CodeKitHubWhat Is This Tool?
This tool lets you redact sensitive regions of an image before sharing it online: faces in a group photo, the number on an ID card or passport, a license plate, a home address printed on an envelope, or a credit card number visible in a screenshot. Draw a rectangle over each region and choose how to cover it — a solid black box, a pixelated mosaic, or a gaussian blur — and the effect is baked permanently into the exported image.
Privacy note: the entire process runs locally using the browser's Canvas API. Your photo is loaded from your device straight into a canvas element and processed there with JavaScript — it is never sent to a server, never uploaded, and never leaves your browser tab. You can even disconnect from the internet after the page loads and the tool keeps working.
Why Use It?
- 100% local processing: your photo never leaves your browser — no upload, no server, no account.
- Three censor modes in one tool: solid black box, pixelated mosaic, or gaussian blur.
- Full-resolution output: censoring is applied at the image's real pixel dimensions, not a shrunk preview.
- Undo and reset: remove the last censored region or start over from the original image at any time.
- One-click PNG download of the finished, permanently redacted image.
How to Use
- Click the upload area or drag in a photo (JPG, PNG or WebP).
- Pick a censor mode: black box, mosaic, or blur.
- Click and drag directly on the image to draw a rectangle over each region you want to hide — a face, an ID number, a plate, and so on.
- Each rectangle is applied immediately when you release the mouse. Use "Undo last" if a rectangle was off, or "Reset" to start over.
- Click "Download PNG" to save the censored image to your device.
Example
Input
A photo with a visible face and a passport number in frameOutput
The same photo with the face pixelated and the passport number covered by a black boxDifferent regions in the same image can use different modes — for example a blurred face and a solid black box over a document number.
Choosing the right censor mode
The three modes trade off differently between security and appearance, so pick based on how sensitive the content is.
| Mode | How it looks | Best for |
|---|---|---|
| Black box | Solid opaque rectangle, fully hides the region | ID numbers, card numbers, addresses — anything that must be completely unreadable |
| Mosaic (pixelated) | Region broken into flat-colored blocks, shape still visible | Faces and general context where you still want the viewer to see a person is there, just not who |
| Blur (gaussian) | Softly blurred region blending with the surroundings | Faces or backgrounds where a smoother, less blocky look is preferred |
Common use cases
- Blurring bystanders' faces before posting a photo publicly on social media.
- Black-boxing the ID or passport number before sending a photo of a document for verification.
- Covering a license plate in a photo shared in a marketplace listing or forum post.
- Redacting a credit card number, order number or personal address visible in a screenshot.
Why redact locally instead of using an online "upload and censor" service
Many online photo tools require uploading the full, unredacted image to a server before you can censor anything on it — which means the exact sensitive content you're trying to protect briefly exists, unprotected, on someone else's infrastructure. This tool avoids that entirely: the image is decoded and processed inside your browser tab using the Canvas API, so the unredacted version never crosses the network. This matters most for exactly the kind of content people censor — faces, ID documents, financial details — where a server-side copy, even a temporary or supposedly deleted one, defeats the purpose of censoring it in the first place.
Frequently Asked Questions
Is my photo uploaded anywhere?
No. The image is loaded from your device with a local blob URL and drawn onto an HTML canvas; all censoring happens with JavaScript running in your browser tab. Nothing is sent to a server at any point, and the tool works even if you go offline after the page has loaded.
Can the black box, mosaic or blur be reversed to recover the hidden content?
The black box and mosaic modes destroy the original pixel data permanently — there is nothing to reverse, since the covered pixels are overwritten with flat color or averaged blocks. A gaussian blur is much harder to reverse than it looks, but for maximum certainty use black box or mosaic on anything highly sensitive, like a full ID number.
Will the downloaded image still contain the original, unredacted version somewhere?
No. The download button exports exactly what is currently drawn on the canvas, which already has your censoring baked into the pixels. There's no hidden original layer embedded in the PNG file, and the file's metadata isn't preserved from the source image.
Can I censor multiple faces or regions in the same photo?
Yes. Draw as many rectangles as you need, one at a time, switching censor mode between them if you like — for example blurring several faces and black-boxing one ID number, all in the same image.
What image formats are supported, and what do I get back?
You can upload JPG, PNG or WebP. The censored result downloads as a PNG, which keeps the redacted regions crisp and avoids re-introducing compression artifacts around your edits.