Definition
OCR, or optical character recognition, automatically converts text visible in a scan, photo, or PDF file into searchable, machine-processable data. In a business process, it reads a document number, date, amount, or business partner details, but the result depends on the image quality, document layout, and model used. The extracted data should therefore be validated, and fields with accounting, legal, or personal significance require user review. Astrafox does not train proprietary OCR models; it uses existing solutions and configures them for the client’s process. Explore process automation.