UB-Mannheim/zotero-ocr

What it solves

Zotero OCR allows users to make non-searchable PDFs within their Zotero library searchable by adding a text layer to them. It solves the problem of having scanned documents that cannot be indexed or searched for specific keywords.

How it works

The plugin integrates Tesseract OCR and pdftoppm (from Poppler tools) to process PDF files. It converts PDF pages to images and uses Tesseract to recognize text, then generates a new PDF with a recognized text layer, a text-only note, or HTML (hOCR) files for verification.

Who it’s for

Researchers, academics, and anyone using Zotero to manage a large library of scanned PDFs that need to be converted into searchable documents.

Highlights

• Integration with Zotero: Operates directly from the Zotero context menu for selected PDFs.
• cantidad of Output Options: Can produce a new searchable PDF, a text-only note, or HTML files for verification.
• Customizable Settings: Users can adjust DPI, Tesseract Page Segmentation Mode (PSM), and the OCR language/script.
• Flexible Attachment Handling: Allows choosing between adding new PDFs as normal attachments or as linked files, or overwriting the original PDF.