Abstract:Spectral imaging has revolutionisedvarious fields by capturing detailed spatial and spectral information. However, its high cost and complexity limit the acquisition of a large amount of data to generalise processes and methods, thus limiting widespread adoption. To overcome this issue, a body of the literature investigates how to reconstruct spectral information from RGB images, with recent methods reaching a fairly low error of reconstruction, as demonstrated in the recent literature. This article explores the modification of information in the case of RGB-to-spectral reconstruction beyond reconstruction metrics, with a focus on assessing the accuracy of the reconstruction process and its ability to replicate full spectral information. In addition to this, we conduct a colorimetric relighting analysis based on the reconstructed spectra. We investigate the information representation by principal component analysis and demonstrate that, while the reconstruction error of the state-of-the-art reconstruction method is low, the nature of the reconstructed information is different. While it appears that the use in colour imaging comes with very good performance to handle illumination, the distribution of information difference between the measured and estimated spectra suggests that caution should be exercised before generalising the use of this approach.

What problem does this paper attempt to address?

### Problems Addressed by the Paper This paper primarily explores how to reconstruct spectral information from RGB images and evaluates the accuracy and reliability of this method in practical applications. Specifically: 1. **Limitations of Spectral Reconstruction Methods**: - Researchers conducted experiments using the latest spectral reconstruction methods (such as MST++ and A++) to verify the generalization ability and reproducibility of these methods in different scenarios. - By comparing the differences between the reconstructed spectra and the real spectra, the accuracy of existing methods is evaluated. 2. **Diversity Analysis of Spectral Data**: - PCA analysis is used to reveal the essential differences between the reconstructed spectra and the real spectra by analyzing the differences between the reconstructed spectra and the original spectra. - Results on the SIDQ and CAVE datasets are compared to analyze performance differences in different scenarios. 3. **Color Re-illumination Analysis**: - The color performance of the reconstructed spectra is evaluated under different light sources (such as D65, A, and LED-B1) by calculating the Euclidean distance in the CIELAB color space to assess color accuracy. - The robustness and generalization ability of the reconstructed spectra under different lighting conditions are verified. In summary, this paper aims to evaluate the effectiveness of methods for reconstructing spectral information from RGB images through experiments and data analysis, and to point out their potential application limitations.

Spectral Reconstruction from RGB Imagery: A Potential Option for Infinite Spectral Data?

Sequential Selection Of Representative Color Samples For Spectral Reflectance Reconstruction

Spectral Imaging with Deep Learning.

Optimal Selection of Representative Colors for Spectral Reflectance Reconstruction in a Multispectral Imaging System

Compressive Single-Pixel Hyperspectral Imaging Using RGB Sensors

Spectral Bidirectional Texture Function Reconstruction by Fusing Multiple-Color and Spectral Images.

Improved Reflectance Reconstruction for Multispectral Imaging by Combining Different Techniques.

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

Limitations of Data-Driven Spectral Reconstruction -- Optics-Aware Analysis and Mitigation

A Comprehensive Review of Spatial-Temporal-spectral Information Reconstruction Techniques

Learnable Reconstruction Methods from RGB Images to Hyperspectral Imaging: A Survey.

NTIRE 2018 Challenge on Spectral Reconstruction from RGB Images

Reconstructing hyperspectral images from rgb inputs based on intrinsic image decomposition

Hyperspectral Image Reconstruction from RGB Input Through Highlighting Intrinsic Properties

Spectral Reflectance Recovery From a Single RGB Image

Training-Based Spectral Reconstruction from a Single RGB Image

Spectral Reconstruction Using an Iteratively Reweighted Regulated Model from Two Illumination Camera Responses

Recovering fluorescent spectra with an RGB digital camera and color filters using different matrix factorizations

Spectral reconstruction from RGB image to hyperspectral image: Take the detection of glutamic acid index in beef as an example

Reconstruction of Spectral Images of Ocean Surfaces Using Standard RGB Images and Typical Ocean Spectrum Samples

Hyperspectral Image Reconstruction Based on Improved Residual Dense Network