一种基于小波系数局部特征的选择性遥感图像融合方法

提出了一种基于小波变换系数局部特征的自适应选择性遥感图像融合方法,方法的基本思想是根据多光谱和全色图像融合过程中小波分解后的低频和高频部分融合目的的不同,对分解得到的近似和各层各方向的细节分量分别运用3×3的移动模板逐一计算相应区域小波系数阵的不同局部特征量,然后选择适当的阈值,对图像的近似和细节分量分别应用不同的策略在小波域内进行选择性融合,最后通过小波和IHS逆变换得到融合的高分辨率多光谱图像。对比实验结果表明:本文的方法在较好地保留空间细节信息的同时,图像的光谱信息也得到了最大限度地保持。

A Selective Remote Sensing Image Fusion Method Based on the Local Feature of Wavelet Coefficients

In this paper,an adaptively selective remote sensing images fusion method was proposed based on the local feature of wavelet coefficients to address this problem.The basic idea of the method as follows.First,a 3×3 mobile template was used to calculate the different local features of corresponding wavelet coefficient matrix one by one for the approximate components and detail components of each direction,each layer resulting from wavelet transform according to the different fusion purposes of low and high frequency part in the process of multi-spectral and panchromatic images fusion.Then the approximate images and detail images were infused selectively in wavelet domain by applying different fusion rules based on proper threshold.Finally,the resultant image with high resolution and multi-spectral characteristics was obtained by inverse wavelet transform and inverse IHS transform.The comparative experiments show that the proposed method can preserve spectral information of image to greatest extent as well as better retain the high spatial resolution details.