Near Infrared

Near-infrared (NIR) spectroscopy and imaging are experiencing a surge in research activity, driven by their unique capabilities in diverse applications. Current research focuses on improving image quality through techniques like image-to-image translation using vision foundation models and addressing challenges like domain adaptation between NIR and visible light data with methods such as low-rank adaptation and self-supervised learning. These advancements are significantly impacting fields ranging from medical diagnostics (e.g., non-invasive glucose monitoring, vein detection) to remote sensing (plant health monitoring) and industrial automation (quality control), enabling more accurate, efficient, and accessible solutions.