Skip to main content
Springer Nature Link
Log in

A real-time, power-efficient architecture for mean-shift image segmentation

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

Abstract

Image segmentation is essential to image processing because it provides a solution to the task of separating the objects in an image from the background and from each other, which is an important step in object recognition, tracking, and other high-level image-processing applications. By partitioning the input image into smaller regions, segmentation performs the balancing act of extracting the main areas of interest (objects and important features) that further help to interpret the image, while remaining immune to irrelevant noise and less important background scenes. Image-segmentation applications branch off into a plethora of domains, from decision-making applications in computer vision to medical imaging and quality control to name just a few. The mean-shift algorithm provides a unique unsupervised clustering solution to image segmentation, and it has an established record of good performance for a wide variety of input images. However, mean-shift segmentation exhibits an unfavorable computational complexity of \(O(kN^2)\), where \(N\) represents the number of pixels and \(k\) the number of iterations. As a result of this complexity, unsupervised image segmentation has had limited impact in autonomous applications, where a low-power, real-time solution is required. We propose a novel hardware architecture that exploits the customizable computing power of FPGAs and reduces the execution time by clustering pixels in parallel while meeting the low-power demands of embedded applications. The architecture performance is compared with existing CPU and GPU implementations to demonstrate its advantages in terms of both execution time and energy.

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

Access this article

Log in via an institution

Subscribe and save

Springer+ Basic
$34.99 /Month
  • Get 10 units per month
  • Download Article/Chapter or eBook
  • 1 Unit = 1 Article or 1 Chapter
  • Cancel anytime
Subscribe now

Buy Now

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
Fig. 12
Fig. 13
Fig. 14
Fig. 15
Fig. 16
Fig. 17
Fig. 18
Fig. 19
Fig. 20
Fig. 21
Fig. 22
Fig. 23
Fig. 24
Fig. 25

Similar content being viewed by others

References

  1. Fukunaga, K.L.D.H.: The estimation of the gradient of a density function, with application in pattern recognition. In: IEEE Trans. Information Theory (IT), vol. 21, 32–40 (1987)

  2. Parzen, E.: On estimation of a probability density function and mode. Ann. Math Stat. 21(1), 1065–1076 (1962)

    Article  MathSciNet  MATH  Google Scholar 

  3. Cheng, Y.: Mean shift, mode seeking, and clustering. IEEE Trans. Pattern Anal. Mach. Intell. 17(8), 790–799 (1995)

    Article  Google Scholar 

  4. Comaniciu, D.P.M.: Mean shift analysis and applications. Proc Seventh International Conference on Computer Vision. 1(1), 1197–1203 (1999)

  5. Comaniciu, D.P.M.: Mean shift: A robust approach toward feature space analysis, ieee trans. pattern analysis and machine intelligence. IEEE Trans. Pattern Anal. Mach .Intell. 24(5):603–619 (2002)

  6. Comaniciu, D.P.M., Ramesh, V.: Real-time tracking of non-rigid objects using mean shift. Proc 2000 IEEE Conference on Computer Vision and Pattern Recognition. 2(1):142–149 (2000)

  7. Bai, P., Fu, M.C.Y.H.C.: Improved mean shift segmentation scheme for medical ultrasound images. Fourth Intern. Conf. Bioinf. Biomed. Eng. (iCBBE). 1(1), 1–4 (2010)

    MathSciNet  Google Scholar 

  8. Bottger, J., Schafer, G.L.A.V.D.S.M.A.: Three-dimensional mean-shift edge bundling for the visualization of functional connectivity in the brain. IEEE Trans Vis. Comp. Gr. 20(3), 471–480 (2014)

    Article  Google Scholar 

  9. Yamashita, A., Ito, T.K.H.A.Y.: Human tracking with multiple cameras based on face detection and mean shift. IEEE Intern. Conf. Robot Biom. 1(1), 1664–1671 (2011)

    Google Scholar 

  10. Shotton, J., Blake, R.C.A.: Multiscale categorical object recognition using contour fragments. IEEE Trans. Pattern Anal. Mach. Intell. 30(7), 1270–1281 (2008)

    Article  Google Scholar 

  11. Deilamani, M.J.R.N.A.: Moving object tracking based on mean shift algorithm and features fusion. Artificial Intelligence and Signal Processing (AISP), 2011 International Symposium on. (1):48–53 (2011)

  12. Du-Ming, Tsai, J.Y.L., Yuan-Ze.: Mean shift-based defect detection in multicrystalline solar wafer surfaces. IEEE Trans. Indus. Inf. 7(1), 125–135 (2011)

    Article  Google Scholar 

  13. Ranchin, T.M.M., Wald, L.: The arsis method: a general solution for improving spatial resolution of images by the mmean of sensor fusion. Fusion of Earth Data: Merging Point Measurements, Raster Maps and Remotely Sensed Images (EARSeL) 1(1):53–58 (1996)

  14. Faro, A.S.P., Giordano, D.: Integrating unsipervised and supervised clustering methods on a gpu platform for fast image segmentation. 3rd International Conference on Image Processing Theory, Tools and Applications (IPTA) 85–90 (2012)

  15. Alexey, A., FWBD Tomas Kulvicius: Facing the Multicore-Challenge, Real-Time Image Segmentation on a GPU. Lecture Notes in Computer Science, vol. 6310, 1st edn. Springer, Berlin Heidelberg (2010)

  16. Fulkerson, B., SS: Really quick shift: Image segmentation on a gpu. In: Kutulakos, K.N. (ed.) Trends and Topics in Computer Vision. Lecture Notes in Computer Science, vol 6554, 350–358. Springer, Berlin Heidelberg (2012)

