mode decomposition. AEU-Int. J. Electron. Commun. 72:200–209CrossRefGoogle Scholar
  11. 11.
    Sahinbegovic H, Music L, and Alic B (2017) Distinguishing physical actions using an artificial neural network. In: 26th International conference on information, communication and automation technologies, pp 1– 5Google Scholar
  12. 12.
    Jana GC, Swetapadma A, and Pattnaik P (2017) An intelligent method for classification of normal and aggressive actions from electromyography signals. In: 1st International conference on electronics, materials engineering and nano-technology, pp 1–5Google Scholar
  13. 13.
    Abdullah AA, Subasi A, Qaisar SM (2017) Surface EMG signal classification by using WPD and ensemble tree classifiers. In: CMBEBIH. Springer, pp 475–481Google Scholar
  14. 14.
    Sukumar N, Taran S, Bajaj V (2018) Physical actions classification of surface EMG signals using VMD. In: International conference on communication and signal processing. April 3–5, India, pp 715–719Google Scholar
  15. 15.
    Krizhevsky A, Sutskever I, and Hinton GE (2012) Imagenet classification with deep convolutional neural networks. In: Advances in neural information processing systems, pp 1097–1105Google Scholar
  16. 16.
    Simonyan K, Zisserman A (2014) Very deep convolutional networks for large-scale image recognition. Computing Research Repository (CoRR), vol. arXiv:1409.1556
  17. 17.
    Orenstein EC, Oscar B (2017) Transfer learning and deep feature extraction for planktonic image data Sets. In: 2017 IEEE winter conference on applications of computer vision (WACV). IEEEGoogle Scholar
  18. 18.
    Russakovsky O, Deng J, Su H, Krause J, Satheesh S, Ma S, Huang Z, Karpathy A, Khosla A, Bernstein M, Berg AC, Fei-Fei L (2015) Imagenet large scale visual recognition challenge. Int J Comput Vision 115(3):211–252MathSciNetCrossRefGoogle Scholar
  19. 19.
  20. 20.
    Hearst MA, Dumais ST, Osuna E, Platt J, Scholkopf B (1998) Support vector machines. IEEE Intell Syst Appl 13(4):18–28CrossRefGoogle Scholar
  21. 21.
    Vedaldi A, Zisserman A (2012) Efficient additive kernels via explicit feature maps. IEEE Trans Pattern Anal Mach Intell 34(3):480–492CrossRefGoogle Scholar
  22. 22.
    He K, Zhang X, Ren S, and Sun J (2016) Deep residual learning for image recognition. In: Proceedings of the IEEE conference on computer vision and pattern recognition, pp 770–778Google Scholar
  23. 23.
  24. 24.
    Huang G, Liu Z, Van Der Maaten L, Weinberger KQ (2017) Densely connected convolutional networks. In: Proceedings of the IEEE conference on computer vision and pattern recognition, pp 4700–4708Google Scholar
  25. 25.
    Iandola FN, Han S, Moskewicz MW, Ashraf K, Dally WJ, Keutzer K (2016) SqueezeNet: AlexNet-level accuracy with 50× fewer parameters and < 0.5 MB model size. arXiv:1602.07360
  26. 26.

Copyright information

© Springer-Verlag London Ltd., part of Springer Nature 2019

Authors and Affiliations

  1. 1.Electrical and Electronics Engineering Department, Technology FacultyFirat UniversityElazigTurkey
  2. 2.Discipline of Electronics and Communication EngineeringPDPM Indian Institute of Information Technology, Design and ManufacturingJabalpurIndia

Personalised recommendations