Skip to content

Blur or pixelate image

Hide faces, number plates and private details before you share.

Drop an image here

JPG, PNG, WebP, HEIC, SVG and more · or paste with Ctrl+V / ⌘V

How to use Blur or pixelate image

  1. Open an image. Drop a photo, choose a file, or paste a screenshot directly onto the page.
  2. Mark areas to hide. Drag boxes over faces, plates, or text, click + Add area, or tick Whole picture.
  3. Pick Effect and Strength. Choose Pixelate, Blur, or Black box, and adjust Strength from 1 to 12 if needed.
  4. Save the redacted file. Select Save picture to download a flattened image with camera metadata removed.

Your files are processed on this device and never uploaded. They stay private, and there's nothing to delete afterwards.

Can blurred or pixelated areas in a saved image be undone?

Pixelated, blurred, and blacked-out regions overwrite the canvas pixels directly before export, so the saved file contains no reversible layers or hidden original pixels. However, light blurs over short, predictable strings like passwords or account numbers can sometimes be guessed, which is why Black box is safer for sensitive text.

According to NIST SP 800-88 Rev. 1 Guidelines for Media Sanitization, effective sanitization must leave no recoverable data signal rather than relying on weak masking. If you need to annotate a censored screenshot for a bug report, you can label redacted screenshots with callout notes after saving.

Permanent pixel replacement versus reversible layers

Desktop image editors often keep redactions on separate layers that can be toggled off if saved in a project format. This tool flattens every Pixelate, Blur, and Black box region into a single raster grid, removing all layer history.

In Blur mode, the canvas explicitly runs clearRect on each box before drawing the two-pass downsampled blur. That step prevents sharp edges on semi-transparent PNG screenshots from bleeding through the blurred overlay.

Because every pixel inside the rectangle is overwritten in memory before encoding, there is no undo data, alpha mask, or hidden layer stored in the downloaded JPG, PNG, or WebP file.

Why Black box is safer than Blur for short text

Standard fonts have predictable character heights, kerning, and word lengths. Even when averaged into mosaic blocks or smoothed by downsampling, short numbers or API keys can leave statistical clues for brute-force guessing attacks.

Choosing Black box replaces every pixel inside the rectangle with solid #000000, reducing entropy to zero so nothing inside the box can be reconstructed.

Choosing between Pixelate, Blur, and Black box modes

Pick Pixelate to blur faces and number plates in photos privately with an unmistakable mosaic look, Blur for softer background obscuring, and Black box for passwords, emails, or financial IDs. Strength (1 to 12, default 5) tunes Pixelate and Blur, while Black box is always 100% opaque.

Effect mode How it alters pixels Adjustable strength Best use case
Pixelate Averages RGBA values into square blocks 1 to 12 (default 5) Faces, license plates, and street signs
Blur Two-pass downsample and smooth upsample 1 to 12 (default 5) Soft background obscuring and whole photos
Black box Solid #000000 opaque rectangle fill None (always 100% solid) Passwords, API keys, and account numbers
Whole picture Applies chosen effect to full canvas Uses active mode setting Teaser previews or full-frame obscuring

Under the MDN CanvasRenderingContext2D.getImageData() Reference, Pixelate reads raw RGBA buffers inside a Web Worker to pixelate sensitive information in screenshots locally. For multi-page documents with selectable text layers rather than screenshots, use a browser-based PDF redaction tool instead.

Drawing, resizing, and nudging boxes on the stage

You can draw as many redaction boxes as you need directly over the 720 px preview stage, or click + Add area to drop a centered box that you can drag into position. Clicking Clear areas removes all boxes at once if you want to start over.

Using Whole picture and Strength from 1 to 12

Ticking Whole picture hides individual boxes and applies the selected Effect across the entire image dimensions. This is useful when creating blurred background cards or obscuring an entire document preview.

The Strength slider scales block size dynamically based on the image’s shorter side divided by 160. Because it scales proportionally with image dimensions, a Strength of 5 looks consistent on both a small screenshot and a 4K photo.

Drawn rectangles are clamped to a minimum side of 8 px (or 3% of the image’s shorter edge) so tiny accidental clicks do not leave invisible 1-pixel boxes behind on the stage.

Keyboard shortcuts for precise box placement

Every drawn box is keyboard focusable for accurate alignment without fighting mouse drift. Press Arrow keys to nudge a box by 2.5% of the image’s shorter side, or hold Alt with Arrow keys for 0.5% micro-adjustments.

Hold Shift plus Arrow keys to resize the box’s width or height, and press Delete or Backspace to remove the focused box. If you are preparing scanned paperwork, you can also convert redacted document scans into high-contrast monochrome to shrink file size and improve legibility.

Local browser processing and automatic metadata removal

Every pixel transformation runs inside your browser tab, so sensitive screenshots containing banking details, medical records, or private messages never touch a remote server. Images larger than 4096 px on a side or 16,777,216 total pixels are clamped proportionally to fit browser canvas memory limits.

Keeping confidential screenshots off remote servers

Uploading unredacted photos to cloud editors exposes the original file before you even draw a mask. Here, Pixelate runs in an isolated Web Worker on your device, and Blur and Black box render directly on your local 2D canvas.

Tip: Before sharing a photo of a document or vehicle, check both the visible corners and reflected surfaces, and use Black box over any alphanumeric code or QR pattern.

Automatic EXIF and GPS coordinate stripping on export

As explained in the MDN HTMLCanvasElement.toBlob() Reference, exporting the canvas at 0.92 quality produces a brand-new raster blob (-blurred or -redacted) with zero original EXIF or GPS tags. To strip location data from photos you do not need to blur, you can clear hidden GPS coordinates from untouched photos directly.

Questions

Why should I use Black box instead of Blur for passwords?

Light blurs and coarse pixelation leave predictable character shapes and widths that automated pattern-matching tools can sometimes guess. Selecting Black box fills the area with solid black (#000000), destroying every underlying pixel value completely.

How does the Strength slider change pixelation and blur?

Strength ranges from 1 to 12 (default 5) and scales the block size or downsampling factor proportionally to the image's shortest side. Higher Strength values produce larger mosaic tiles in Pixelate mode and heavier smoothing in Blur mode.

Does blurring an image remove its camera and location data?

Yes. Exporting the canvas creates a fresh raster file without the original image's camera model, timestamp, or GPS location tags, protecting both visual details and embedded metadata.

How do I censor multiple areas on a single photo?

Drag a box over the first face, plate, or detail, then draw additional boxes or click + Add area for each extra region. You can drag any box to reposition it, resize it from its bottom-right handle, or delete it individually.

Can I paste a screenshot directly into the blur tool?

Yes. Copy any screenshot or image to your clipboard and press Ctrl+V (or Cmd+V on Mac) to load it onto the redaction stage without saving a temporary file first.

How do I resize or delete a redaction box with the keyboard?

Tab to focus any box, then press Arrow keys to move it (hold Alt for fine nudges), Shift plus Arrow keys to resize width or height, and Delete or Backspace to remove it.

Updated