The ability to automatically extract, understand, and convert information from document images (like scans or forms) into structured, machine-readable formats.