Performance Evaluation of Shadow Features as a Data Preprocessing Method in Data Mining for Human Activities Recognitions

  • Simon FongEmail author
  • Shimin Hu
  • Ni Ren
  • Wei Song
  • Kyungeun Cho
  • Raymond Wong
  • Sabah Mohammed
Part of the International Series on Computer Entertainment and Media Technology book series (ISCEMT)


A Human Activity Recognition (HAR) classification model is used to predict the class or predefined type of human activity. With the limited amount of available original features of human activity, the classification performance is usually mediocre. One solution is to enrich the information of the original data attributes. The objective of this study is to find a suitable feature transformation method for inducing an accurate classifier for HAR. A novel concept for enriching the feature information of HAR is called Shadow Feature. Two versions of Shadow Features are implemented here. They are being tested via RapidMiner to see which version is more suitable for HAR. The experiment results show that the four data pre-processing strategies could be ranked by their performance as follow: shadow feature 2 > shadow feature 1 > statistical features > original features. Algorithm-wise, ensemble algorithms are able to improve the HAR classification performance while a single decision tree is shown to be a weak classifier. Finally, it is observed that good performance can be achieved when shadow features are applied over datasets of drastic activity; in this case shadow feature 2 is better than shadow feature 1. For datasets of subtle activity shadow features do have advantages too, though slightly; in this case shadow feature 1 works better than shadow feature 2.


Human activity recognition Classification model Feature transformation Shadow features Statistical features 



The authors are thankful for the financial support from the Research Grants (1) title: “Temporal Data Stream Mining by Using Incrementally Optimized Very Fast Decision Forest (iOVFDF)”, Grant no. MYRG2015-00128-FST, offered by the University of Macau, and Macau SAR government. (2) title: “A Scalable Data Stream Mining Methodology: Stream-based Holistic Analytics and Reasoning in Parallel”, Grant no. FDCT/126/2014/A3, offered by FDCT of Macau SAR government.


