Redact Image
Black out names, numbers and faces with solid boxes or a brush. Runs in your browser — the image is never uploaded.
JPG, PNG, WebP, GIF or BMP — processed entirely on your device
—
Drag to draw a box over anything that must not be readable.
How to Redact an Image
- Drop in a photo or screenshot, or paste one from the clipboard.
- Choose Box to drag rectangles, or Brush to paint freehand.
- Cover everything that must not be readable; use Undo if you miss.
- Pick an output format and click Download Redacted Image.
- Check the downloaded file before you send it.
Your image is never uploaded. The redaction happens on your device.
Four ways a redaction fails
Covering something up looks finished long before it is. These are the ways it goes wrong in practice, and what this tool does about each.
A translucent highlighter
Marking over a card number with a dark highlighter looks solid on screen. Turn up the brightness or adjust the levels and the digits come straight back, because a semi-transparent stroke only darkens what is underneath.
Here: Every fill here is fully opaque. There is no opacity setting to get wrong.
Blur or pixelation on text
A pixelated block still encodes the rough shape of each character. Recovering pixelated text from screenshots has been demonstrated publicly — the attacker renders every candidate string, pixelates it the same way, and keeps the one that matches.
Here: Boxes carry no information about what they cover: every pixel inside is the same colour.
An editor that keeps the original
In 2023 the Pixel Markup tool and the Windows Snipping Tool were both found to leave the original image's data at the end of edited files — cropped or redacted screenshots could be partly restored.
Here: The output is encoded fresh from a canvas. It contains the pixels you see and nothing after them.
Metadata nobody looked at
A photo can carry GPS coordinates, the camera serial, the time it was taken, and an embedded thumbnail of the unredacted original.
Here: A canvas holds pixels only, so all of it is gone from the download.
Redact or blur — which one?
| Redact (this tool) | Blur | |
|---|---|---|
| Covered area holds | Nothing — one solid colour | A smeared version of the original |
| Text underneath | Unrecoverable | Sometimes guessable, especially short strings |
| Looks | Deliberate and obvious | Natural, unobtrusive |
| Best for | Card and account numbers, names in documents, addresses, passwords, anything legal or financial | Faces in a crowd, a background, a number plate in a holiday photo |
The rule of thumb: if someone recovering it would cost money or safety, redact it. If you just do not want it to be the focus, blur is kinder to the picture.
Before you send it
- Open the downloaded file, not the editor, and look again. You are checking the file other people will receive.
- Look for the same information elsewhere in the picture — a name in a heading and again in an email signature, a card number on the front and its last four digits in a receipt.
- Reflections and the edge of the frame give things away: a monitor reflected in glasses, a letter half out of shot.
- If the original file was a PDF, redacting a screenshot of it does not redact the PDF. Send the image, not the document.
Specifications
- Accepts
- Any image
- Gives you
- <name>-redacted .png, .jpg, .webp download
- Where it runs
- Your browser — the file is never uploaded
- Sign-up
- None
- Cost
- Free, with no usage limits
FAQ
Is my image uploaded to a server?
No. The redaction is drawn on a canvas in your browser and the new file is handed straight to your downloads. Nothing is uploaded — which matters more here than anywhere, since the whole point is that the thing you are hiding never reaches anyone.
Can a redaction from this tool be undone?
Not from the file you download. The boxes are painted into the pixels and the image is encoded again from scratch, so the covered area holds only the fill colour — there is no layer to remove and no copy of the original underneath. Undo works while you are editing, because the original is still open in your browser; it is not part of the output.
Why not just blur it?
Because blur and pixelation keep some of the information. For text especially, a pixelated block still carries the rough shape of each character, and tools exist that recover pixelated text from screenshots by brute force. A solid box carries nothing. Use the Blur tool when you want the photo to still look natural, and this one when the content must be unreadable.
Does it remove the photo's location data?
Yes, as a side effect. A canvas holds pixels and nothing else, so GPS coordinates, the camera model, timestamps and the embedded thumbnail some phones add are all gone from the output. That embedded thumbnail is worth knowing about: it is a small copy of the original photo, and editors that only change the main image have been known to leave it intact.
Can I redact on my phone?
Yes. Drawing uses pointer events, so a finger works the same as a mouse. For small text, zoom the page first and use the box tool, which is easier to place precisely than a stroke.
Which output format should I pick?
PNG for screenshots — it keeps text edges crisp and the fill exactly solid. JPEG for photos, where it gives a much smaller file. Either way the redacted area is fully covered; JPEG's compression cannot bring back what was painted over.