Multimedia Tools and Applications

, Volume 75, Issue 21, pp 13041–13056 | Cite as

Human action recognition using modified slow feature analysis and multiple kernel learning

  • Yongliang Xiao
  • Limin Xia


A novel human action recognition method is proposed, which includes two periods of action feature extraction and action recognition. Firstly, we use a modified slow feature analysis (SFA) to extract video local feature. Unlike slow feature analysis, we redefine the objective function with supervised information, which make the modified SFA more suitable to preserve the slow feature and label information. Meanwhile, in effort to cope with the dimension explosion in SFA, locality preserving projections (LPP) is used to reduce the quadratic expansion dimension. Secondly, we use a multiple kernel learning method (MKL) to classify human action, in which the weights of different kernels are optimized by combining Bacterial Chemotaxis method and Powell method. The results of experiments indicate the efficiency of our method.


Human action recognition Slow feature analysis Multiple kernel learning Powell method 



This work is supported by the Postdoctoral Science Foundation of Central South University, the Construct Program of the Key Discipline in Hunan Province, Hunan Province Education and Science Issue “Performance Evaluation for College Teacher Based on Adaptive Learning” (no. XJK013CGD083), the Teaching Reform Research Foundation of Hunan Province Ordinary College under Grant (no. [2014]247-612), and the Research Foundation of Science & Technology Office of Hunan Province under Grant (no. 2014FJ3057).


