Hyperspectral Image Super-Resolution Based on Spatial-Spectral Feature Extraction Network

Constrained by the physics of hyperspectral sensors, spatial resolution images (HSI) is low. Hyperspectral image super-resolution (HSI SR) a task to obtain high-resolution from low-resolution images. Existing algorithms have problem losing important spectral information while improving resolution. To handle this problem, spatial-spectral feature extraction network (SSFEN) for HSI SR proposed in paper. It enhances preserving information. The SSFEN composed three parts: mapping network, reconstruction and fusing network. And joint loss function with constraints designed guide training SSFEN. Experiment results show that method improves effectively preserves simultaneously.