Wavelet-based PAN and Multispectral Image Fusion

A new algorithm is developed to fuse a high-resolution panchromatic (PAN) image and a low-resolution multispectral (MS) image based on wavelet transform. Firstly, this paper clarifies the concepts of and relationship between image fusion rule and operator based on wavelet transform; secondly, we propose a new strategy of calculating spatial frequency in accordance with wavelet decomposition characteristic. We adopt activity level comparison based on the spatial frequency to fuse the detail coefficient and weighted average method to fuse approximate image. Finally, this paper compares the fused results based on various wavelets, and presents objective and quantitative evaluation and analysis for the fused images