基于替换规则的对偶树复小波变换遥感影像融合  

Remote sensing image fusion based on replace rule of dual-tree complex wavelet transform

作　　者：汪西莉[1] 高卫平[1] 

机构地区：[1]陕西师范大学计算机科学学院,西安710062

出　　处：《计算机应用研究》2009年第4期1552-1553,1571,共3页Application Research of Computers

基　　金：国家自然科学基金资助项目(40671133)

摘　　要：针对对偶树复小波变换(DT-CWT)比离散小波变换具有更好的方向选择性、平移不变性,能更好地表征影像特征的优点,提出了基于替换规则的对偶树复小波变换遥感多光谱和全色影像融合算法。通过对IKO-NOS影像的实验表明,本方法比基于实数小波变换的融合方法和IHS变换法有优势,在保留了光谱特征的基础上,更好地提高了融合影像的清晰度,且方法简单高效。The dual-tree complex wavelet transform has the advantages of better direction selectivity and translation invariance compared with traditional wavelet transform. It can describe the image＇ s features much better. This paper proposed a remote sensing multi-spectral and panchromatic image fusion method based on the DT-CWT and replace rule. The experiment results of IKONOS images show that the proposed method is better than wavelet and IHS transform methods. It preserves more spectral and spatial information. It is simple and effective.

关 键 词：对偶树复小波变换 遥感影像融合 替换规则 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]