Skip to main content
SpringerLink
Log in

Efficient adaptive load balancing approach for compressive background subtraction algorithm on heterogeneous CPU–GPU platforms

  • Original Research Paper
  • Published:
Journal of Real-Time Image Processing Aims and scope Submit manuscript

Abstract

Mixture of Gaussians (MoG) and compressive sensing (CS) are two common approaches in many image and audio processing systems. The combination of these algorithms is recently used for the compressive background subtraction task. Nevertheless, the result of this combination has not been exploited to take advantage of the evolution of parallel computing architectures. This paper proposes an efficient strategy to implement CS-MoG on heterogeneous CPU–GPU computing platforms. This is achieved through two elements. The first one is ensuring the better acceleration and accuracy that can be achieved for this algorithm on both CPU and GPU processors: The obtained results of the improved CS-MoG are more accurate and performant than other published MoG implementations. The second contribution is the proposition of the Optimal Data Distribution Cursor ODDC, a novel adaptive data partitioning approach to exploit simultaneously the heterogeneous processors on any given platform. It aims to ensure an automatic workload balancing by estimating the optimal data chunk size that must be assigned to each processor, with taking into consideration its computing capacity. Furthermore, our method ensures an update of the partitioning at runtime to take into account any influence of data content irregularity. The experimental results, on different platforms and data sets, show that the combination of these two contributions allows reaching 98% of the maximal possible performance of targeted platforms.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Fig. 7
Fig. 8
Fig. 9
Fig. 10
Fig. 11

Similar content being viewed by others

References

  1. Akhter, S., Roberts, J.: Multi-core Programming: Increasing Performance Through Software Multithreading. Intel Press, California (2016)

    Google Scholar 

  2. Augonnet, C., Thibault, S., Namyst, R., Wacrenier, P.A.: StarPU: a unified platform for task scheduling on heterogeneous multicore architectures. Eur. Conf. Parall. Process. 23, 187–198 (2009)

    Google Scholar 

  3. Barnich, O., Droogenbroeck, M.V.: ViBe: a universal background subtraction algorithm for video sequences. IEEE Trans. Image Process. 20, 1709–1724 (2011)

    Article  MathSciNet  Google Scholar 

  4. Chapman, B., Jost, G., van der Paas, R.: Using OpenMP. Portable Shared Memory Parallel Programming. MIT Press, Cambridge (2007)

    Google Scholar 

  5. Davenport, M.: The Fundamentals of compressive sensing. SigView. (2013)

  6. Friedman, N., Russell, S.: Image segmentation in video sequences: a probabilistic approach. In: Proceedings of the thirteenth conference on uncertainty in artificial intelligence, pp. 175–181. (1997)

  7. Grewe, D., Wang, Z., O’Boyle, M.F.P.: OpenCL task partitioning in the presence of GPU contention. In: International workshop on languages and compilers for parallel computing. Springer, pp 87–101. (2013)

  8. Guler, P., Emeksiz, D., Temizel, A., Teke, M., Temizel, T.T.: Real-time multi-camera video analytics system on GPU. J. Real-Time Image Process. 11, 457–472 (2016)

    Article  Google Scholar 

  9. KaewTraKulPong, P., Bowden, R.: An improved adaptive background mixture model for real-time tracking with shadow detection. In: Proceedings of 2nd european workshop on advanced video based surveillance systems. (2001)

  10. Kovacev, P., Misic, M., Tomasevic, M.: Parallelization of the mixture of gaussians model for motion detection on the GPU. In: Zooming innovation in consumer technologies. (2018)

  11. Kulkarni, A., Mohsenin, T.: Accelerating compressive sensing reconstruction OMP algorithm with CPU, GPU, FPGA and domain specific many-core. In: International symposium on circuits and systems. (2015)

  12. Kumar, P., Singhal, A., Mittal, M.: Real-time moving object detection algorithm on high-resolution videos using GPUs. J. Real-Time Image Process. 11, 93–109 (2016)

    Article  Google Scholar 

  13. Kyungnain, K., Chalidabhongse, TH., Hanuood, D., Davis, L.: Background modeling and subtraction by codebook construction. In: International conference on image processing ICIP. (2004)

  14. LI, L., GEDA, R., HAYES, A.B., CHEN, Y., CHAUDHARI, P., ZHANG, E.Z. SZEGEDY, M.: A simple yet effective balanced edge partition model for parallel computing. In: Proceedings of the 2017 ACM SIGMETRICS/international conference on measurement and modeling of computer systems, pp. 6-6 Urbana-Champaign, Illinois, USA. (2017)

