A hand gesture recognition algorithm based on DC-CNN

  • Xiao Yan WuEmail author


In the process of hand gesture recognition, the diversity and complexity of gesture will greatly influence the recognition rate and reliability. In the task of hand gesture recognition, the traditional method based on manual feature extraction is time-consuming, and the recognition rate is low. In order to improve the recognition rate, a novel recognition algorithm based on double channel convolutional neural network (DC-CNN) is proposed. Firstly, the preprocessing, denoising and edge detection of original gesture images are performed to obtain the hand edge images. Secondly, the hand gesture images and the hand edge images are respectively selected as two input channels of the CNN. Each channel contains the same number of convolutional layers and the same parameters, but each has a separate weight. Finally, the feature fusion is performed at the full connection layer and the output result is classified by softmax classifier. Experiments on the Jochen Triesch Database (JTD) and the NAO Camera hand posture Database (NCD) show that the proposed algorithm has improved the rate of hand gesture recognition and has enhanced the generalization ability of the CNN.


CNN Double channel DC-CNN Gesture recognition Deep learning 



  1. 1.
    Abdel-Hamid O, Deng L, Yu D (2013) Exploring convolutional neural network structures and optimization techniques for speech recognition[C]. Interspeech, pp 3366–3370Google Scholar
  2. 2.
    Barros P, Magg S, Weber C et al (2014) A multichannel convolutional neural network for hand posture recognition[C]. In: International conference on artificial neural networks. Springer, Cham, pp 403--410Google Scholar
  3. 3.
    Cai J, Cai JY, Liao XD et al (2015) Preliminary study on hand gesture recognition based on convolutional neural network[J]. Comput Syst Appl 24(4):113–117MathSciNetGoogle Scholar
  4. 4.
    Canny J (1986) Acomputationalapproachtoedgedetection[J]. IEEE Trans Pattern Anal Mach Intell 8(6):679–698CrossRefGoogle Scholar
  5. 5.
    Cao X, Bo H (2016) Study on gesture recognition based on CNN [J]. Microcomputer Appl 35(9):55–57Google Scholar
  6. 6.
    Dong L, Ruan J, Ma Q, Wang L (2012) The gesture identification based on invariant moments and SVM[J]. Image Process Multimed Technol 31(6):32–35Google Scholar
  7. 7.
    Dumoulin V, Visin F (2016) A guide to convolution arithmetic for deep learning[J]Google Scholar
  8. 8.
    Farfade S S, Saberian MJ, Li L-J (2015) Multi-view face detection using deep convolutional neural networks[C]. In: Proceedings of the 5th ACM on international conference on multimedia retrieval, 2015. ACM, pp 643–650Google Scholar
  9. 9.
    Goldberg Y (2017) Neural network methods for natural language processing[J]. Synthesis Lectures on Human Language Technologies 10(1):1–309MathSciNetCrossRefGoogle Scholar
  10. 10.
    He K, Sun J (2015) Fast guided filter[J]. Computer ScienceGoogle Scholar
  11. 11.
    He K, Sun J, Tang X (2010) Guided image filtering. In: ECCV, pp 1–14Google Scholar
  12. 12.
    He K, Sun J, Tang X (2013) Guided image filtering. TPAMI 35(6):1397–1409CrossRefGoogle Scholar
  13. 13.
    Hinton GE, Salakhutdinov RR (2006) Reducing the diensionality of data with neural networks. Science 313(5786):504–507MathSciNetCrossRefGoogle Scholar
  14. 14.
    Hua-Fu LV (2018) Research on the static hand gesture recognition base on convolutional neural network[J]. Modern ComputerGoogle Scholar
  15. 15.
    Jin LW, Zhong ZY, Yang Z (2016) Applications of deep learning for handwritten chinese character recognition: A review[J]. Acta Automatica Sinica 42(8):1125–1141zbMATHGoogle Scholar
  16. 16.
    John V, Mita S, Liu Z et al (2015) Pedestrian detection in thermal images using adaptive fuzzy C-means clustering and convolutional neural networks[C]. In: Proceedings of the 14th IAPR international conference on machine vision applications. IEEE, Tokyo, pp 246–249Google Scholar
  17. 17.
    Lazebnik S, Schmid C, Ponce J (2006) Beyond bags of features: spatial pyramid matching for recognizing natural scene categories[C]. In: 2006 IEEE computer society conference on computer vision and pattern recognition. IEEE, vol 2, pp 2169–2178Google Scholar
  18. 18.
    Le QV (2013) Building high-level features using large scale unsupervised learning[C]. In: 2013 IEEE international conference on acoustics, speech and signal processing (ICASSP). IEEE, pp 8595–8598Google Scholar
  19. 19.
    Liu Y, Yin Y, Zhang S (2012) Hand gesture recognition based on HU moments in interaction of virtual reality[C]. In: International conference on intelligent human-machine systems and cybernetics. IEEE, pp 145–148Google Scholar
  20. 20.
    Murthy GRS, Jadon RS (2010) Hand gesture recognition using neural networks[C]. In: Advance computing conference. IEEE, pp 134–138Google Scholar
  21. 21.
    Nagi J, Ducatelle F, Di Caro GA et al (2011) Max-pooling convolutional neural networks for vision-based hand gesture recognition[C]. In: 2011 IEEE international conference on signal and image processing applications (ICSIPA). IEEE, pp 342–347Google Scholar
  22. 22.
    Ngiam J, Chen Z, Chia D et al (2010) Tiled convolutional neural networks[C]. In: Advances in neural information processing systems, pp 1279–1287Google Scholar
  23. 23.
    Ranzato MA, Poultney C, Chopra S (2007) Efficient learning of sparse representations with an energy-based model. In: Proceedings of the 2007 advances in neural information processing systems. MIT Press, USA, pp 1137–1144Google Scholar
  24. 24.
    Razavian AS, Azizpour H, Sullivan J et al (2014) CNN features off-the-shelf: an astounding baseline for recognition[C]. In: 2014 IEEE conference on computer vision and pattern recognition workshops (CVPRW). IEEE, pp 512–519Google Scholar
  25. 25.
    Scherer D, Muller A, Behnke S (2010) Evaluation of pooling operations in convolutional architectures for object recognition[J]. Artificial Neural Networks–ICANN 2010:92–101Google Scholar
  26. 26.
    Sui Y, Guo Y (2014) Hand gesture recognition based on combing Hu moments and BoF-SURF support vector machine[J]. Appl Res Comput 31(3):953–956Google Scholar
  27. 27.
    Wang L, Liu H, Wang B (2017) Gesture recognition method combining skin color models and convolution neural network [J]. Comput Eng Appl 53(6):209–214Google Scholar
  28. 28.
    Xie SJ, Lu Y, Yoon S et al (2015) Intensity variation normalization for finger vein recognition using guided filter based singe scale retinex[J]. Sensors 15(7):17089–17105CrossRefGoogle Scholar
  29. 29.
    Yamashita T, Watasue T Hand posture recognition based on bottom-up structured deep convolutional neural network with curriculum learning[C]. In: 2014 IEEE international conference on image processing (ICIP). IEEE, pp 853–857, vol 2014Google Scholar

Copyright information

© Springer Science+Business Media, LLC, part of Springer Nature 2019

Authors and Affiliations

  1. 1.Sichuan University of Arts and ScienceDaZhou CityChina

Personalised recommendations