A HIGH PERFORMANCE ARCHITECTURE FOR IMPLEMENTATION OF 2-D CONVOLUTION WITH QUADRANT SYMMETRIC KERNELS

M.Z. Zhang,∗ H.T. Ngo,∗ A.R. Livingston,∗ and V.K. Asari∗

References

  1. [1] A.F. Breitzman, Automatic derivation and implementation offast convolution algorithms, Ph.D. thesis, Drexel University,2003.
  2. [2] E. Jamro, Parameterised automated generation of convolversimplemented in FPGAs, Ph.D. thesis, University of Miningand Metallurgy, 2001.
  3. [3] H.M. Chang & M.H. Sunwoo, An efficient programmable 2-D convolver chip, Proc. of the 1998 IEEE Intl. Symp. onCircuits and Systems, ISCAS, part 2, Monterey, CA, USA,1998, 429–432.
  4. [4] V. Hecht, K. Ronner, & P. Pirsch, An advanced programmable2-D convolution for real time image processing, Proc. of IEEEIntl. Symp. on Circuits and Systems, ISCAS, Singapore, 1991,1897–1900.
  5. [5] J.H. Kim & W.E. Alexander, A multiprocessor architecturefor 2-D digital filters, IEEE Trans. on Computer, C-36, 1987,876–884.
  6. [6] A.E. Nelson, Implementation of image processing algorithmson FPGA hardware, Masters thesis, Vanderbilt University,USA, 2000.
  7. [7] S.Y. Lee & J.K. Aggarwal, Parallel 2-D convolution on a meshconnected array processor, IEEE Trans. on Pattern Analysisand Machine Intelligence, PAMI-9, 1987, 590–594.
  8. [8] B. Bosi & G. Bois, Reconfigurable pipelined 2-D convolversfor fast digital signal processing, IEEE Trans. on Very LargeScale Integration (VLSI) Systems, 7(3), 1999, 299–308.
  9. [9] H.T. Kung, L.M. Ruane, & D.W.L. Yen, A two-level pipelinedsystolic array for multidimensional convolution, Image andVision Computing, 1(1), 1983, 30–36.
  10. [10] H.T. Kung & S.W. Song, A systolic 2-D convolution chip,Proc. IEEE Comput. Soc. Workshop Comput. Architecture forPattern Anal. and Image Database Management, Los Alamitos,CA, USA, 1981, 159–160.
  11. [11] H.T. Kung & R.L. Picard, Hardware pipelines for multi-dimensional convolution and resampling, Proc. IEEE Comput.Soc. Workshop Comput. Architecture for Pattern Anal. andImage Databae Management, Los Alamitos, CA, USA, 1981,273–278.
  12. [12] P. Ruetz & R. Broderson, A custom chip set for real-timeimage processing, Proc. of the 1986 ICASSP, Tokyo, Japan,1986, 801–804.
  13. [13] A. Wong, A new scalable systolic array processor architecturefor discrete convolution, Masters thesis, University of Kentucky,USA, 2003.
  14. [14] H.T. Ngo & K.V. Asari, A pipelined architecture for real-time correction of barrel distortion in wide-angle camera im-ages, IEEE Transactions of Circuits and Systems for VideoTechnology, 15(3), 2005, 436–444.
  15. [15] K.V. Asari, Design of an efficient VLSI architecture for non-linear spatial warping of wide-angle camera images, Journalof Systems Architecture, 50(12), 2004, 743–755.
  16. [16] M.J. Seow, H.T. Ngo, & K.V. Asari, Systolic implementationof 2-D block based hopfield neural network for efficient patternassociation, Journal of Microprocessors and Microsystems,27(8), 2003, 359–366.
  17. [17] H.T. Kung, Why systolic architectures?, IEEE Computer,15(1), 1982, 37–46.
  18. [18] V.R. Daggu & M. Venkatesan, Design and implementation ofan efficient reconfigurable architecture for image processingalgorithms using Handel-C, University of Nevada, Las Vegas.307
  19. [19] K. Wiatr & E. Jamro, Constant coefficient multiplicationin FPGA structures, IEEE Proc. of the 26th EuriomicroConference, Maastricht, The Netherlands, 2000, 252–259.
  20. [20] H. Samueli, An improved search algorithm for the design ofmultiplier-less FIR filters with powers-of-two coefficients, IEEETrans. on Circuits and Systems, 36(7), 1989, 1044–1047.
  21. [21] Z. Ye & C.H. Chang, Local search method for FIR filtercoefficients synthesis, Proc. 2nd IEEE Int. Workshop on Elec-tronic Design, Test and Applications (DELTA-2004), Perth,Australia, 2004, 255–260.
  22. [22] J. Yli-Kaakinen & T. Saram¨aki, A systematic algorithm forthe design of multiplierless FIR filters, Proc. IEEE Int. Symp.Circuits Syst., Sydney, Australia, 2001, 185–188.

Important Links:

Go Back