The following information was submitted:
Transactions: INTERNATIONAL JOURNAL of COMPUTERS
Transactions ID Number: 19-918
Full Name: Dietmar Wippig
Position: Doctor (Researcher)
Age: ON
Sex: Male
Address: Rodderweg 6a, D-50374 Erftstadt
Country: GERMANY
Tel: 02235/4599311
Tel prefix: +49
Fax:
E-mail address: wippig@hsu-hamburg.de
Other E-mails: dietmar.wippig@gmx.de
Title of the Paper: GPU-Based Translation-Invariant 2D Discrete Wavelet Transform for Image Processing
Authors as they appear in the Paper: Dietmar Wippig, and Bernd Klauer
Email addresses of all the authors: wippig@hsu-hamburg.de,bernd.klauer@hsu-hamburg.de
Number of paper pages: 9
Abstract: The Discrete Wavelet Transform (DWT) is applied to various signal and image processing applications. However the computation is computational expense. Therefore plenty of approaches have been proposed to accelerate the computation. Graphics processing units (GPUs) can be used as stream processor to speed up the calculation of the DWT. In this paper, we present a implementation of the translation-invariant wavelet transform using consumer level graphics hardware. As our approach was motivated by infrared image processing our implementation focuses on gray-level images, but can be also used in color image processing applications. Our experiments show, that the computation performance of the DWT could be significantly improved. However, initialisation and data transfer times are still a problem of GPU implementations. They could dramatically reduce the achievable performance, if they cannot be hided by the application. This effect was also observed integrating our imp!
lementation in wavelet-based edge detection and wavelet denoising.
Keywords: Parallel discrete wavelet transform, Algorithme à trous, Image processing, GPU, Shader
EXTENSION of the file: .doc
Special (Invited) Session: Translation-Invariant Two-Dimensional Discrete Wavelet Transform on Graphics Processing Units
Organizer of the Session: 104-149
How Did you learn about congress:
IP ADDRESS: 88.77.163.155
