Accelerating Digital Image Compression Using CUDA
General-purpose computing on graphics processing unit (GPGPU) has been applied to many different fields from solving mathematical problem to a protein folding simulation with remarkable successes.This has been achieved because the computing power of the latest GPU architecture has significantly overtaken the existing commonly used CPU,and the new NVIDIA CUDA (Compute Unified device Architecture) technology makes GPU programming easier than before.In this paper,we present a JPEG digital image compression on GPU using a new CUDA technology.We have implemented JPEG digital image compression on GPU and CPU respectively in two different testing system environments.Our experiment has shown that JPEG digital image compression using CUDA technology is able to achieve a performance improvement of 50-107 times over the compression on CPU.
CUDA digital image GPGPU NVIDIA digital compression
Huifang DENG Zhihong GUO Chunhui DENG
School of Computer Science and Engineering South China University of Technology,University Town,5100 School of Software Engineering South China University of Technology,University Town,51006,Guangzhou, Department of Computer Engineering Guangzhou Auto College South China University of Technology No.l
国际会议
秦皇岛
英文
25-28
2010-11-05（万方平台首次上网日期，不代表论文的发表时间）