  1. 1.
    Bahar Farahani, Farshad Firouzi, Victor Chang, Mustafa Badaroglu, Nicholas Constant, Kunal Mankodiya, Towards fog-driven IoT eHealth: Promises and challenges of IoT in medicine and healthcare, Future Generation Computer Systems, Volume 78, Part 2, January 2018, pp.659–676 CrossRefGoogle Scholar
  2. 2.
    Abdelsalam Helal, Kyungeun Cho, Wonsik Lee, Yunsick Sung, JW Lee, Eunju Kim, 3D modeling and simulation of human activities in smart spaces, 2012 9th International Conference on Ubiquitous Intelligence & Computing and 9th International Conference on Autonomic & Trusted Computing (UIC/ATC), IEEE, September 2012, pp.112–119Google Scholar
  3. 3.
    Hang Yang, Simon Fong, Kyungeun Cho, Junbo Wang, Atmospheric pattern recognition of human activities on ubiquitous sensor network using data stream mining algorithms, International Journal of Sensor Networks, Volume 20 Issue 3, March 2016, pp.147–162Google Scholar
  4. 4.
    Simon Fong, Kexing Liu, Kyungeun Cho, Raymond Wong, Sabah Mohammed, Jinan Fiaidhi, Improvised methods for tackling big data stream mining challenges: case study of human activity recognition, Journal of Supercomputing, Volume 72 Issue 10, October 2016, pp.3927–3959Google Scholar
  5. 5.
    Braun, Marta, 1992, Picturing Time: The Work of Etienne-Jules Marey, University of Chicago Press, ChicagoGoogle Scholar
  6. 6.
    Laurent Mannoni, La grande arte della luce e dell'ombra. Archeologia del cinema, Lindau 1994–2007Google Scholar
  7. 7.
  8. 8.
  9. 9.
    Williams, L. Casting curved shadows on curved surfaces. Computer graphics and interactive techniques. Pages 270–274CrossRefGoogle Scholar
  10. 10.
    S.W. Leung, J.W. Minett , C.F. Chung. An analysis of the shadow feature technique in radar detection. Jul 1999. IEEE. pp. 1104–1106CrossRefGoogle Scholar
  11. 11.
    J. W. Minett, S. W. Leung, Y. M. Siu, K. T. Ng, and W. N. Chau. Estimating the Length of a Radar Shadow in Shadow-Feature-Enhanced Detection Using a Fuzzy System. 2001. IEEEGoogle Scholar
  12. 12.
    Ishida. S, Fukui. S, Iwahori. Y, Bhuyan. M. K, Woodham. R. J. Shadow Detection by Three Shadow Models with Features Robust to Illumination Changes. Procedia Computer Science. Volume 35, 2014, Pages 1219–1228CrossRefGoogle Scholar
  13. 13.
    Liu. H, Li. J. T, Liu. Q, Yian. Y. L, Li. H. J. Moving Cast Shadow Elimination Based on Color and Gradient Features. Journal of Computer-Aided Design & Computer Graphics. Vol. 19, NO.10 Oct., 2007Google Scholar
  14. 14.
    Xie. W. H, Yi. B. S, Xiao. J. S, Gan. L. C. Shadow detection algorithm based on color and Regional gradient direction features. Journal of Central South University (Science and Technology). Vol.44 No.12. Dec. 2013Google Scholar
  15. 15.
    Fang. J. Q, Chen. F, He. H. J, Yin. Z. K. Shadow Detection of Remote Sensing Images Based on Local-classification Level Set and Color Feature. ACTA AUTOMATICA SINICA. Vol. 40, No. 6. June, 2014Google Scholar
  16. 16.
    Ding. A, Yang. K, Qi. H. C. Xiao. F. Vehicle Shadow Removal Based on the Characteristics of the Single side shadow. CAAI Transactions on Intelligent Systems. Vol. 10. No. 2. Apr. 2015Google Scholar
  17. 17.
    Li. Y. C, He. K. Z, Jia. P. F. Forward vehicle detection based on shadow features and Adaboost J Tsinghua Univ CSci &. Tech). 2007. Vol. 47. No. 10. pp. 1713–1716Google Scholar
  18. 18.
    Lin. K. J, Wan. X. D. Shadow Detection Algorithm Based on Edge Information and Illumination Direction. Computer Engineering. Vol.35 No.20. October 2009Google Scholar
  19. 19.
    H.E. Hurst, Long-term storage of reservoirs: an experimental study, Transactions of the American Society of Civil Engineers, Vol. 116, 1951, pp.770–799Google Scholar
  20. 20.
    Lijffijt, J, Papapetrou, P, Puolamäki, K. Size Matters: Finding the Most Informative Set of Window Lengths. Lecture Notes in Computer Science. LNCS, volume 7524Google Scholar
  21. 21.
    Chan, J.H, Visutarrom, T, Cho, S.B, Engchuan, W, Mongolnam, P, Fong, S. A Hybrid Approach to Human Posture Classification during TV Watching. 6(4), 1119–1126CrossRefGoogle Scholar
  22. 22.
    Fothergill, S, Mentis, H. M, Kohli, P, Nowozin, S. Instructing People for Training Gestural Interactive Systems. ACM, Conference on Human Factors in Computing Systems. 2012, pp.1737–1746Google Scholar
  23. 23.
    Wei, S.X, Wang, W. Y. Improvement and Implementation of SVM and Integrated Learning Algorithm. Computer System & Applications. 2015, Volume 24, No. 7Google Scholar
  24. 24.
    Wang, X.D, Gao, X.F, Yao, X, Lei, L. Research and Application of SVM Ensemble. Journal of Air Force Engineering University. Apr 2012, Vol. 13, No.2Google Scholar
  25. 25.
    Eriksson, D, Glansberg, S, Johan, J. Cost-sensitive Classifiers. Dec 4, 2009Google Scholar
  26. 26.
    Dietterich, T. An experimental comparison of three methods for constructing ensembles of decision trees: Bagging, boosting, and randomization. Machine Learning 40, 2 (2000), 139–158CrossRefGoogle Scholar

Copyright information

© Springer International Publishing AG, part of Springer Nature 2018

Authors and Affiliations

  • Simon Fong
    • 1
    Email author
  • Shimin Hu
    • 1
  • Ni Ren
    • 1
  • Wei Song
    • 2
  • Kyungeun Cho
    • 3
  • Raymond Wong
    • 4
  • Sabah Mohammed
    • 5
  1. 1.Department of Computer and Information ScienceUniversity of MacauTaipaChina
  2. 2.School of Computer Science and TechnologyNorth China University of TechnologyBeijingChina
  3. 3.Department of Multimedia EngineeringDongguk UniversitySeoulRepublic of Korea
  4. 4.School of Computer Science and EngineeringUniversity of New South WalesSydneyAustralia
  5. 5.Department of Computer ScienceLakehead UniversityThunder BayCanada

Personalised recommendations