一种新的基于主分量变换与小波变换的图像融合方法

A New Method for Image Fusion Based on Principal Component Analysis and Wavelet Transforms

为了更好地进行不同分辨率图像的融合,提出了一种基于主分量变换与小波变换结合的多光谱图像与高分辨率图像融合方法。该新方法首先对多光谱图像进行主分量变换;然后分别对其第1主分量与高分辨率图像进行小波变换,并采用成像强度对比法有效地将经小波分解的高分辨率图像的低频分量信息融合到经小波分解的多光谱图像的第1主分量的低频分量中;最后,通过将小波融合结果作为多光谱图像的第1主分量再做逆主分量变换来得到最终的融合图像。实验结果分析表明,该新方法使融合图像在较好地保留光谱信息的同时,空间细节信息也得到了增强,比典型的IHS变换、主分量变换及小波变换融合方法具有更好的融合效果。

In this paper,a new method of image fusion method was proposed based on principal component analysis(PCA) and wavelet transforms(WT) to fuse a low-resolution multi-spectral image and high-resolution panchromatic image.Firstly,the multi-spectral image is transformed with PCA.Then,the first principal component of multi-spectral image and high-resolution panchromatic image were transformed by WT,respectively.The high-resolution panchromatic image approximation can be merged into the multi-spectral image approximation using antitheses of imaging intensity by means of multi-resolution wavelet decomposition,and then the merged image is reconstructed by inverse wavelet transform and replaces the first principal component of multi-spectral image.Finally,the fusion image is obtained by taking the inverse PCA transform.The experiment indicates that the performance of new method is better than those of IHS transform-based fusion method,PCA transform-based one,and WT transform-based one.It not only preserves spectral information of the original multi-spectral image well,but also enhances spatial detail information.