OCR PDF

About OCR PDF Tool

This free online OCR PDF tool extracts text from scanned PDFs and image-based PDF documents. It uses advanced optical character recognition (OCR) technology powered by Tesseract.js to convert images of text into editable, searchable text. All processing happens in your browser - your files are never uploaded to any server.

What types of PDFs can this tool process?

This tool is designed for scanned PDFs and image-based PDFs where the text is embedded as images rather than selectable text. If your PDF already has selectable text, you may want to use our PDF to Text tool instead for faster results.

What languages are supported?

The tool supports 15+ languages including English, Vietnamese, Chinese (Simplified and Traditional), Japanese, Korean, French, German, Spanish, Russian, Arabic, Hindi, Portuguese, Italian, and Thai. Select the primary language of your document for best results.

How does render quality affect results?

Higher render quality creates larger, more detailed images of each page, which generally produces more accurate OCR results. However, it also takes longer to process. The 'High (2x)' setting is recommended for most documents.

Can I process specific pages only?

Yes! You can choose to process all pages or specify particular pages. Use page numbers or ranges like '1-3, 5, 7-10' to process only the pages you need.

What does the confidence score mean?

The confidence score indicates how certain the OCR engine is about its text recognition. Higher scores (80%+) indicate reliable results. Lower scores may indicate poor image quality, unusual fonts, or complex layouts.

Why is OCR taking so long?

OCR is computationally intensive. Processing time depends on the number of pages, render quality, and your device's performance. Each page must be rendered to an image and then analyzed by the OCR engine.

Is my PDF file secure?

Absolutely! All OCR processing happens locally in your browser using JavaScript. Your PDF file is never uploaded to any server, ensuring complete privacy and security.

What is the maximum file size?

The maximum file size is 100MB. For very large documents, consider processing them in smaller batches by selecting specific page ranges.

My PDF already has selectable text - do I still need OCR?

No. If a page already carries a real text layer (a born-digital or export PDF), OCR would only slow it down and risk introducing recognition errors into text that was already perfect. Use the default 'Auto' mode: it detects existing text layers and extracts them instantly at 100% accuracy, running OCR only on the genuinely scanned pages. The results panel shows how many pages came from the text layer vs OCR. Choose 'Force OCR on all pages' only when you specifically want to re-recognise everything (for example a flattened or corrupted text layer).

Which render quality should I pick for small fonts, fine print or tables?

Higher render quality produces a larger image with more pixels per character, which is exactly what OCR needs for small fonts, footnotes, dense tables and fine print. Use 'High (2x)' for typical documents and 'Best (3x)' for tiny text or detailed tables. Very large pages (A3, posters) are automatically clamped to a safe canvas size so rendering never silently produces a blank image.

How do I handle mixed-language documents and what about handwriting?

OCR works best when the selected language matches the document. For a mixed-language file, pick the dominant language, or split it into page ranges and OCR each section with its matching language, then combine the output. Handwriting - especially cursive - is not reliably recognised by Tesseract; expect clean printed type to score well (80%+ confidence) while handwriting, stamps and low-resolution scans will score low. Use the confidence score and 'Best (3x)' quality to gauge and improve accuracy.