  17. Jun Zhang, X.L., Luo, S.: Weighted mean shift object tracking implemented on gpu for embedded sustems. Intern. Conf. Control Eng. Commun. Technol. 1(1), 982–985 (2012)

    Google Scholar 

  18. F Galluzzo, H.H.N.S., Barbosa D.: Segmentation framework for 3d echocardiography. IEEE International Ultrasonics Symposium (IUS) 2639–2642 (2012)

  19. Feng W.Y.Z, Xiang, H.: An improved graph-based image segmentation algorithm and its gpu acceleration. 2011 Workshop on Digital Media and Digital Content Management (DMDCM) 237–241 (2011)

  20. Rao, S., de Martins, A.M., Principe, J.C.: Mean shift: An information theoretic perspective. Trans. Pattern Anal. Mach. Intell. 30(3), 222–230 (2009)

  21. Renyi, A.: On measure of entropy and information. Proc Fourth Berkeley Symp Math Stat and Prob. 1(1), 547–561 (1961)

    MathSciNet  Google Scholar 

  22. Carreira-Perpinan, M.A.: Acceleration strategies for gaussian mean-shift image segmentation. IEEE Comp. Soc. Conf. Comp. Vision Pattern Recognit. 1(1), 1160–1167 (2006)

    Google Scholar 

  23. Saegusa, T.T.M.: An fpga implementation of k-means clustering for color images based on kd-tree. Intern. Conf. Field Program Logic Appl (FPL) 1(1), 1–6 (2006)

  24. Wang, H., HLJW JZhao.: Parallel clustering algorithms for image processing on multi-core cpus. Intern. Conf. Comp. Sci. Softw Eng. 2(1), 450–453 (2008)

  25. Ali, U., Malik, K.M.M.B.: Fpga/soft-processor based real-time object tracking system. Fifth South. Conf. Program. Logic (SPL) 1(1), 33–37 (2009)

    Google Scholar 

  26. Pandey, M.S.J.U.K.S.R., Borgohain, D.: Real-time histogram computation in kernel-based tracking system. International Conference on Advanced Electronic Systems (ICAES) 171–174 (2013)

  27. Trieu, D.B.K.T.M.: An implementation of the mean-shift filter on fpga. Intern. Conf. Field Program. Logic Appl. (FLP) 219–224 (2011)

  28. Lu, X.S.Y., Ren, D.: Fpga-based real-time object tracking for mobile robot. Intern. Conf. Audio Lang. Imag. Process. (ICALIP) 1657–1662 (2010)

  29. Stolkin, R., Florescu, M.B.C.H.I.: Efficient visual servoing with the abcshift tracking algorithm. IEEE Intern. Conf. Robot. Automation (ICRA). 1(1), 3219–3224 (2008)

    Google Scholar 

  30. Carreira-Perpinan, M.A.: Gaussian mean-shift is an em algorithm. IEEE Trans. Pattern Anal. Mach. Intell. 29(5), 767–776 (2007)

    Article  Google Scholar 

  31. Kirchgessner, R.A.G., Lam, H.: Reconfigurable computing middleware for application portability and productivity. Intern. Conf. Appl-Spec. Syst., Arch Process. (ASAP). 1(1):211–218 (2013)

  32. Kalgin, K.: Implementation of fine-grained algorithms on graphical processing units. 10th International Conference on Prallel Computing Technologies 207–215 (2009)

  33. Sirotkivic, J., Dujmic, V.P.H.: Accelerating mean-shift image segmsegmentation ifgt on massively parallel gpu. 36th International Conference on Information & Communications Technology Electronics and Microelectronics (MIPRO) 279–285 (2013)

  34. Altera, C.: Altera announces breakthrough advantages with generation 10; http://newsroom.altera.com/press-releases/nr-altera-generation-10.htm. Tech. rep., Altera (2013)

  35. Coombes, D.: Tegra k1 whitepaper. Tech. rep, NVIDIA (2014)

    Google Scholar 

Download references

Acknowledgments

This work was supported in part by the I/UCRC Program of the National Science Foundation under Grant Nos. EEC-0642422 and IIP-1161022.

Author information

Authors and Affiliations

  1. Department of Electrical and Computer Engineering, NSF Center for High-Performance Reconfigurable Computing (CHREC), University of Florida, Gainesville, FL, 32611-6200, USA

    Stefan Craciun, Robert Kirchgessner, Alan D. George & Herman Lam

  2. Computational Neuro-Engineering Laboratory (CNEL), Department of Electrical and Computer Engineering, University of Florida, Gainesville, FL, 32611-6200, USA

    Jose C. Principe

Authors
  1. Stefan Craciun
    View author publications

    You can also search for this author inPubMed Google Scholar

  2. Robert Kirchgessner
    View author publications

    You can also search for this author inPubMed Google Scholar

  3. Alan D. George
    View author publications

    You can also search for this author inPubMed Google Scholar

  4. Herman Lam
    View author publications

    You can also search for this author inPubMed Google Scholar

  5. Jose C. Principe
    View author publications

    You can also search for this author inPubMed Google Scholar

Corresponding author

Correspondence to Stefan Craciun.

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Craciun, S., Kirchgessner, R., George, A.D. et al. A real-time, power-efficient architecture for mean-shift image segmentation. J Real-Time Image Proc 14, 379–394 (2018). https://doi.org/10.1007/s11554-014-0459-1

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s11554-014-0459-1

Keywords