The Journal of Supercomputing

, Volume 75, Issue 3, pp 1670–1685 | Cite as

Optimization of lateral interaction in accumulative computation on GPU-based platform

  • Aurelio Bermúdez
  • Francisco Montero
  • María T. López
  • Antonio Fernández-Caballero
  • José L. SánchezEmail author


The lateral interaction in accumulative computation (LIAC) algorithm is a biologically inspired method that allows us to detect moving objects from image sequences acquired from fixed surveillance cameras. This method achieves excellent precision but requires a high processing time. Sequential implementation is too slow and cannot achieve real-time processing. In this paper, we present several improvements to the LIAC algorithm that increase its efficiency in terms of execution time and energy consumption. In particular, a GPU-based implementation delivers the same precision and is notably faster and more energy efficient than the sequential implementation.


HPC Motion detection Acceleration Optimization GPU Lateral interaction in accumulative computation 



This work was partially supported by Spanish Ministerio de Ciencia, Innovación y Universidades, Agencia Estatal de Investigación (AEI) / European Regional Development Fund (FEDER, UE) under DPI2016-80894-R and TIN2015-66972-C5-2-R grants.


  1. 1.
    Aydin S, Samet R, Bay OF (2018) Real-time parallel image processing applications on multicore CPUs with OpenMP and GPGPU with CUDA. J Supercomput 74(6):2255–2275. CrossRefGoogle Scholar
  2. 2.
    Bako L, Hajdu S, Brassai ST, Morgan F, Enachescu C (2016) Embedded implementation of a real-time motion estimation method in video sequences. Proc Technol 22:897–904. CrossRefGoogle Scholar
  3. 3.
    Bermúdez A, López MT, Montero F, Fernández-Caballero A, Sánchez JL (2018) Accelerating bioinspired moving object detection with FPGAs and GPUs. In: 18th International Conference on Computational and Mathematical Methods in Science and Engineering (CMMSE 2018)Google Scholar
  4. 4.
    Campmany V, Silva S, Espinosa A, Moure J, Vázquez D, López A (2016) GPU-based pedestrian detection for autonomous driving. Proc Comput Sci 80:2377–2381. CrossRefGoogle Scholar
  5. 5.
    Corporation N (2018) CUDA: Compute unified device architecture.
  6. 6.
    Corporation N (2018) Wattsup?.net meter.
  7. 7.
    Cox DD, Dean T (2014) Neural networks and neuroscience-inspired computer vision. Curr Biol 24(18):R921–R929. CrossRefGoogle Scholar
  8. 8.
    Delgado AE, López MT, Fernández-Caballero A (2010) Real-time motion detection by lateral inhibition in accumulative computation. Eng Appl Artif Intell 23(1):129–139. CrossRefGoogle Scholar
  9. 9.
    Fernández-Caballero A, López M, Castillo J, Maldonado-Bascón S (2009) Real-time accumulative computation motion detectors. Sensors 9(12):10044–10065. CrossRefGoogle Scholar
  10. 10.
    Fernández-Caballero A, López MT, Carmona EJ, Delgado AE (2011) A historical perspective of algorithmic lateral inhibition and accumulative computation in computer vision. Neurocomputing 74(8):1175–1181. CrossRefGoogle Scholar
  11. 11.
    García-Rodríguez J, Orts-Escolano S, Angelopoulou A, Psarrou A, Azorín-López J, García-Chamizo JM (2016) Real time motion estimation using a neural architecture implemented on GPUs. J Real-Time Image Process 11(4):731–749. CrossRefGoogle Scholar
  12. 12.
    Gascueña JM, Serrano-Cuerda J, Castillo JC, Fernández-Caballero A, López MT (2014) A multi-agent system for infrared and color video fusion. In: Bajo Perez J, Corchado JM, Mathieu P, Campbell A, Ortega A, Adam E, Navarro EM, Ahrndt S, Moreno MN, Julián V (eds) Trends in Practical Applications of Heterogeneous Multi-Agent Systems. The PAAMS Collection. Springer, Cham, pp 131–138Google Scholar
