Real Time Object Oriented 6-Point Skeleton Extraction Component from Human Silhouette for Video Surveillance and Analysis Application
This paper describes the design and implementation of a fast 6-Point skeleton extraction algorithm from human silhouette images which can be used for real-time video surveillance and analysis applications. The 6 Points of Interest (POIs) are sacroiliac support point (P c ), head (P h ), right and left shoulders (P sr and P sl ) and right and left foot (P fr and P fl ). The algorithm was implemented as an object class library and can be used in Microsoft.NET development environment across multiple.NET compatible programming language such as C#, VB.Net and IronRuby. The developed class library successfully tracked the POIs across live video frames in real-time. The applicability of the developed class library for video surveillance and analysis application was proven via its application in the development of the Intelligent Video Surveillance System (InViSSTM). A simple, general purpose case study, which shows the use of the developed.NET components for simple gait analysis application is also discussed in the end of this paper to prove that it is general enough to be used for video surveillance and analysis related applications.
Keywords6-Point skeleton Human silhouette analysis Object oriented programming
The authors would like to express their gratitude to the Government of Malaysia and Universiti Kebangsaan Malaysia for financing this research via the GUP-2013-035 Research Grant.
- 1.Siyuan, G.: An intelligent video surveillance system. In: 2010 International Conference on E-Product E-Service and E-Entertainment (ICEEE), pp. 1–4 (2010)Google Scholar
- 3.Wann-Yun, S., Ju-Chin, H.: Speedup the multi-camera video-surveillance system for elder falling detection. In International Conference Embedded Software and Systems, ICESS 2009, pp. 350–355 (2009)Google Scholar
- 4.Saad, M.H.M., Hussain, A.: Design and development of intelligent anomalous behaviour and event detection system. In: 2010 6th International Colloquium on Signal Processing and Its Applications, CSPA 2010, Melaka, Category number CFP1079G-ART, 21–23 May 2010Google Scholar
- 5.Saad, M.H.M., Hussain, A., Loong, L.X., Baharuddin, W.N.A., Tahir, M.: Event description from video stream for anomalous human activity and behaviour detection. In: Proceedings - 2011 IEEE 7th International Colloquium on Signal Processing and Its Applications, pp. 503–506 (2011)Google Scholar
- 6.Schwartz, W.R., Kembhavi, A., Harwood, D., Davis, L.S.: Human detection using partial least squares analysis. In: 2009 IEEE 12th International Conference on Computer Vision, pp. 24–31, 29 September–22 October 2009Google Scholar
- 7.Chen, H.S., Chen, H.T., Chen, Y.W., Lee, S.Y.: Human action recognition using star skeleton. In: VSSN 2006, pp. 171–178, October 2006Google Scholar
- 8.Guo, Y., Xu, G., Saburo, T.: Understanding human motion pattern. In: Proceeding of the 12th IAPR International Conference, vol. 2, pp. 325–329, 9–13 October 1994Google Scholar
- 9.Mo, H.C., Leou, J.J., Lin, C.S.: Human behaviour analysis using 2D features and multicategory support vector machine. In: MVA 2009, pp. 46–49, May 2009Google Scholar
- 10.Ding, J., Wang, Y., Yu, L.: Extraction of human body skeleton based on silhouette images. In: Second International Workshop on Education Technology and Computer Science (2010)Google Scholar