Skip to content

🟦 Image Mosaic / Blur

Drag to select any region and apply mosaic or blur. Useful for hiding faces, license plates, and personal info. Processed in-browser via Canvas API.

100% Free No signup Browser-only 5 languages Dark mode

🔒 About Privacy

📂

Drag & drop, click to select, or paste an image

📖 Where people get stuck

Select any region by dragging and apply a mosaic, a blur or a solid black fill, with rectangular, elliptical or freehand selection. Processing uses the Canvas API entirely in the browser. If the purpose is to hide something, which effect you choose is the decisive question: a mosaic or a blur weakens the information without destroying it. There are documented cases of recovery, so do not use them for anything that genuinely must not be seen.

Case What happens What to do
Mosaic and blur can be reversed A blur averages neighbouring pixels, and averaging does not fully destroy information — something close to the inverse operation can recover an approximation of the original. A mosaic is worse still: where the set of possible strings behind it is small — a numeric licence plate, a product code in a fixed format, a card number — you can simply mosaic every candidate and compare. That is brute force, and it needs no sophisticated technique. A face obscured with a swirl distortion has in fact been reconstructed and used in a criminal investigation. Use the solid black fill for anything that must genuinely be hidden. Filling replaces the pixels in that region with a single colour, so the original information is gone completely — that is the only approach that is safe in principle. Understand a mosaic or a blur as a presentational effect for making something less noticeable, not a feature for making it unreadable. As a rule of thumb: a mosaic is fine when the goal is consideration, such as blurring passers-by in a published photograph; for an address, a name or a number — anything where disclosure is a problem — the black fill is the only option. When in doubt, fill it black: this is not a place to trade safety against appearance.
The masked area is not enough to prevent identification Hiding only the face still leaves hairstyle, build, clothing, accessories, the buildings behind, a reflection in a mirror or window, a lanyard badge, or the document in someone hand. In a photograph of a document, removing the name leaves the address, phone number, reference number, an official seal and the form layout itself as clues. A single missed spot is fatal, and it is exactly what you are least likely to notice yourself — information you already know does not register visually, so you overlook what is obvious to anyone else. Before you start, put into words what you are protecting against. Not wanting an acquaintance to recognise someone and not wanting the general public to identify them call for entirely different coverage. For the former, the face alone is not enough; for the latter, retaking the photograph or cropping down to only what is needed is more reliable (image crop tool). Always have someone else look at it before you publish — simply asking can you tell who this is surfaces clues you could not see. Reviewing it yourself after a gap also helps, since fading memory brings you closer to a stranger eye.
The unedited original survives somewhere Even when the editing itself is correct, the original can leak because you attached it alongside, an unedited screenshot is still sitting in a chat, or a cloud backup captured the original. Harder to anticipate is the behaviour of the editing application itself: the Windows Snip and Sketch tool and the Pixel screenshot editor both had a defect that left the original image data at the end of a file you had cropped — known as aCropalypse. The assumption that saving over a file removes what was there does not always hold. Check the output filename (_masked) and hand over only that. This tool never overwrites your original and always saves under a new name, so that particular accident cannot happen here. As a final check before publishing, open the file you are about to send and look at it — the file you meant to edit and the file you actually attached being different is a startlingly common mistake. If the file size looks oddly large, suspect leftover data. For extra certainty, take a fresh screenshot of the edited image: by construction, nothing beyond what is on screen can be in it.

Going through a canvas strips all EXIF from the output — losing GPS coordinates and the capture timestamp is a welcome side effect before publishing. Other ways of hiding things have traps, though. Covering something with a black rectangle in a PDF, a Word file or a slide deck is entirely ineffective: the shape can be moved, and if it is text underneath it can be selected and copied. This has caused real disclosures again and again, including repeatedly in documents published by public bodies. To remove content from a document, export it to an image first and edit the image — it takes longer, and it is the only reliable procedure. One more thing: loading an animated GIF or WebP processes only the first frame, so this tool cannot help you hide something inside a video or an animation.

📖 How to Use

  1. 1
    Load an image
    Drag and drop, click to select, or paste from clipboard with Ctrl+V.
  2. 2
    Choose effect and strength
    Choose mosaic / blur / black fill. Use the slider to adjust block size.
  3. 3
    Drag a region to apply
    Drag on the image. The effect is applied to that region as soon as you release. Drag multiple times for multiple regions.
  4. 4
    Download
    Download as PNG or JPEG. Filename gets _masked suffix.

❓ Frequently Asked Questions

Is my image uploaded to your server?
No. All processing happens entirely in your browser via the Canvas API. Image data never leaves your device. You can verify this in the Network tab of your developer tools.
Can the mosaic be reversed?
Mosaic and blur lose information, but with low strength they can be partially recovered by AI. For sensitive info, use Black fill or strong mosaic (block size 30+).
What image formats are supported?
All formats your browser can display (PNG, JPEG, WebP, GIF, BMP, SVG, AVIF, HEIC, etc.). Output as PNG or JPEG.
🐛 Found a bug or issue with this tool?

Free to use, no signup. Even just the steps to reproduce are helpful. Reports go directly to the operator and help us fix issues.

* Browser info (UA / screen / language / URL) is sent automatically to help reproduce the issue