Blur Image

No. Blurring happens entirely inside your browser using the HTML Canvas API's built-in filter support — your image is decoded locally, blurred locally, and re-encoded locally, with no server-side component involved in processing it at any point. This matters most for exactly the kind of content people use a blur tool for: photos containing faces, addresses, license plates, or documents with sensitive information. Since the image data never leaves your device, you can redact sensitive regions with confidence, and you can independently verify this by watching your browser's network tab while using the tool — you won't see the image data being transmitted anywhere.

Full Image blur applies the same blur intensity evenly across the entire photo, which is useful for creating a soft, out-of-focus background effect, an abstract placeholder image, or obscuring an entire scene at once. Area blur leaves the rest of the image perfectly sharp and applies the blur only within a rectangular region you draw and position yourself, which is the right choice whenever you need to redact or obscure one specific detail — a face in a group photo, a license plate, a visible screen, a signature, or a phone number printed on a document — while keeping everything else in the photo fully legible.

The intensity slider sets the blur radius in pixels, from 1 (a light softening that barely changes the image) up to 20 (a strong blur that thoroughly obscures fine detail like text or facial features). Under the hood this value is passed directly to the Canvas API's native blur filter, which applies a Gaussian-style blur — each pixel becomes an average of its surrounding neighborhood, and a larger radius means a wider neighborhood is averaged together, producing a stronger, softer effect. For genuinely obscuring sensitive details like a face or a license plate, use a higher intensity (14 and above); for a gentle background-softening effect, a lower intensity (3–8) usually looks better.

Yes — click and drag from inside the selection box to move it without changing its size, or drag one of its edge or corner handles to resize it. The blurred preview updates live as you adjust the box, so you can fine-tune its exact position and size while watching the effect in real time until it precisely covers the area you want obscured.

A sufficiently strong blur (intensity 15–20) applied over a tightly-fitted selection box will make most faces, plates, and printed text illegible to a casual viewer, but it's worth understanding this isn't a cryptographically irreversible redaction — heavily blurred images have occasionally been partially reconstructed using specialized deblurring or AI upscaling techniques, particularly for large, high-contrast text or a blur radius that's too small relative to the detail being hidden. For genuinely sensitive information where reversal would be a serious problem, use a maximum intensity setting, make sure your selection box fully covers the detail with a comfortable margin, and consider whether a solid black box (which this tool doesn't currently draw, but which you could add afterward in an image editor) would be a safer, fully irreversible alternative.

This is a normal, minor side effect of how canvas-based blurring works at image boundaries: a blur filter averages each pixel with its neighbors, and pixels right at the edge of the image have fewer real neighbors to average with (the canvas beyond the image edge is empty), which can very slightly lighten or soften the extreme border compared to the interior. It's usually only noticeable at high intensity settings and doesn't affect the overall usefulness of the blur — if it's a problem for your use case, using a slightly lower intensity or cropping a thin strip off the final result usually resolves it.

Not meaningfully by design — blurring changes pixel data, not compression settings, so the output file size depends mainly on your image's dimensions and the export format (JPG, PNG, or WebP) rather than the blur intensity you chose. In practice, a heavily blurred image can sometimes compress slightly smaller than the sharp original under lossy formats like JPG, since blurred regions have less fine detail for the compressor to encode, but this is an incidental side effect, not something to rely on. If reducing file size is your actual goal, use the dedicated Image Compressor tool afterward.

You can upload JPG, PNG, or WebP images, and the blurred result downloads in the same format as your original upload — a JPG in produces a JPG out, a PNG in produces a PNG out, preserving transparency for PNG sources. This keeps the tool predictable: you always get back the same format you started with, already blurred, without needing a separate conversion step unless you specifically want to change formats (in which case, use one of the dedicated format converter tools on the blurred result afterward).