计算机工程与应用 ›› 2012, Vol. 48 ›› Issue (35): 166-173.

• 图形、图像、模式识别 • 上一篇    下一篇

基于小波变换一致性检测的遥感影像融合

周  惠1,2,高红民2,田小敏1   

  1. 1.南京工业职业技术学院 计算机与软件学院,南京 210046
    2.河海大学 计算机与信息学院,南京 211100
  • 出版日期:2012-12-11 发布日期:2012-12-21

Remotely sensed image fusion based on wavelet transform and consistency detection

ZHOU Hui1,2, GAO Hongmin2, TIAN Xiaomin1   

  1. 1.College of Computer and Software, Nanjing Institute of Industry Technology, Nanjing 210046, China
    2.College of Computer and Information Engineering, Hohai University, Nanjing 211100, China
  • Online:2012-12-11 Published:2012-12-21

摘要: 通过分析一些经典遥感影像融合方法的不足, 提出了一种基于小波变换一致性检测的遥感影像融合方法。其基本思想是在基于局部方差融合方法的基础上,采用“多数”原则对各个像素的8个邻域进行一致性检测,并进行相应调整,以保持融合后的高频细节分量具有一致性。通过仿真实验以及相应评价指标的衡量证明了该方法在遥感影像融合方面的有效性和优越性。

关键词: 遥感影像融合, 小波变换, 一致性检测

Abstract: Through analyzing the shortages of some classical remote sensing image fusion methods, a fusion method based on wavelet transform and consistency detection is proposed in this paper. The basic idea is using the “majority” principle to detect the consistency of eight neighborhoods of each pixel and making corresponding adjustment, which is based on the fusion method of local variance. The purpose of consistency detection is to maintain the consistency of high-frequency component after the fusion. Experimental results and evaluating indicators show that it is possible to significantly improve the classical fusion methods with the proposed approach.

Key words: remotely sensed image fusion, wavelet transform, consistency detection