| 基于CUDA的JPEG压缩算法优化 |
| |
| 引用本文： | 童炜,刘铎.基于CUDA的JPEG压缩算法优化[J].通信技术,2011,44(4):1-4,9. |
| |
| 作者姓名： | 童炜 刘铎 |
| |
| 作者单位： | 1. 中国电子科技集团公司第三十研究所,四川成都,610041
2. 西北工业大学计算机学院,陕西西安,710072 |
| |
| 摘 要： | 随着图形处理器（GPU）的处理能力的不断增强,图形处理器越来越多地被应用于计算密集型的数据运算处理中。JPEG图像压缩算法中的部分步骤存在典型的并行特性,针对大分辨率图像JPEG压缩串行顺序执行时间开销较大的问题,利用CUDA的并行计算和图形硬件的可编程性,可实现对JPEG图像压缩的加速,同时结合GPGPU硬件结构可实现JPEG压缩程序程序的优化设计。通过程序测试实验,与串行程序比较加速比在20以上。
|
| 关 键 词： | 图形处理器 统一计算设备架构 图像压缩 并行计算 |
Optimization of JPEG Compression Algorithm based on CUDA |
| |
| TONG Wei,LIU Duo.Optimization of JPEG Compression Algorithm based on CUDA[J].Communications Technology,2011,44(4):1-4,9. |
| |
| Authors: | TONG Wei LIU Duo |
| |
| Affiliation: | ②(①No.30 Institute of CETC,Chengdu Sichuan 610041,China;②School of Computer Science and Engineering,Northwestern Polytechnical University,Xi’an Shaanxi 710072,China) |
| |
| Abstract: | With the improvement of GPU(Graphic Processing Unit) performance,GPU is increasingly applied in the computing-concentrated data operation.According to the parallelism and programmability of CUDA,the acceleration for JPEG compression is addressed in this paper.Based on analysis of the GPGPU(General Purpose GPU) structure,an optimal algorithm is also proposed.Finally,the speeding-up factor,no less than 20,is concluded in experiment. |
| |
| Keywords: | GPU CUDA picture compression parallel computing |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
