Extraction models on general shopping receipts, identity cards and most of the official licenses are already available.
FormExtractor leverages the power of existing OCR services to deliver accurate results. Google Vision API, Azure Computer Vision OCR and Tesseract are currently supported.
Users can submit forms, configure extraction parameters, label fields, add and manage existing models on our decorated web-based portal.
Uploaded photos are passed a set of optimazation processes including light and contrast adjustment, skew correction and edge detection to boost the extraction accuracy.
Learn more about how FormExtractor can assist your Business Process Automation
Traditional OCR engines only locate fragments of text and yield a string combining all of them. No structure nor labeling.
Traditional ones only allow customization and management via complex code.
This charity scheme aimed to provide financial support to small and medium-sized enterprises (SMEs) who were going through a difficult time in Hong Kong. More than 45000 applications were submitted and comfortably handled by an online registration system, where FormExtractor played an integral part.
Applicants could simply take photos of their licenses / documents and had relative parts of the registration form automatically filled.
FormExtractor assisted Wilson Parking, one of the largest car park management company to automate the process of submitting and scanning shopping receipts.