Abstract
Foreground detection plays an important role in many content based video processing applications. To detect moving objects in a scene, the changes inherent to the background need to be modelled. In this work we propose a non-parametric statistical background modeling technique. Moreover, the proposed modeling framework is designed to utilize Nvidia’s CUDA architecture to accelerate the overall foreground detection process. We present three main contributions: (1) a novelty detection mechanism capable of building accurate statistical models for background pixels; (2) an adaptive mechanism for classifying pixels based on their respective statistical background model; and (3) the complete implementation of the proposed approach based on the Nvidia’s CUDA architecture. Comparisons and both qualitative and quantitative experimental results show that the proposed work achieves considerable accuracy in detecting foreground objects, while reaching orders of magnitude speed-up compared to traditional approaches.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
References
Benezeth, Y., Jodoin, P.M., Emile, B., Laurent, H., Rosenberger, C.: Comparative study of background subtraction algorithms. J. Electron. Imaging 19, 033003–033003 (2010)
Wang, B., Dudek, P.: A fast self-tuning background subtraction algorithm. In: 2014 IEEE Conference on Computer Vision and Pattern Recognition Workshops (CVPRW), pp. 401–404. IEEE (2014)
Zivkovic, Z., van der Heijden, F.: Efficient adaptive density estimation per image pixel for the task of background subtraction. Pattern Recogn. Lett. 27, 773–780 (2006)
Stauffer, C., Grimson, W.E.L.: Adaptive background mixture models for real-timetracking. In: IEEE ComputerSociety Conference on Computer Vision and Pattern Recognition, 1999, vol. 2. IEEE (1999)
Zivkovic, Z.: Improved adaptive gaussian mixture model for background subtrac-tion. In: Proceedings of the 17th International Conference on Pattern Recognition, 2004. ICPR 2004, vol. 2, pp. 28–31. IEEE (2004)
Varadarajan, S., Miller, P., Zhou, H.: Spatial mixture of gaussians for dynamic background modelling. In: 2013 10th IEEE International Conference on Advanced Video and Signal Based Surveillance (AVSS), pp. 63–68. IEEE (2013)
Maddalena, L., Petrosino, A.: A fuzzy spatial coherence-based approach to background/foreground separation for moving object detection. Neural Comput. Appl. 19, 179–186 (2010)
Maddalena, L., Petrosino, A.: The sobs algorithm: what are the limits? In: 2012 IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops (CVPRW), pp. 21–26. IEEE (2012)
Gregorio, M.D., Giordano, M.: Change detection with weightless neural networks. In: 2014 IEEE Conference on Computer Vision and Pattern Recognition Workshops (CVPRW), pp. 409–413. IEEE (2014)
Liang, D., Kaneko, S.: Improvements and experiments of a compact statistical background model. arXiv preprint arXiv:1405.6275 (2014)
Elgammal, A., Harwood, D., Davis, L.: Non-parametric model for background subtraction. In: Vernon, D. (ed.) ECCV 2000. LNCS, vol. 1843, pp. 751–767. Springer, Heidelberg (2000)
Sedky, M., Moniri, M., Chibelushi, C.C.: Spectral-360: a physics-based technique for change detection. In: 2014 IEEE Conference on Computer Vision and Pattern Recognition Workshops (CVPRW), pp. 405–408 (2014)
Tavakkoli, A., Nicolescu, M., Bebis, G.: Non-parametric statistical background modeling for efficient foreground region detection. Int. J. Mach. Vis. Appl. 20, 395–409 (2008)
Wang, R., Bunyak, F., Seetharaman, G., Palaniappan, K.: Static and moving object detection using flux tensor with split gaussian models. In: 2014 IEEE Conference on Computer Vision and Pattern Recognition Workshops (CVPRW), pp. 420–424. (2014)
Goyette, N., Jodoin, P.M., Porikli, F., Konrad, J., Ishwar, P.: changedetection.net: a new change detection benchmark dataset (2012)
Lu, X.: A multiscale spatio-temporal background model for motion detection. In: 2014 IEEE International Conference on Image Processing (ICIP), pp. 3268–3271. IEEE (2014)
Acknowledgements
This material is based upon work supported in part by the U. S. Army Research Laboratory and the U. S. Department of Defense under grant number W911NF-15-1-0024 and W911NF-15-1-0455, and by the NASA MUREP ASTAR program under grant number NNX15AU31H. This support does not necessarily imply endorsement by the DoD or NASA.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2015 Springer International Publishing Switzerland
About this paper
Cite this paper
Wilson, B., Tavakkoli, A. (2015). An Efficient Non-parametric Background Modeling Technique with CUDA Heterogeneous Parallel Architecture. In: Bebis, G., et al. Advances in Visual Computing. ISVC 2015. Lecture Notes in Computer Science(), vol 9474. Springer, Cham. https://doi.org/10.1007/978-3-319-27857-5_19
Download citation
DOI: https://doi.org/10.1007/978-3-319-27857-5_19
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-27856-8
Online ISBN: 978-3-319-27857-5
eBook Packages: Computer ScienceComputer Science (R0)