How to redact a scanned PDF (without uploading it anywhere)
Scanned PDFs are the documents people most often need to redact — court records, medical forms, IDs, signed contracts — and they're also the easiest to redact wrong. This guide explains the two traps and shows how to do it properly, free, without your document ever leaving your computer.
Why scanned PDFs are different
A scanned PDF is a photograph of a page. There's no text layer, so you can't search for a name and most "redaction" tools can't find anything to redact. Worse, tools that just draw a black rectangle on top of the image leave the original picture underneath — anyone who opens the file in an editor can remove the box and read what you hid. Real redaction has to destroy the pixels, not cover them.
The second trap: uploading sensitive scans
Most online redaction sites work by uploading your file to their server. For exactly the documents you'd want to redact — the ones full of Social Security numbers and account details — that's the worst possible move. The FBI has warned about malicious free file-conversion sites, and even honest services are a breach away from leaking your document. The safest upload is the one that never happens.
Redact a scanned PDF in four steps
- Open redactoronline.com and drop your scanned PDF onto the page. It loads in your browser only — nothing is uploaded, and the app keeps working with your internet disconnected.
- Click “Scan (OCR)”. Text recognition runs on your device (the OCR engine ships with the site), turning the scanned image into searchable text without sending it anywhere.
- Click “Auto-detect PII” to automatically find and mark Social Security numbers, emails, phone numbers, and credit-card numbers — or search for specific names, or left-drag across any text (the box snaps to the line height automatically).
- Hit “Redact & download”. Every page is rebuilt with the marked areas destroyed at the pixel level, and all document metadata is stripped. What's under the black box is gone, not hidden.
How do I know it's really redacted? Open the downloaded file and try select-all + copy: there is no hidden text to select. The output contains only flattened page images with the sensitive regions destroyed.
Common questions
Is it really free? Yes — your first pages are free, and heavy users can buy a one-time lifetime license. No subscription, no account.
Does OCR upload my file? No. The OCR engine (Tesseract) and its language model are served from this site and run inside your browser. You can verify in your browser's network tab: no file data leaves the page.
What about handwriting? OCR is built for printed text. For handwriting, use the manual tools: left-drag snaps to the line, right-drag draws a box of any size — both destroy what they cover.
Will the output be searchable? The redacted file is intentionally image-based — that's what makes the redaction bulletproof. Keep your original for your own records.