  1. 1.
    Alain R, Francis RB, Phane C, Yves SG (2008) Simple MKL. J Mach Learn Res 9(3):2491–2521MathSciNetGoogle Scholar
  2. 2.
    Ben AN, Mejdoub M, Chokri BA (2014) Graph-based approach for human action recognition using spatio-temporal features. J Vis Commun Image Represent 25(2):329–338CrossRefGoogle Scholar
  3. 3.
    Benmokhtar R (2014) Robust human action recognition scheme based on high-level feature fusion. Multimed Tools Appl 69(2):253–275CrossRefGoogle Scholar
  4. 4.
    Berkes P, Wiskott L (2005) Slow feature analysis yields a rich repertoire of complex cell properties. J Vis 5(6):579–602MATHCrossRefGoogle Scholar
  5. 5.
    Bobick AF, Davis JW (2001) The recognition of human movement using temporal templates. IEEE Trans Pattern Anal Mach Intell 23(3):257–267CrossRefGoogle Scholar
  6. 6.
    Boser B, Guon I, Vapnik V (1992) A training algorithm for optimal margin classifiers. In: Proceedings of 5th ACM workshop computing learn theoryGoogle Scholar
  7. 7.
    Christodoulou CA, Vita V, Ekonomou L, Chatzarakis GE, Stathopulos IA (2010) Application of Powell’s optimization method to surge arrester circuit models’ parameters. Energy 35(8):3375–3380CrossRefGoogle Scholar
  8. 8.
    Ding H, Bian ZH (2009) A sub-pixel registration approach based on powell algorithm. Acta Photomica Sinica 38(12):46–49Google Scholar
  9. 9.
    Guo P, Miao ZJ, Shen Y (2014) Continuous human action recognition in real time. Multimed Tools Appl 68(3):827–844CrossRefGoogle Scholar
  10. 10.
    He XF, Niyogi P (2003) Locality preserving projections. In: Proceedings of Neural Information Processing SystemsGoogle Scholar
  11. 11.
    Kalinin YV, Jiang LL, Tu YH, Wu MM (2009) Logarithmic sensing in escherichia coli bacterial chemotaxis. Biophys J 96(6):2439–2448CrossRefGoogle Scholar
  12. 12.
    Kim H, Lee SH, Sohn MK, Ju D (2014) Illumination invariant head pose estimation using random forests classifier and binary pattern run length matrix. Hum Centric Comput Inf Sci 9(7):4–9Google Scholar
  13. 13.
    Laptev I (2005) On space-time interest points. Int J Comput Vis 64(2):107–123MathSciNetCrossRefGoogle Scholar
  14. 14.
    Moghaddam Z, Piccardi M (2014) Training initialization of hidden Markov models in human action recognition. IEEE Trans Autom Sci Eng 11(2):394–408CrossRefGoogle Scholar
  15. 15.
    Müller SD, Marchetto J, Airaghi S, Koumoutsakos P (2002) Optimization based on bacterial chemotaxis. IEEE Trans Evol Comput 6(1):16–29CrossRefGoogle Scholar
  16. 16.
    Narang N, Dhillon JS, Kothari DP (2012) Multiobjective fixed head hydrothermal scheduling using integrated predator–prey optimization and Powell search method. Energy 47(1):237–252CrossRefGoogle Scholar
  17. 17.
    Sadek S, Al-Hamadi A, Michaelis B, Sayed U (2010) Toward robust action retrieval. In: Proceedings of 2010 21st British Machine Vision Conference, pp 1–11Google Scholar
  18. 18.
    Schüldt C, Laptev I, Caputo B (2004) Recognizing human actions: a local SVM approach. In: Proceedings of the 17th International Conference on Pattern Recognition, pp 32–36Google Scholar
  19. 19.
    Shen B, Wang WY, Lu ZH, Lu XB, Yu H (2006) Parallel implementation of the global optimization algorithm based on uniform distributional design and Powell method. In: Proceedings of 2006 International Conference on Parallel Processing Workshops, pp 1–6Google Scholar
  20. 20.
    Shon JG, Kim BW (2014) Design and implementation of a content model for m-learning. J Inf Proces Syst 10(4):543–554MathSciNetCrossRefGoogle Scholar
  21. 21.
    Tan LZ, Xia LM, Huang JX, Xia SP (2013) Human action recognition based on pLSA model. J Natl Univ Def Technol 35(5):102–109Google Scholar
  22. 22.
    Theriault C, Thome N, Cord M (2014) Perceptual principles for video classification with slow feature analysis. IEEE J Sel Top Sign Proces 8(3):428–437CrossRefGoogle Scholar
  23. 23.
    Tonaru T, Takiguchi T, Ariki Y (2009) Extraction of human activities as action sequences using pLSA and prefixspan. Int J Hybrid Inf Technol 2(1):13–29Google Scholar
  24. 24.
    Tu HB, Xia LM (2014) The approach for action recognition based on the reconstructed phase spaces. Sci World J Article ID 495071. doi:  10.1155/2014/495071
  25. 25.
    Tu HB, Xia LM, Tan LZ (2013) Adaptive self-occlusion behavior recognition based on pLSA. J Appl Math 2013, Article ID 506752Google Scholar
  26. 26.
    Tuani HT, Li C, Jian Z, Li W (2012) Structured learning of local features for human action classification and localization. Image Vis Comput 30(1):1–14CrossRefGoogle Scholar
  27. 27.
    Tuia D, Camps G, Matasci G, Kanevski M (2010) Learning relevant image feature with multiple-kernel classification. IEEE Trans Geosci Remote Sens 48(2):3780–3791CrossRefGoogle Scholar
  28. 28.
    Uddin J, Islam R, Kim JM (2014) Texture feature extraction techniques for fault diagnosis of induction motors. J Convergence 5(2):15–20Google Scholar
  29. 29.
    Wang YY, Li YB, Ji XF (2014) Human action recognition based on normalized interest points and super-interest points. Int J Humanoid Rob 11(1):329–338Google Scholar
  30. 30.
    Yang GL, Liu S (2014) Distributed cooperative algorithm for k-M Set with negative integer k by fractal symmetrical property. Int J Distrib Sens Netw. doi: 10.1155/ 2014/ 398583 Google Scholar
  31. 31.
    Yang W, Payam S, Greg M (2007) Semi-latent dirichlet allocation: a hierarchical model for human action recognition. Lect Notes Comput Sci 4814(1):240–254Google Scholar
  32. 32.
    Zhang Y, Guo Y, Gu YF, Zhong WZ (2009) Particle swarm optimization with Powell’s direction set method. In: Proceedings of 5th International Conference on Natural Computation, pp 388–392Google Scholar
  33. 33.
    Zhang Z, Tao DC (2012) Slow feature analysis for human action recognition. IEEE Trans Patt Anal Mach Intell 34(3):436–450CrossRefGoogle Scholar
  34. 34.
    Zhou W, Zheng JR, Yan JJ, Wang JF (2011) A novel hybrid algorithm for assembly sequence planning combining bacterial chemotaxis with genetic algorithm. Int J Adv Manuf Technol 52(5):715–724CrossRefGoogle Scholar

Copyright information

© Springer Science+Business Media New York 2015

Authors and Affiliations

  1. 1.School of Information Science and EngineeringCentral South UniversityChangshaChina
  2. 2.Department of Information ManagementHunan University of Finance and EconomicsChangshaChina

Personalised recommendations