Title:Hyperspectral Image Super-Resolution via Deep Prior Regularization with Parameter Estimation

Abstract:Hyperspectral image (HSI) super-resolution is commonly used to overcome the hardware limitations of existing hyperspectral imaging systems on spatial resolution. It fuses a low-resolution (LR) HSI and a high-resolution (HR) conventional image of the same scene to obtain an HR HSI. In this work, we propose a method that integrates the physical model and deep prior information. A novel, yet effective two-stream fusion network is designed and its output serves as a regularization term of the fusion problem. Then, we optimize the fusion problem by solving a Sylvester equation, and simultaneously estimate the regularization parameter to automatically adjust contribution of the physical model and learnt prior to reconstruct the final HR HSI. Experimental results on two public datasets demonstrate the superiority of the proposed method over several state-of-the-art methods on both quantitative and qualitative comparisons.