• 中国计算机学会会刊
  • 中国科技核心期刊
  • 中文核心期刊

J4 ›› 2010, Vol. 32 ›› Issue (9): 34-38.doi: 10.3969/j.issn.1007130X.2010.

• 论文 • 上一篇    下一篇

像素级遥感图像融合并行算法研究与实现

张灿峰,周海芳   

  1. (国防科学技术大学计算机学院,湖南 长沙 410073)
  • 收稿日期:2010-03-03 修回日期:2010-06-12 出版日期:2010-09-02 发布日期:2010-09-08
  • 通讯作者: 张灿峰
  • 作者简介:张灿峰(1983),男,湖南岳阳人,硕士,研究方向为遥感图像融合与并行计算;周海芳,博士,副教授,研究方向为高性能图形图像处理与并行计算。
  • 基金资助:

    国家863计划资助项目(2007AA12Z147)

Research and Implementation of the Parallel Algorithms of PixelLevel Image Fusion for RemoteSensing

ZHANG Canfeng,ZHOU Haifang   

  1. (School of Computer Science,National University of Defense Technology,Changsha 410073,China)
  • Received:2010-03-03 Revised:2010-06-12 Online:2010-09-02 Published:2010-09-08

摘要:

本文针对遥感图像IHS、HPF、DWT等典型的像素级融合算法,提出并实现了相应的基于数据并行的并行融合算法PIHS、PHPF、PDWT,并在算法时空复杂度分析的基础上进行了通信、I/O优化。针对IKONOS卫星遥感图像在机群系统上的测试结果表明,我们提出的并行算法可获得良好的并行加速比,并行效率较高。这三类算法适合于对实时性要求比较高的遥感应用领域。

关键词: 遥感图像融合, IHS, HPF, DWT, 并行化

Abstract:

In this paper,corresponding to the image fusion algorithms of IHS,HPF,DWT and other typical  remote sensing image fusion algorithms,the data parallel algorithms such as PIHS,PHPF,PDWT are proposed. Based on the algorithm complexity analysis of spacetime,the communication and I/O optimization are carried out. The IKONOS satellite images tested on a cluster system  show that our proposed parallel algorithm can obtain good parallel speedup and high parallel efficiency. These three algoriths are suitable for realtime remote sensing applications.

Key words: remote sensing image fusion, IHS, HPF, SWT, parallel algorithm