Open-Capture is a web-based, open-source document capture and OCR platform developed by Edissyum, a French company based in Carpentras. It combines OCR with machine learning to ingest documents from scanners, email, uploads and APIs. Metadata is extracted using regular expressions, learning, QR codes or a pre-trained AI model; operators then validate it and export it to document management and business systems. Modules include Splitter, Verifier and Mailcollect. Open-Capture is written in Python, runs on Linux and has an official Docker setup; Edissyum offers demonstrations, integration, configuration, training and support.
Edissyum is based in Carpentras, France. The source code is published on GitHub under the GNU GPL v3 licence, so you can install and operate the software yourself on your own servers in the EU, without an external cloud provider. This suits European organisations that want to process documents under their own control. No foreign parent company is mentioned.