Accelerated 2D Image Processing on GPUs
Georgia College & State University, Department of ISCM, Milledgeville, GA 31061
Lecture Notes in Computer Science, Vol. 3515 (January 2005), pp. 256-264
@article{payne2005accelerated,
title={Accelerated 2D image processing on GPUs},
author={Payne, B.R. and O. Belkasim, S. and Owen, G.S. and C. Weeks, M. and Zhu, Y.},
journal={Computational Science–ICCS 2005},
pages={256–264},
year={2005},
publisher={Springer}
}
Graphics processing units (GPUs) in recent years have evolved to become powerful, programmable vector processing units. Furthermore, the maximum processing power of current generation GPUs is roughly four times that of current generation CPUs (central processing units), and that power is doubling approximately every nine months, about twice the rate of Moore’s law. This research examines the GPU’s advantage at performing convolutionbased image processing tasks compared to the CPU. Straight-forward 2D convolutions show up to a 130:1 speedup on the GPU over the CPU, with an average speedup in our tests of 59:1. Over convolutions performed with the highly optimized FFTW routines on the CPU, the GPU showed an average speedup of 18:1 for filter kernel sizes from 3×3 to 29×29.
November 30, 2010 by hgpu
