Spatial resolution enhancement for hyperspectral image based on wavelet Bayesian fusion

The fusion of multispectral (MS) and hyperspectral (HS) images to enhance the spatial resolution of the latter is considered in this paper, and a Bayesian estimation based fusion technique is proposed. The technique works in the wavelet domain, assuming a joint normal model for the images and a blurring and additive noise imaging model for the HS image. To keep the calculation feasible, an appropriately simplified estimation strategy is also proposed. In the simulated experiment, the technique is validated and also compared to its counterpart in the image domain.