Deep Hyperspectral and Multispectral Image Fusion with Inter-image Variability

TL;DR

This paper introduces a novel hyperspectral and multispectral image fusion method that models inter-image variability and uses CNN-based priors, improving fusion accuracy under realistic acquisition conditions.

Contribution

It presents a new imaging model considering inter-image variability, a reweighted optimization scheme, and a zero-shot learning approach for image-specific priors.

Findings

Effective fusion on real data with inter-image variability

Outperforms existing methods in realistic scenarios

Unsupervised, image-specific prior learning enhances results

Abstract

Hyperspectral and multispectral image fusion allows us to overcome the hardware limitations of hyperspectral imaging systems inherent to their lower spatial resolution. Nevertheless, existing algorithms usually fail to consider realistic image acquisition conditions. This paper presents a general imaging model that considers inter-image variability of data from heterogeneous sources and flexible image priors. The fusion problem is stated as an optimization problem in the maximum a posteriori framework. We introduce an original image fusion method that, on the one hand, solves the optimization problem accounting for inter-image variability with an iteratively reweighted scheme and, on the other hand, that leverages light-weight CNN-based networks to learn realistic image priors from data. In addition, we propose a zero-shot strategy to directly learn the image-specific prior of the latent images in an unsupervised manner. The performance of the algorithm is illustrated with real data subject to inter-image variability.