基于小波变换的适合Quick Bird遥感影像的融合方法

Fusion method fitting Quick Bird remote sensing image based on wavelet transform.

在分析了IHS变换、PCA变换、Brovey变换、加权融合、PMⅠ(Proposed methodⅠ)等一般融合方法的基础上,针对Quick Bird全色与多光谱影像的融合,提出了基于átrous小波变换的小波比值方法;然后通过目视判别、定量统计参数两种方法对各融合结果进行了比较.结果表明,小波分解层数为3的小波比值方法既很大程度地提高了空间纹理细节信息,又高度保持了原始多光谱影像的光谱特征,适合于Quick Bird遥感影像融合.

By analyzing the general fusion methods of IHS transform, PCA transform, Brovey transform, weighting fusion and PMⅠ（Proposed Method Ⅰ ）, the wavelet-ratio method based on wavelet transform is proposed for the fusion of Quick Bird panchromatic and multispeetral images. Visual judgment and quantitative statistical parameters are used to assess the results of different fusion methods. The result shows the wavelet-ratio method whose wavelet decomposition number is 3 is the very method that suits for the fusion of Quick Bird images. It improves the high spatial information apparently and keeps the spectral characteristics of original multispectral images highly at the same time.