  13. 13.
    Goyette N, Jodoin PM, Porikli F, Konrad J, Ishwar P (2012) a new change detection benchmark dataset. In: 2012 IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops, pp 1–8.
  14. 14.
    Guler P, Emeksiz D, Temizel A, Teke M, Temizel TT (2016) Real-time multi-camera video analytics system on GPU. J Real-Time Image Process 11(3):457–472. CrossRefGoogle Scholar
  15. 15.
    Kirk DB, Hwu WW (2012) Programming massively parallel processors, 2nd edn. Morgan Kaufmann Publishers Inc., San FranciscoGoogle Scholar
  16. 16.
    Kriegeskorte N (2015) Deep neural networks: a new framework for modeling biological vision and brain information processing. Annu Rev Vis Sci 1(1):417–446. CrossRefGoogle Scholar
  17. 17.
    López MT, Bermúdez A, Montero F, Sánchez JL, Fernández-Caballero A (2018) A finite state machine approach to algorithmic lateral inhibition for real-time motion detection. Sensors 18(5):1420. CrossRefGoogle Scholar
  18. 18.
    Lyu C, Chen H, Jiang X, Li P, Liu Y (2017) Real-time object tracking system based on field-programmable gate array and convolution neural network. Int J Adv Robot Syst 14(1):1729881416682705. CrossRefGoogle Scholar
  19. 19.
    Medathati NVK, Neumann H, Masson GS, Kornprobst P (2016) Bio-inspired computer vision: towards a synergistic approach of artificial and biological vision. Comput Vis Image Underst 150:1–30. CrossRefGoogle Scholar
  20. 20.
    Mira J, Delgado AE (2001) What can we compute with lateral inhibition circuits? In: Mira J, Prieto A (eds) Connectionist models of neurons, learning processes, and artificial intelligence. Springer, Heidelberg, pp 38–46CrossRefGoogle Scholar
  21. 21.
    Mori JY, Arias-Garcia J, Sánchez-Ferreira C, Muñoz DM, Llanos CH, Motta JM (2012) An FPGA-based omnidirectional vision sensor for motion detection on mobile robots. Int J Reconfigurable Comput 12:1–16. CrossRefGoogle Scholar
  22. 22.
    Sánchez JL, Viana R, López MT, Fernández-Caballero A (2017) Acceleration of moving object detection in bio-inspired computer vision. In: Fernández Vicente JM, Álvarez-Sánchez JR, de la Paz López F, Toledo Moreo J, Adeli H (eds) Biomedical applications based on natural and artificial computing. Springer, Cham, pp 364–373CrossRefGoogle Scholar
  23. 23.
    Sehairi K, Benbouchama C, Chouireb F (2015) A real time implementation on FPGA of moving objects detection and classification. Int J Circuits Syst Signal Process 9:160–167Google Scholar
  24. 24.
    Shehab M, Al-Ayyoub M, Jararweh Y (2017) Accelerating compute-intensive image segmentation algorithms using GPUs. J Supercomput 73(5):1929–1951. CrossRefGoogle Scholar
  25. 25.
    Singh S, Shekhar C, Vohra A (2017) Real-time fpga-based object tracker with automatic pan-tilt features for smart video surveillance systems. J Imaging 3(2):1–28. Google Scholar
  26. 26.
    Tang JW, Shaikh-Husin N, Sheikh UU, Marsono MN (2016) Fpga-based real-time moving target detection system for unmanned aerial vehicle application. Int J Reconfigurable Comput 2016:8457908. CrossRefGoogle Scholar
  27. 27.
    Ullman S, Assif L, Fetaya E, Harari D (2016) Atoms of recognition in human and computer vision. Proc Natl Acad Sci 113(10):2744–2749. CrossRefGoogle Scholar
  28. 28.
    Wang Y, Jodoin PM, Porikli F, Konrad J, Benezeth Y, Ishwar P (2014) CDnet 2014: an expanded change detection benchmark dataset. In: 2014 IEEE Conference on Computer Vision and Pattern Recognition Workshops, pp 393–400.

Copyright information

© Springer Science+Business Media, LLC, part of Springer Nature 2019

Authors and Affiliations

  1. 1.Departamento de Sistemas InformáticosUniversidad de Castilla-La ManchaAlbaceteSpain

Personalised recommendations