摘要
研究基于小波变换的图像融合方法中图像边缘效应产生的原因和消除方法.依据小波变换的定义和重构过程中所用多种边界延拓的方法,分析了边缘效应产生的原因,据此改进了传统边界延拓方法.利用先截短后扩展的延拓方法满足了图像边缘完全重建的要求.实验证明,改进后的边界延拓方法消除了原有融合图像中的边缘效应,有效地改善了融合图像质量.
Image boundary effects occur in image fusion due to wavelet transforms. To overcome it, causes leading to boundary effects are analyzed, based on the definition of wavelet and the boundary extension methods in its reconstruction. According to the analysis, the traditional boundary extension method is modified. The method introduced by shortening the signal followed by its extension satisfies the demand of perfect reconstruction. It is proved through experiments that the image fusion algorithm with the modified signal extension method eliminates the boundary effects in image fusion, and improves the quality of image fusion.
出处
《北京理工大学学报》
EI
CAS
CSCD
北大核心
2003年第4期525-528,共4页
Transactions of Beijing Institute of Technology
基金
国家"八六三"计划项目(2002AA783050)
关键词
图像融合
边界延拓
小波变换
image fusion
boundary extension
wavelet transform
