Scalable Hyperspectral Enhancement via Patch-Wise Sparse Residual Learning: Insights from Super-Resolved EnMAP Data

In this contribution, we propose a novel parallel patch-wise sparse residual learning (P2SR) algorithm for resolution enhancement based on fusion of hyperspectral imaging data (HSI) and multispectral imaging data (MSI). The proposed method uses multi-decomposition techniques to extract spatial and spectral features to form a sparse dictionary. The spectral and spatial characteristics of the scene encoded as a dictionary enables reconstruction through a first order optimization algorithm to ensure an efficient sparse representation. The final spatially enhanced HSI is reconstructed using sparse-dictionary features from low resolution HSI and applying a MSI regulated guided filter to enhance spatial fidelity while minimizing artifacts. P2SR is deployable on a high-performance computing (HPC) system with parallel processing, ensuring scalability and computational efficiency for large HSI datasets. Extensive evaluations on three diverse study sites demonstrate that P2SR consistently outperforms traditional and state-of-the-art (SOA) methods in both quantitative metrics and qualitative spatial assessments. P2SR displays superior spatio-spectral reconstruction contributing to sharper spatial features, reduced mixed pixels, and enhanced geological features. Importantly, we show that P2SR preserves critical spectral signatures such as Fe²⁺ absorption and improves the detection of fine-scale environmental and geological structures. P2SR’s ability to maintain spectral fidelity while enhancing spatial detail makes it a powerful tool for high-precision remote sensing applications, including mineral mapping, land-use analysis, and environmental monitoring.