OnlineOCR vs Open-Capture

OnlineOCR has no write-up on Specifiction yet; its figures are above.

Open-Capture collects documents, extracts information with OCR and AI, and exports results to third-party management software. It accepts material from scanners, copiers, file drops, email and attachments, web forms, uploads, and web services. Extraction options include Tesseract OCR, pre-trained models, machine-learning classification, key-value extraction, reference-data matching, dictionaries, and barcode or QR-code reading. Extracted fields can be checked against rules, formats, reference data, and scripted business rules. The full-web application runs natively on Linux and can be hosted on-premise or in a public or private cloud. It is free and open-source under the GNU GPLv3 license, with no limit on users. Optional modules include Splitter for preparing batches, Verifier for checking extracted data, and MailCollect for email processing. MailCollect analyzes messages, identifies senders, extracts metadata, and routes attachments into processing workflows. Verifier is described for structured or semi-structured documents such as invoices, delivery notes, contracts, and purchase orders. Listed connections include a native MEM Courrier connector, CMIS, OpenADS/Oxalis, OpenCRM, Alfresco, and SharePoint; export formats include PDF, PDF/A, XML, CSV, and plain text.