Deep Learning Hyperspectral Pansharpening on large scale PRISMA dataset

TL;DR

This paper evaluates deep learning methods for hyperspectral pansharpening using a new large-scale dataset from the PRISMA satellite, demonstrating neural networks outperform traditional approaches in both supervised and real-world scenarios.

Contribution

Introduces a new extensive hyperspectral dataset from PRISMA and assesses deep learning strategies, showing their superiority over traditional methods in pansharpening tasks.

Findings

Neural network methods outperform traditional approaches.

Deep learning models adapt better to hyperspectral pansharpening.

The new dataset enables comprehensive evaluation of methods.

Abstract

In this work, we assess several deep learning strategies for hyperspectral pansharpening. First, we present a new dataset with a greater extent than any other in the state of the art. This dataset, collected using the ASI PRISMA satellite, covers about 262200 km2, and its heterogeneity is granted by randomly sampling the Earth's soil. Second, we adapted several state of the art approaches based on deep learning to fit PRISMA hyperspectral data and then assessed, quantitatively and qualitatively, the performance in this new scenario. The investigation has included two settings: Reduced Resolution (RR) to evaluate the techniques in a supervised environment and Full Resolution (FR) for a real-world evaluation. The main purpose is the evaluation of the reconstruction fidelity of the considered methods. In both scenarios, for the sake of completeness, we also included machine-learning-free approaches. From this extensive analysis has emerged that data-driven neural network methods outperform machine-learning-free approaches and adapt better to the task of hyperspectral pansharpening, both in RR and FR protocols.