Overview
Umi-OCR is a free, open-source OCR application for desktop platforms that works offline. It bundles multiple recognition engines and language libraries to provide screenshot OCR, batch image import, PDF OCR, and barcode/QR code recognition. The tool is suitable for extracting text from images or documents quickly and reliably, especially in privacy-sensitive or network-restricted environments.
Key features
- Screenshot capture and recognition, batch image processing for common image formats.
- PDF OCR with options to export searchable layered PDF or plain text.
- Multi-language support and post-processing for layout and ordering.
- Barcode and QR code recognition, command-line and HTTP interfaces for integration.
Use cases
- Quick extraction of text from screenshots into notes or documents.
- Bulk conversion of scanned images or PDFs into searchable archives or structured text.
- Offline OCR in restricted or privacy-sensitive environments.
- Integration into automated document-processing pipelines via CLI or HTTP APIs.
Technical notes
- Built with Python and QML/Qt: QML provides a cross-platform desktop UI while Python handles OCR engines and plugins.
- Plugin architecture, Weblate-based localization, and multiple export formats (txt, jsonl, md, csv).
- Licensed under MIT and widely adopted by the community, making it easy to reuse or extend in commercial and research projects.