Color transfer based remote sensing image fusion using non-separable wavelet frame transform

In order to fuse two registered high spatial resolution panchromatic image and low spatial resolution multispectral image of the same scene, we proposed a new color transfer based fusion algorithm by using the non-separable wavelet frame transform (NWFT). Three bands are selected from the source multispectral image as the channels to be fused. A grayscale image is obtained by averaging these three bands. Histogram matching is performed on the source panchromatic image to obtain a new panchromatic image with a uniform histogram as the grayscale image acquired from the source multispectral image. The histogram matched panchromatic image is decomposed by the NWFT. The lowest frequency subband of the NWFT coefficients is substituted by the grayscale image acquired from the source multispectral image in order to produce the composite NWFT coefficients. A composite image is obtained by performing the inverse NWFT transform on the combined coefficients. Three bands selected from the source multispectral image are mapped into the RGB (red-green-blue) color space. The color information (strictly speaking, it is the spectral information of the source multispectral image) is transferred into the composite image by using a color transfer method in order to get the finally fused image. Experiment results show that the proposed algorithm works well in remote sensing image fusion. (c) 2005 Elsevier B.V. All rights reserved.