Self-Supervised Hyperspectral Inpainting with the Optimisation inspired Deep Neural Network Prior
This addresses data quality issues in hyperspectral imaging for applications like remote sensing, though it appears incremental as it builds on existing plug-and-play and deep prior methods.
The paper tackles the problem of missing pixels and bands in hyperspectral images due to noise and instrumental errors, introducing a self-supervised model called LRS-PnP-DIP that achieves state-of-the-art inpainting performance in experiments with real data.
Hyperspectral Image (HSI)s cover hundreds or thousands of narrow spectral bands, conveying a wealth of spatial and spectral information. However, due to the instrumental errors and the atmospheric changes, the HSI obtained in practice are often contaminated by noise and dead pixels(lines), resulting in missing information that may severely compromise the subsequent applications. We introduce here a novel HSI missing pixel prediction algorithm, called Low Rank and Sparsity Constraint Plug-and-Play (LRS-PnP). It is shown that LRS-PnP is able to predict missing pixels and bands even when all spectral bands of the image are missing. The proposed LRS-PnP algorithm is further extended to a self-supervised model by combining the LRS-PnP with the Deep Image Prior (DIP), called LRS-PnP-DIP. In a series of experiments with real data, It is shown that the LRS-PnP-DIP either achieves state-of-the-art inpainting performance compared to other learning-based methods, or outperforms them.