Blur an image — soft Gaussian blur from subtle to heavy

How to use Image Blur

1. Choose one or more images.
2. Drag Blur radius from 0.5 to 60 px (it starts at 6 px; steps of 0.5).
3. Choose Save as and, for lossy formats, Quality.
4. Press Apply and download the files ending in -blurred.

What it does and when to use it

A blur spreads each pixel’s colour over its neighbours, so sharp lines turn into gentle gradients. Common uses:

• Backgrounds for slides, banners and phone wallpapers, where text must sit on top of a photo and stay readable. A 20–40 px blur keeps the colours but removes the clutter. Add the words with Image Text Overlay.
• Soft-focus and dreamy styles with a small radius.
• Reducing visible noise or moiré in scans of printed pictures, at 1–2 px.

If you want a blocky censor look instead, Image Pixelate does that. To bring back crispness after resizing, the opposite tool is Image Sharpen.

How it works

A true Gaussian blur is slow on large photos, so the tool uses a well-known shortcut: three box-blur passes in a row, horizontally and vertically, sized so that together they match a Gaussian curve whose standard deviation equals your radius. For the default 6 px, the three boxes are 11, 11 and 13 pixels wide.

All of this happens in a Web Worker on your device. Colours are multiplied by their opacity before blurring and divided again afterwards, so semi-transparent edges keep their true colour.

Worked examples

To show the strength of each radius, we blurred a hard black-to-white edge with the tool’s own function and measured how wide the soft transition (from 10% to 90% grey) became:

So at the default setting, a crisp line in your photo becomes a fade about 16 pixels wide. On a 1080 px wide social image, 30 px spreads each edge over roughly 7% of the width, which is why fine text becomes unreadable.

A slide background. A 1920 × 1080 px photo at 30 px keeps its size and colours but loses all detail, ready for a white headline.

Limits and tips

• The whole image is blurred; there is no brush or mask.
• Blurred photos compress very well, so the saved JPEG is often smaller than the original.
• Large radii on very big images take a few seconds, especially on phones.
• EXIF data is not carried into the blurred copy.