• 中国计算机学会会刊
  • 中国科技核心期刊
  • 中文核心期刊

计算机工程与科学

• 论文 • 上一篇    下一篇

基于OpenCL 的JPEG压缩算法并行化设计与实现

张敏华,张剑贤,裘雪红,周端   

  1. (西安电子科技大学计算机学院,陕西 西安 710071)
  • 收稿日期:2017-01-11 修回日期:2017-03-20 出版日期:2017-05-25 发布日期:2017-05-25
  • 基金资助:

    国家自然科学基金(61474087);中央高校基本科研业务费专项资金(JB150315)

Parallel design and implementation of
JPEG compression algorithm based on OpenCL

ZHANG Min-hua,ZHANG Jian-xian,QIU Xue-hong,ZHOU Duan   

  1. (School of Computer,Xidian University,Xi’an 710071,China)
  • Received:2017-01-11 Revised:2017-03-20 Online:2017-05-25 Published:2017-05-25

摘要:

随着图像数据的大量增加,传统单处理器或多处理器结构的计算设备已无法满足实时性数据处理要求。异构并行计算技术因其高效的计算效率和并行的实时性数据处理能力,正得到广泛关注和应用。利用GPU在图形图像处理方面并行性的优势,提出了基于OpenCL的JPEG压缩算法并行化设计方法。将JPEG算法功能分解为多个内核程序,内核之间通过事件信息传递进行顺序控制,并在GPU+CPU的异构平台上完成了并行算法的仿真验证。实验结果表明,与CPU串行处理方式相比,本文提出的并行化算法在保持相同图像质量情况下有效提高了算法的执行效率,大幅降低了算法的执行时间,并且随着图形尺寸的增加,算法效率获得明显的提升。
 
 

关键词: 异构计算, OpenCL , 并行化, JPEG

Abstract:

As the scale of information data increases enormously, traditional singleprocessor or multiprocessor structure based computing devices are unable to meet the requirements of realtime data processing. The heterogeneous parallel computing technology attracts  much attention and is widely applied for its effective computation efficiency and parallel realtime processing capability. We propose a parallel design of the JPEG compression algorithm based on OpenCL by using the advantages of the GPU in image processing. The JPEG algorithm is divided into multiple kernel programs, and the kernels are sequentially controlled by the event information transfer. The parallel algorithm is simulated and verified on the GPU+CPU platform. Experimental results show that under the same image quality condition, the parallel algorithm can improve algorithm implementation efficiency  and reduce time substantially. And as the graph size increases, the efficiency of the algorithm obtains obvious improvement.

Key words: heterogeneous computing, OpenCL, parallel processing, JPEG