Infusing Learned Priors into Model-Based Multispectral Imaging
This work addresses multispectral image reconstruction for imaging applications, but it is incremental as it builds on existing regularization by denoising methods.
The paper tackles the problem of reconstructing multispectral images from noisy linear measurements by introducing a new algorithm that uses a learned 3D deep neural net denoiser as a prior, achieving generalizability where a single trained model solves multiple imaging problems.
We introduce a new algorithm for regularized reconstruction of multispectral (MS) images from noisy linear measurements. Unlike traditional approaches, the proposed algorithm regularizes the recovery problem by using a prior specified \emph{only} through a learned denoising function. More specifically, we propose a new accelerated gradient method (AGM) variant of regularization by denoising (RED) for model-based MS image reconstruction. The key ingredient of our approach is the three-dimensional (3D) deep neural net (DNN) denoiser that can fully leverage spationspectral correlations within MS images. Our results suggest the generalizability of our MS-RED algorithm, where a single trained DNN can be used to solve several different MS imaging problems.