Research news on multispectral imaging

Multispectral imaging is an optical acquisition method that records scene radiance or reflectance in multiple discrete spectral bands, typically spanning the visible, near-infrared, and sometimes shortwave infrared regions. Using bandpass filters, tunable filters, or sensor arrays with spectrally selective elements, it generates co-registered images where each band captures wavelength-dependent material properties. Quantitative analysis of these bands enables spectral feature extraction, classification, and segmentation based on differences in absorption, scattering, or fluorescence. In research, multispectral imaging is applied to non-destructive assessment, tissue characterization, remote sensing, and cultural heritage analysis, often serving as a lower-complexity, higher-throughput alternative to full hyperspectral imaging.

DESI releases biggest 2D map of the universe

Hold on to your telescopes: The DESI Legacy Imaging Surveys team has released the largest-ever 2D color map of the universe. The 5.6-trillion-pixel map contains nearly 4 billion celestial objects, primarily stars and galaxies. ...

Reducing night lights in cities has ecological effects

Artificial light at night (ALAN) can cause plants to open their buds and sprout leaves earlier in the spring. ALAN can also cause trees to delay changing the color of their leaves in the fall. These changes can ripple through ...

Observing oscillations, flares and tornados on the sun

For six and a half days in July 2024, the balloon-borne solar observatory Sunrise III kept its gaze fixed on the sun. The stratospheric flight, which stretched from the northernmost tip of Sweden to Canada's Northwest Territories, ...

page 1 from 9