World Scientific
  • Search
Skip main navigation

Cookies Notification

We use cookies on this site to enhance your user experience. By continuing to browse the site, you consent to the use of our cookies. Learn More
×
Our website is made possible by displaying certain online content using javascript.
In order to view the full content, please disable your ad blocker or whitelist our website www.worldscientific.com.

System Upgrade on Tue, Oct 25th, 2022 at 2am (EDT)

Existing users will be able to log into the site and access content. However, E-commerce and registration of new users may not be available for up to 12 hours.
For online purchase, please visit us again. Contact us at [email protected] for any enquiries.

PARALLELIZATION OF WAVELET FILTERS USING SIMD EXTENSIONS

    Much work has been done to optimize wavelet transforms for SIMD extensions of modern CPUs. However, these approaches are mostly restricted to the vertical part of 2-D transforms with line-wise organized memory layouts because this leads to a rather straight forward SIMD-implementation. This work shows for an example of a common wavelet filter new approaches to use SIMD operations on 1-D transforms that are able to produce reasonable speedups. As a result, the performance of algorithms that use wavelet transforms, such as JPEG2000, can be increased significantly. Various variants of parallelization are presented and compared. Their advantages and disadvantages for general filters are discussed.

    References

    • ISO/IEC JPEG committee. JPEG 2000 image coding system — ISO/IEC 15444-1:2000, December 2000 . Google Scholar
    • R. Kutil and A. Uhl, Journal of Computing and Information Technology (Special Issue on Parallel Numerics and Parallel Computing in Image Processing, Video Processing, and Multimedia) 8(1), 31 (2000). Google Scholar
    • M.-L. Woo, Parallel discrete wavelet transform on the Paragon MIMD machine, Proceedings of the seventh SIAM conference on parallel processing for scientific computing, eds. R. S. Schreiberet al. (1995) pp. 3–8. Google Scholar
    • J. Fridman and E. S. Manolakos, Multidimensional Systems and Signal Processing 8(1–2), 185 (1997). Crossref, ISIGoogle Scholar
    • M. M. Pic, H. Essafi and D. Juvin, Wavelet transform on parallel SIMD architectures, SPIE Proceedings 1961, Visual Information Processing II (), eds. F. O. Huck and R. D. Juday (SPIE, 1993) pp. 316–323. Google Scholar
    • C. Chakrabarti and M. Vishvanath, IEEE Transactions on Signal Processing 3(43), 759 (1995). ISIGoogle Scholar
    • M. Feil and A. Uhl, Wavelet packet decomposition and best basis selection on massively parallel SIMD arrays, Proceedings of the International Conference "Wavelets and Multiscale Methods" (IWC'98) (INRIA, Rocquencourt, 1998) p. 4. Google Scholar
    • C.   Tenllado et al. , Vectorization of the 2D wavelet lifting transform using SIMD extensions , Workshop on Parallel and Distributed Image Processing, Video Processing, and Multimedia, PDIVM '03 ( 2003 ) . Google Scholar
    • D.   Chaver et al. , 2-D wavelet transform enhancement on general-purpose microprocessors: Memory hierarchy and SIMD parallelism exploitation , Proceedings of the 2000 International Conference on High Performance Computing . Google Scholar
    • C. Chrysafis and A. Ortega, IEEE Transactions on Image Processing 9(3), 378 (2000). Crossref, ISIGoogle Scholar
    • I. Daubechies and W. Sweldens, Journal of Fourier Analysis Applications 4(3), 245 (1998). ISIGoogle Scholar
    • Markus Püschelet al., Proceedings of the IEEE, special issue on "Program Generation, Optimization, and Adaptation" 93(2), 232 (2005). Google Scholar