Abstract Existing snapshot hyperspectral (HS) cameras suffer from low spatial resolution, which limits their applications. Here, we develop a novel fusion architecture for high-resolution video-rate HS imaging. The architecture adopts a coaxial dual optical path structure, where one path acquires the HS video with high signal-to-noise ratio (SNR) via mosaic coding, and the other path acquires the panchromatic (PAN) video with higher spatial resolution. Besides, we customize an efficient fusion algorithm that enhances the spatial resolution of HS video by leveraging the spatial information from PAN video. Following this architecture, we develop a visible hyperspectral camera (VIS-HS) and a near-infrared hyperspectral camera (NIR-HS), both capable of capturing 16 spectral bands. The VIS-HS offers a spatial resolution of 2048 × 2048 pixels at up to 90 frames per second (fps), while the NIR-HS provides 2048×2208 pixels at up to 60 fps. Furthermore, we showcase the broad applications of our cameras including the recognition of real and artificial objects for material discrimination, segmentation of sand and soil for land desertification monitoring, identification of drug components for pharmaceutical safety, and drone tracking for dynamic target monitoring.
Li et al. (Wed,) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: