Umi-OCR
Free, MIT-licensed OCR software for Windows and Linux that reads screenshots, image batches, and scanned documents offline.
Umi-OCR turns screenshots, images, and scanned documents into text on Windows and Linux. It works offline. The software is free under the MIT License, and recognition runs locally with built-in language libraries. It suits people who need to copy text from a screen, process a folder of images, or make scans searchable without sending them to an online OCR service.
Screenshot recognition captures text from a selected area or an image pasted into the app. For larger jobs, batch recognition handles local images and saves the results as plain text, Markdown, JSONL, or CSV. Its text processing can put columns and paragraphs into reading order, retain spacing in code screenshots, and handle vertical layouts when the OCR engine supports them. You can also exclude areas that contain recurring watermarks or logos, so they don't appear in batch results.
Document recognition extracts text from PDF, XPS, EPUB, MOBI, FB2, and CBZ files. For scanned PDFs, Umi-OCR can produce a searchable PDF with the recognized text behind the page image. Ignore areas can keep headers and footers out of the extracted text. The app also reads QR codes and barcodes from images and can generate code images from text.
Umi-OCR supports the offline RapidOCR-json and PaddleOCR-json engines through plugins. It has a multilingual interface, including English, Japanese, and Traditional Chinese, and provides command-line and HTTP interfaces for people who want to connect OCR to other software.