  15. Lee, J., Park, M.: An adaptive background subtraction method based on kernel density estimation. Sensors 12, 12279–12300 (2012)

    Article  Google Scholar 

  16. Li, H.F., Liang, T.Y., Lin, Y.J.: An OpenMP programming toolkit for hybrid CPU/GPU clusters based on software unified memory. J. Inf. Sci. Eng. 32, 517–539 (2016)

    Google Scholar 

  17. Liang, M., Li, Y., Neifeld, MA., Xin, H.: Principal component analysis (PCA) based compressive sensing millimeter wave imaging system. In: USNC-URSI Radio Science Meeting (Joint with AP-S Symposium) (2015)

  18. Liu, B., Qiu, W., Jiang, L., Gong, Z.: Software pipelining for graphic processing unit acceleration: Partition, scheduling and granularity. Int. J. High Perform. Comput. Appl. 2015, 1–17 (2015)

    Google Scholar 

  19. Luk, C.K., Hong, S., Kim, H.: Qilin: exploiting parallelism on heterogeneous multiprocessors with adaptive mapping. In: Proceedings of the 42nd Annual IEEE/ACM international symposium on microarchitecture, pp. 45–55. (2009)

  20. Mabrouk, L., Sylvain, Huet S., Houzet, D., Belkouch, S., Hamzaoui, A., and Zennayi, Y.: Performance and scalability improvement of GMM background segmentation algorithm on multi-core parallel platforms. In: International conference on electronic engineering and renewable energy ICEERE, Saidia, Morocco. (2018)

  21. Mabrouk, L., Sylvain, Huet S., Houzet, D., Belkouch, S., Hamzaoui, A., and Zennayi, Y.: Single core SIMD parallelization of GMM background subtraction algorithm for vehicles detection. In: IEEE international colloquium on information science and technology (CiSt). (2018)

  22. Mabrouk, L., Sylvain, Huet S., Houzet, D., Belkouch, S., Hamzaoui, A., Zennayi, Y.: Efficient parallelization of GMM background subtraction algorithm on a multi-core platform for moving objects detection. In: International conference on advanced technologies for signal and image processing. (2018)

  23. Navarro, A., Corbera, F., Rodriguez, A., Vilches, A., Asenjo, R.: Heterogeneous parallel for template for CPU-GPU Chips. Int. J. Parall. Program. 47, 213–233 (2018)

    Article  Google Scholar 

  24. Nurhadiyatna, A., Wijayanti, R., Fryantoni, D.: Extended Gaussian mixture model enhanced by hole filling algorithm (GMMHF) utilize GPU acceleration. Inf. Sci. Appl. 2016, 459–469 (2016)

    Google Scholar 

  25. Nvidia: NVIDIA CUDA Compute unified device architecture. Version 2.0. (2008)

  26. Pham, V., Vo, P., Hung, V.T., Bac, L.H.: GPU Implementation of Extended Gaussian Mixture Model for Background Subtraction. In: International conference on computing & communication technologies, research, innovation, and vision for the future (RIVF). (2010)

  27. Piccardi, M.: Background subtraction techniques: a review (PDF). IEEE Int. Conf. Syst. Man Cybern. 4, 3099–3104 (2004)

    Google Scholar 

  28. Ponomarenko, N., Lukin, V., Egiazarian, K., Astola, J.: DCT based high quality image compression. Tampere international center for signal processing, tampere university of technology (2005)

  29. Powers, D.M.W.: Evaluation: from precision, recall and F-measure to ROC, informedness, markedness and correlation (PDF). J. Mach. Learn. Technol. 2(1), 37–63 (2011)

    MathSciNet  Google Scholar 

  30. Raju, K., Niranjan, N.C.: A survey on techniques for cooperative CPU-GPU computing. Sustain. Comput. Inf. Syst. 19, 72–85 (2018)

    Google Scholar 

  31. Rennich, S.: Cuda C/C++ streams and concurrency. In: GPU Technology Conference. San Jose, California (2011). https://developer.download.nvidia.com/CUDA/training/StreamsAndConcurrencyWebinar.pdf

  32. Sanders, J., Kandrot, E.: CUDA by example: an introduction to general-purpose GPU programming, Portable Documents. Addison-Wesley Professional (2010)

  33. SBMnet dataset: http://jacarini.dinf.usherbrooke.ca/dataset2014/. (2018)

  34. Shen, Y., Hu, W., Yang, M., Liu, J., Wei, B., Lucey, S., Chou, C.T.: Real-time and robust compressive background subtraction for embedded camera networks. IEEE Trans. Mobile Comput. 15, 406–418 (2016)

