A survey on computational spectral reconstruction methods from RGB to hyperspectral imaging

TL;DR

This paper reviews over 25 spectral reconstruction methods from RGB images to hyperspectral imaging, highlighting the superiority of data-driven deep learning approaches in accuracy and quality, and serving as a comprehensive reference for future research.

Contribution

It provides a systematic comparison and analysis of existing spectral reconstruction methods, emphasizing the advancements of deep learning techniques over prior-based methods.

Findings

Deep learning methods outperform prior-based methods in accuracy and quality.

Most data-driven approaches are faster and more effective.

The survey offers a valuable reference for future research directions.

Abstract

Hyperspectral imaging enables versatile applications due to its competence in capturing abundant spatial and spectral information, which are crucial for identifying substances. However, the devices for acquiring hyperspectral images are expensive and complicated. Therefore, many alternative spectral imaging methods have been proposed by directly reconstructing the hyperspectral information from lower-cost, more available RGB images. We present a thorough investigation of these state-of-the-art spectral reconstruction methods from the widespread RGB images. A systematic study and comparison of more than 25 methods has revealed that most of the data-driven deep learning methods are superior to prior-based methods in terms of reconstruction accuracy and quality despite lower speeds. This comprehensive review can serve as a fruitful reference source for peer researchers, thus further inspiring future development directions in related domains.