Pansharpening of Multispectral Images based on Cycle-spinning Quincunx Lifting Transform

Pansharpening is a process of fusing the multispectral (MS) images with the panchromatic (PAN) image to improve the spatial resolution of the MS images. The key of pansharpening is how to extract the lost detail from the PAN image and add it to the MS images with an appropriate injection model. In this paper, a pansharpening approach based on cycle-spinning quincunx lifting transform (CQLT) is proposed. The CQLT features translation invariance and vanishing moments which can extract the detail properly. In order to reduce the spectral distortion of the fused image, the histogram matching along with two popular injection models are involved in the fusion. Experimental results show that the proposed method has better quantitative results as well as competitive visual results compared with some other state-of-the-art algorithms.