Free image redaction tool

Redact an image online for free.

Open a photo, scan, or screenshot in the private editor. Review suggested text and faces, add solid boxes over any missing details, and download a new PNG without the source image metadata.

Choose a PNG, JPEG, or WebP up to 25 MB. Review every suggestion and draw any missing boxes, then download a new PNG. Automatic detection can miss details.

Start redacting an image Try fictional sample files

Worked image example

Keep the error message. Cover the customer details.

This invented support ticket was reviewed in the real private editor. The email was suggested automatically. The full name and demo token were found with exact-text search after the first review; the token was missed by automatic detection.

Fictional support ticket before redaction, containing Riley Demo, a test email, a demo token, and an upload timeout error — Before: all names and identifiers are invented.

The same fictional image after solid redactions cover the name, email, and token while preserving the timeout error and code — After: three solid boxes become part of a new PNG.

Download the fictional original to try it yourself, then inspect the worked PNG at full size. Check every edge and the rest of the image before sharing your own result.

How this example was checked

The live editor completed the export and showed a file fingerprint. The published worked PNG was generated with the same unchanged image-export function and the reviewed pixel coordinates, using a local canvas encoder. It is not a screenshot or an identical copy of the browser download.

View the real editor export receipt · Inspect the example provenance and output hash

For a fuller review, follow the support screenshot guide or check the product screenshot privacy checklist. These checks also apply to scans and photos with labels or screens in the background.

What this tool does

Images can reveal identity in more places than the obvious face

A photo can expose a home address on a package, a license plate in the background, a school name on clothing, a patient label, a computer screen, or a handwritten signature. Scanned images can contain the same names, account numbers, and signatures as a PDF while lacking a clean text layer.

FreeRedact renders the image in your browser and offers suggestions for likely text and faces. Treat those marks as a starting point. Pan across the full image, inspect reflections and background objects, and draw manual redaction boxes anywhere context or image quality makes automatic detection uncertain.

Image uploads are exported as new PNG files. The approved solid redactions become part of the output pixels, and metadata from the source PNG, JPEG, or WebP is not copied into the new file.

Files and output

Supported formats and results

PNG, JPEG, and WebP input

Open a supported image up to 25 MB, including a photo, scan, screenshot, or exported document page.

Text and face suggestions

Review likely private text, faces, and handwriting regions directly on the image.

Manual redaction

Draw a precise solid box over a license plate, signature, tattoo, label, screen, or any other visual clue.

Fresh PNG output

Download a new PNG with redactions applied and source image metadata left behind.

How to use FreeRedact

A careful three-step workflow

Choose the image

Open a PNG, JPEG, or WebP from your device. The editor decodes and displays it locally.

Review text and visual clues

Check suggested text and faces, then inspect objects, labels, screens, backgrounds, and reflections for additional clues.

Export a clean PNG

Apply black or white redactions, download the new file, and inspect it at full resolution.

Automatic suggestions

What image detection can—and cannot—recognize

Text recognition and visual models work best on clear, upright, high-contrast images. Real photos often contain perspective, glare, shadows, and clutter.

Details FreeRedact looks for

Email addresses, common U.S. phone formats, account IDs, payment details, and U.S. SSN-style numbers
Person names, U.S.-style street addresses, IPv4 addresses, private-looking URLs, and access tokens
Faces and regions that may contain hard-to-read handwriting
Custom text repeated across labels, forms, badges, signs, and screens

Visual clues to check manually

License plates, house numbers, tattoos, uniforms, badges, and location landmarks are context-dependent
Text at an angle, behind glare, on a curved surface, or at low resolution may be missed
Posters, illustrations, statues, and logos can create false face suggestions
Reflections, mirrors, windows, screens, and photo backgrounds can reveal additional people or information

Before you share

Verify a redacted image at full resolution

Small previews can hide gaps around a redaction or details in the background.

Open the downloaded PNG at 100% size and zoom in around every redaction edge.
Inspect faces, name tags, documents, mail, screens, plates, mirrors, and reflections.
Confirm solid boxes cover the complete detail rather than only its center.
Check the new file's properties if the source image may have contained location or camera metadata.
Ask a second person to scan the final image when disclosure would carry meaningful risk.
Rename the downloaded copy if its filename includes a person's name, case number, health condition, customer, project, or other private context.

Questions and answers

Frequently asked questions

Which image formats can I redact?

FreeRedact accepts PNG, JPEG, and WebP images up to 25 MB. The redacted download is a new PNG.

Can FreeRedact automatically find faces?

It can suggest likely faces, but visual detection can miss faces or mistake illustrations and objects for faces. Review the full image and add or reject boxes as needed.

Can it read handwriting in a scanned image?

FreeRedact can flag regions that may contain handwriting, but handwriting recognition is less reliable than clear printed text. Inspect signatures, filled fields, and handwritten notes manually.

Will the exported image keep EXIF location data?

No source EXIF data is copied. FreeRedact creates a new PNG from the visible image and approved redactions.

Related tools and guides

Ready to review the file?

Open the private editor to apply these checks. No signup is required, and the file stays on your device.

Open private editor