Image Pixelate
Pixelates an image entirely in your browser using two canvas passes: it first scales the picture down so each block of the original becomes a single pixel (the browser's smoothing averages the colours inside the block), then scales it back up with nearest-neighbour interpolation so every block comes out as a crisp, hard-edged square. The preview re-renders live as you drag the block-size slider from 2 to 64 px, and nothing is ever uploaded — the file stays in memory on your device. Typical uses: masking faces, usernames, or license plates in a screenshot before sharing it, adding a mosaic effect, or turning a photo into a colour-block reference for pixel art. Because the output is re-encoded from a clean canvas, EXIF metadata such as GPS location is not carried into the download. For genuinely sensitive text, prefer a large block size — or better, an opaque rectangle — since fine pixelation of short strings can sometimes be reconstructed.
Pixelation alone is not a security boundary for sensitive text — for that, paint over with a solid colour.
How to use
- Drop an image onto the target or click to pick one — it stays on your device.
- Drag the block-size slider (2–64 px); the preview re-renders automatically as you move it.
- Choose an output format: PNG is lossless and keeps transparency, while JPEG and WebP reveal a quality slider and usually produce smaller files.
- Compare original and result side by side, then hit Download; Clear starts over with a new image.
Frequently asked questions
- What if I just want to redact a small area?
- Crop to that area first in another tool, pixelate the crop here, then paste it back over the original. This tool applies the effect to the whole image at once, which is what you want for a full mosaic or retro look.
- Is pixelation reversible?
- No. Each block stores a single averaged colour — the per-pixel detail is discarded and can't be recovered from the file. That said, don't treat weak pixelation as a security boundary: with small blocks over short, known-font text, attackers have reconstructed originals by brute-forcing candidate strings.
- Does the download keep EXIF metadata like GPS location?
- No. The result is freshly encoded from a canvas, which carries pixels only — EXIF tags, GPS coordinates, and camera info from the original file are not copied into the output. (The browser applies EXIF orientation while decoding, so the image still comes out the right way up.)
- What block size should I use?
- For a retro pixel-art look, 8–16 px usually reads well. For censoring, the block must be large relative to the detail you're hiding — for on-screen text, use blocks at least as tall as the characters, and zoom into the preview to confirm nothing is guessable. When in doubt, go bigger or paint over with a solid colour.
- Is there a file size or resolution limit?
- The tool sets no limit of its own, but the whole image is processed on an in-memory canvas, so extremely large photos can hit the browser's canvas size ceiling — mobile Safari is the strictest. If processing fails on a huge image, scale it down first and pixelate the smaller copy.
Related tools
Image → ASCII Art
Convert any image to ASCII art with adjustable width, character set, and inversion — all in your browser.
Image Sharpen (Unsharp Mask)
Sharpen any photo or screenshot in your browser using the unsharp-mask algorithm — three sliders (amount, radius, threshold), four presets (subtle → extreme), live before/after comparison, and PNG download. Nothing is uploaded.
QR Code Decoder
Drop a screenshot or photo of a QR code and read what's inside — entirely in your browser, no upload.
DPI / Print Size Calculator
Pixels ↔ physical print size at any DPI. Common print presets (72 web, 150 draft, 300 photo, 600 fine art), megapixels, aspect ratio, and file size estimates.
Image Posterize
Reduce a photo to N levels per channel for that flat poster-art look, with optional Floyd-Steinberg dithering and per-channel level control.
Lorem Picsum URL Builder
Build placeholder image URLs from picsum.photos — set dimensions, grayscale, blur, a seed for stable images, or a specific image ID.