    Article  Google Scholar 

  35. Singh, A.K., Prakash, A., Basireddy, K.R., Merrett, G.V., Al-Hashimi, B.M.: Energy-efficient run-time mapping and thread partitioning of concurrent OpenCL applications on CPU-GPU MPSoCs. In: ACM transactions on embedded computing systems (TECS)—Special Issue ESWEEK (2017)

  36. Stafford, E., Perez, B., Bosque, J.L., Beivide, R., Valero, M.: To distribute or not to distribute: the question of load balancing for performance or energy. Eur. Conf. Parall. Process. 2017, 710–722 (2017)

    Google Scholar 

  37. Stauffer, C. and Grimson, WEL.: Adaptive background mixture models for real-time tracking. In: Proceedings of IEEE conference on computer vision and pattern recognition, pp. 246–252. USA (1999)

  38. Stone, J.E., Gohara, D., Shi, G.: OpenCL: a parallel programming standard for heterogeneous computing systems. Comput. Sci. Eng. 12, 66 (2010)

    Article  Google Scholar 

  39. Szwoch, G., Ellwart, D., Czyżewski, A.: Parallel implementation of background subtraction algorithms for real-time video processing on a supercomputer platform. J. Real-Time Image Proc 11, 111–125 (2012)

    Article  Google Scholar 

  40. Tamersoy, B.: Background subtraction—lecture notes (PDF). University of Texas at Austin. (2009)

  41. Vilches, A., Asenjo, R., Navarro, A., Corbera, F., Gran, R., Garzaran, M.: Adaptive partitioning for irregular applications on heterogeneous CPU-GPU chips. Int. Conf. Comput. Sci. 51, 140–149 (2015)

    Google Scholar 

  42. Wen, Y., Wang, Z., O’Boyle, M.F.P.: Smart multi-task scheduling for OpenCL programs on CPU/GPU heterogeneous platforms. In High Performance Computing (HiPC). In: 2014 21st International Conference on. IEEE, pp. 1–10. (2014)

  43. Yang, J., Yuan, X., Liao, X., Llull, P., Brady, D.J., Sapiro, G., Carin, L.: Video Compressive sensing using Gaussian mixture models. IEEE Trans. Image Process. 23, 4863–4878 (2014)

    Article  MathSciNet  Google Scholar 

  44. Zhang, F., Zhai, J., He, B., Zhang, S., Chen, W.: Understanding co-running behaviours on integrated CPU-GPU architectures. IEEE Trans. Parall. Distribut. Syst. 28, 905–918 (2016)

    Article  Google Scholar 

  45. Zivkovic, Z., Van Der Heijden, F.: Efficient adaptive density estimation per image pixel for the task of background subtraction. Pattern Recogn. Lett. 27(7), 773–780 (2006)

    Article  Google Scholar 

Download references

Author information

Authors and Affiliations

  1. CNRS, Grenoble-INP, GIPSA-Lab, Univ. Grenoble Alpes, 38000, Grenoble, France

    Lhoussein Mabrouk, Sylvain Huet & Dominique Houzet

  2. LGECOS Lab, ENSA Marrakesh, Cadi Ayyad University, Marrakesh, 40000, Morocco

    Said Belkouch & Abdelkrim Hamzaoui

  3. Smart Communications Research Team, Research Centre E3S, Department of Electrical Engineering at Mohammadia School of Engineering, Mohamed V University, Rabat, 10000, Morocco

    Yahya Zennayi

Authors
  1. Lhoussein Mabrouk
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Sylvain Huet
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Dominique Houzet
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Said Belkouch
    View author publications

    You can also search for this author in PubMed Google Scholar

  5. Abdelkrim Hamzaoui
    View author publications

    You can also search for this author in PubMed Google Scholar

  6. Yahya Zennayi
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Lhoussein Mabrouk.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

This work has been partially supported by the LabEx PERSYVAL-Lab (ANR-11-LABX-0025-01), and partially supported by the National Centre for Scientific and Technical Research in Morocco (CNRST), through the MOVITS project.

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Mabrouk, L., Huet, S., Houzet, D. et al. Efficient adaptive load balancing approach for compressive background subtraction algorithm on heterogeneous CPU–GPU platforms. J Real-Time Image Proc 17, 1567–1583 (2020). https://doi.org/10.1007/s11554-019-00916-4

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s11554-019-00916-4

Keywords

Search

Navigation