SSAU-Net: A Spectral–Spatial Attention-Based U-Net for Hyperspectral Image Fusion

Compared with traditional remoting image, there is a large amount of spectral information in the hyperspectral image (HSI), which makes HSI better reflect actual condition surface features. However, due to limitations imaging conditions, tends have lower spatial resolution. In order overcome this issue, we propose spectral-spatial attention-based U-Net named SSAU-Net for and multispectral (MSI) fusion. The constructs attention module by coordinate-attention (CA) an efficient pyramid split (ESPA) module, can enhance image’s information. Meanwhile, proposed network fully extracts shallow deep features images, finally generates high-resolution (HR) images. state-of-the-art HSI-MSI fusion methods, experimental results verify that method has subjective objective effect.