The Parallel Algorithm for the 2-D Discrete Wavelet Transform

Czech title

Paralelní algoritmus pro 2D diskrétní vlnkovou transformaci

Type

conference paper

Language

English

Authors

Bařina David, Ing., Ph.D. (DCGM)
Najman Pavel, Ing. (DCGM)
Klepárník Petr, Ing., Ph.D. (DCGM)
Kula Michal, Ing., Ph.D. (DCGM)
Zemčík Pavel, prof. Dr. Ing., dr. h. c. (DCGM)

Keywords

discrete wavelet transform, lifting scheme, multi-core processors, parallel architecture

Abstract

The discrete wavelet transform can be found at the heart of many image-processing algorithms. Until now, the transform on general-purpose processors (CPUs) was mostly computed using a separable lifting scheme. As the lifting scheme consists of a small number of operations, it is preferred for processing using single-core CPUs. However, considering a parallel processing using multi-core processors, this scheme is inappropriate due to a large number of steps. On such architectures, the number of steps corresponds to the number of points that represent the exchange of data. Consequently, these points often form a performance bottleneck. Our approach appropriately rearranges calculations inside the transform, and thereby reduces the number of steps. In other words, we propose a new scheme that is friendly to parallel environments. When evaluating on multi-core CPUs, we consistently overcome the original lifting scheme. The evaluation was performed on 61-core Intel Xeon Phi and 8-core Intel Xeon processors.

Annotation

The discrete wavelet transform can be found at the heart of many image-processing algorithms. Until now, the transform on general-purpose processors (CPUs) was mostly computed using a separable lifting scheme. As the lifting scheme consists of a small number of operations, it is preferred for processing using single-core CPUs. However, considering a parallel processing using multi-core processors, this scheme is inappropriate due to a large number of steps. On such architectures, the number of steps corresponds to the number of points that represent the exchange of data. Consequently, these points often form a performance bottleneck. Our approach appropriately rearranges calculations inside the transform, and thereby reduces the number of steps. In other words, we propose a new scheme that is friendly to parallel environments. When evaluating on multi-core CPUs, we consistently overcome the original lifting scheme. The evaluation was performed on 61-core Intel Xeon Phi and 8-core Intel Xeon processors.

Published

2017

Pages

1–6

Journal

Proceedings of SPIE, vol. 10615, no. 4, ISSN 0277-786X

Proceedings

Ninth International Conference on Graphic and Image Processing (ICGIP 2017)

ISBN

978-1-5106-1741-4

Publisher

SPIE - the international society for optics and photonics

Place

Qingdao

DOI

10.1117/12.2302881

UT WoS

000434707200060

EID Scopus

2-s2.0-85046470425

BibTeX

@inproceedings{BUT144475,
  author="David {Bařina} and Pavel {Najman} and Petr {Klepárník} and Michal {Kula} and Pavel {Zemčík}",
  title="The Parallel Algorithm for the 2-D Discrete Wavelet Transform",
  booktitle="Ninth International Conference on Graphic and Image Processing (ICGIP 2017)",
  year="2017",
  journal="Proceedings of SPIE",
  volume="10615",
  number="4",
  pages="1--6",
  publisher="SPIE - the international society for optics and photonics",
  address="Qingdao",
  doi="10.1117/12.2302881",
  isbn="978-1-5106-1741-4",
  issn="0277-786X",
  url="https://www.fit.vut.cz/research/publication/11530/"
}