Publication detail

GPU Acceleration of 2D-DWT Image Compression in MATLAB with CUDA

ŠIMEK, V.

Original Title

GPU Acceleration of 2D-DWT Image Compression in MATLAB with CUDA

English Title

GPU Acceleration of 2D-DWT Image Compression in MATLAB with CUDA

Type

conference paper

Language

en

Original Abstract

This article will present the details about the acceleration of 2D wavelet-based medical data (image) compression on MATLAB with CUDA. It is obvious that the diagnostic materials (mostly as a certain type of image) are increasingly acquired in a digital format. Therefore, common need to daily manipulate huge amount of data brought about the issue of compression within a very less stipulated amount of time. Attention will be given to the acceleration processing flow which exploits the massive parallel computational power offered by the latest NVIDIA graphics processor unit (GPU). It brings a compute device that can be programmed using a C-like language using CUDA, (Compute Unified Device Architecture). In the same time, a number of attractive features can be exploited for a broad class of intensive data parallel computation tasks. The final part of discussion outlines possible directions towards future improvements of compression ratio and processing speed.

English abstract

This article will present the details about the acceleration of 2D wavelet-based medical data (image) compression on MATLAB with CUDA. It is obvious that the diagnostic materials (mostly as a certain type of image) are increasingly acquired in a digital format. Therefore, common need to daily manipulate huge amount of data brought about the issue of compression within a very less stipulated amount of time. Attention will be given to the acceleration processing flow which exploits the massive parallel computational power offered by the latest NVIDIA graphics processor unit (GPU). It brings a compute device that can be programmed using a C-like language using CUDA, (Compute Unified Device Architecture). In the same time, a number of attractive features can be exploited for a broad class of intensive data parallel computation tasks. The final part of discussion outlines possible directions towards future improvements of compression ratio and processing speed.

Keywords

GPU, CUDA, 2D wavelet transform, image compression, Matlab

RIV year

2008

Released

08.09.2008

Publisher

IEEE Computer Society

Location

Liverpool

ISBN

978-0-7695-3325-4

Book

Proceedings 2nd UKSim European Symposium on Computer Modelling and Simulation

Edition

NEUVEDEN

Edition number

NEUVEDEN

Pages from

274

Pages to

277

Pages count

4

Documents

BibTex


@inproceedings{BUT32107,
  author="Václav {Šimek}",
  title="GPU Acceleration of 2D-DWT Image Compression in MATLAB with CUDA",
  annote="This article will present the details about the acceleration of 2D wavelet-based
medical data (image) compression on MATLAB with CUDA. It is obvious that the
diagnostic materials (mostly as a certain type of image) are increasingly
acquired in a digital format. Therefore, common need to daily manipulate huge
amount of data brought about the issue of compression within a very less
stipulated amount of time. Attention will be given to the acceleration processing
flow which exploits the massive parallel computational power offered by the
latest NVIDIA graphics processor unit (GPU). It brings a compute device that can
be programmed using a C-like language using CUDA, (Compute Unified Device
Architecture). In the same time, a number of attractive features can be exploited
for a broad class of intensive data parallel computation tasks. The final part of
discussion outlines possible directions towards future improvements of
compression ratio and processing speed.",
  address="IEEE Computer Society",
  booktitle="Proceedings 2nd UKSim European Symposium on Computer Modelling and Simulation",
  chapter="32107",
  edition="NEUVEDEN",
  howpublished="print",
  institution="IEEE Computer Society",
  year="2008",
  month="september",
  pages="274--277",
  publisher="IEEE Computer Society",
  type="conference paper"
}