Journal of Real-Time Image Processing

, Volume 9, Issue 1, pp 141–157 | Cite as

High performance implementation of texture features extraction algorithms using FPGA architecture

  • Ali Reza Akoushideh
  • Asadollah Shahbahrami
  • Babak Mazloom-Nezhad Maybodi
Special Issue


The most popular second-order statistical texture features are derived from the co-occurrence matrix, which has been proposed by Haralick. However, the computation of both matrix and extracting texture features are very time consuming. In order to improve the performance of co-occurrence matrices and texture feature extraction algorithms, we propose an architecture on FPGA platform. In the proposed architecture, first, the co-occurrence matrix is computed then all thirteen texture features are calculated in parallel using computed co-occurrence matrix. We have implemented the proposed architecture on Virtex 5 fx130T-3 FPGA device. Our experimental results show that a speedup of 421[× yields over a software implementation on Intel Core i7 2.0 GHz processor. In order to improve much more performance on textures, we have reduced the computation of 13 texture features to 3 texture features using ranking of Haralick’s features. The performance improvement is 484×.


Haralick texture feature Image processing FPGA 


  1. 1.
    Matreka, A., Strzelecki, M.: Texture analysis methods: a review. Eur Cooperation Sci Technol COST B11, 90–924 (1998)Google Scholar
  2. 2.
    Valkealahti, K., Oja, E.: Reduced multidimensional co-occurrence histograms in texture classification. IEEE Trans Pattern Anal Mach Intell 12, 5–45 (1980)Google Scholar
  3. 3.
    Tahir, M.A., Bouridane, A., Kurugollu, F.: An FPGA based coprocessor for GLCM and Haralick texture features and their application in prostate cancer classification. Analog Integr. Circ. Sig. Process 43, 205–215 (2005)CrossRefGoogle Scholar
  4. 4.
    Shahbahrami, A., Pham, T., Bertels, K.: Parallel implementation of gray level co-occurrence matrices and Haralick texture features on cell architecture. J Supercomput 59(3), 1455–1477 (2012)CrossRefGoogle Scholar
  5. 5.
    Akoushideh, A. R., Shahbahrami, A.: “Accelerating texture features extraction algorithms using FPGA architecture,” International Conference on Reconfigurable Computing and FPGAs (ReConFig), pp. 232–237 (2010)Google Scholar
  6. 6.
    Haralick, R.M., Shanmugam, K., Dinstein, I.H.: Textural features for image classification. IEEE Trans Syst Man Cybern 3, 610–621 (1973)CrossRefGoogle Scholar
  7. 7.
    Hall-Beyer, M.: The GLCM Tutorial home page. (2008)
  8. 8.
    Boland, M. V.:”Quantitative description and automated classification of cellular protein localization patterns in fluorescence microscope images of mammalian cells,” Ph.D. Thesis., Pittsburgh (1999)Google Scholar
  9. 9.
    Santos, L.-E., Rene, C.:”Decision tree based FPGA-architecture for texture sea state classification.” In: IEEE International Conference on Reconfigurable Computing and FPGA’s (ReConFig), pp. 1–7 (2006)Google Scholar
  10. 10.
    du Buf, J.M.H., Kardan, M., Spann, M.: Texture feature performance for image segmentation. Pattern Recogn 23, 291–309 (1990)CrossRefGoogle Scholar
  11. 11.
    Tahir, M.A., Bouridane, A., Kurugollu, F., Amira, A.: Accelerating the computation of GLCM and Haralick texture features on reconfigurable Hardware. Int Conf Image Process 5, 2857–2860 (2004)Google Scholar
  12. 12.
    Bariamis, D.G., Iakovidis, D.K., Maroulis, D.E.: Dedicated Hardware for real-time computation of second-order statistical features for high resolution images. Lecture Notes in Computer Science in Advanced Concepts for Intelligent Vision Systems 4179, 67–77 (2006)CrossRefGoogle Scholar
  13. 13.
    Iakovidis, D.K., Maroulis, D.E., Bariamis, D.G.: FPGA architecture for fast parallel computation of co-occurrence matrices. Microprocess Microsyst 31, 160–165 (2007)CrossRefGoogle Scholar
  14. 14.
    Bariamis, D. G., Iakovidis, D. K., Maroulis, D. E., Karkanis, S. A.:”An FPGA-based architecture for real time image feature extraction. In: Proceedings of the 17th International Conference on Pattern Recognition, pp. 801–804 (2004)Google Scholar
  15. 15.
    Maroulis, D.E., Iakovidis, D.K., Bariamis, D.G.: FPGA-based System for real-time video texture analysis. J Signal Process Syst 53, 419–433 (2008)CrossRefGoogle Scholar
  16. 16.
    Gipp, M., Marcus, G., Harder, N., Suratanee, A., Rohr, K., König, R., Männer, R.:”Accelerating the Computation of Haralick texture features using graphics processing units (GPUs),” in World Congress on Engineering (2008)Google Scholar
  17. 17.
    Gipp, M., Marcus, G., Harder, N., Suratanee, A., Rohr, K., König, R., Männer, R.:”Haralick’s texture features computed by GPUs for biological applications,” IAENG Int J Comput Sci 36 (2009)Google Scholar
  18. 18.
    Picard, R., Graczyk, C., Mann, S., Wachman, J., Picard, L., Campbell, L.: MIT Media Lab: VisMod Group.
  19. 19.
    Brodatz, P. Brodatz Textures. (1966)
  20. 20.
    Tahir, M. A., Bouridane, A., Kurugollu, F.:”An FPGA Based Coprocessor for the Classification of Tissue Patterns in Prostatic Cancer,” Lecture Notes in Computer Science in Field Programmable Logic and Application (2004)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2012

Authors and Affiliations

  • Ali Reza Akoushideh
    • 1
  • Asadollah Shahbahrami
    • 2
  • Babak Mazloom-Nezhad Maybodi
    • 1
  1. 1.Electrical and Computer DepartmentShahid-Beheshti University, G.CTehranIran
  2. 2.Department of Computer EngineeringUniversity of GuilanRashtIran

Personalised recommendations