Blur Image

Blurring softens a whole picture: a background for text or slides, a dreamy effect, a deliberately unreadable preview, or the illusion of speed. This tool applies a Gaussian, box or directional motion blur to one image or a batch, keeps the edges clean and preserves transparency. The processing runs in your browser rather than relying on browser filters, so results look the same everywhere.

How to use this tool

1. Choose one or more images.
2. Pick Gaussian for a soft natural blur, Box for an even average, or Motion for streaks in one direction.
3. Set the radius in pixels (and the direction for motion blur). Bigger photos need bigger radii for the same look.
4. Download the blurred images, or all of them as a ZIP.

Formula

Gaussian weight at distance d: e^(−d² ÷ 2σ²); approximated by three box blurs whose widths add up to the same variance

Three successive box blurs give a curve within a few percent of a true Gaussian at a constant cost per pixel, whatever the radius. The odd box widths wᵢ are chosen so that Σ (wᵢ² − 1) ÷ 12 ≈ σ². Pixels beyond the border repeat the edge pixel, so edges do not darken, and colours are weighted by alpha so transparent areas do not create grey halos. Motion blur averages 2r + 1 pixels along the chosen direction.

Example

A 1920 × 1080 px wallpaper blurred with Gaussian σ = 20 px becomes a smooth background on which white text is easy to read. The tool uses three box passes of width 39, 39 and 41 px, whose combined variance 2 × (39² − 1) ÷ 12 + (41² − 1) ÷ 12 ≈ 393 is within 2% of 20² = 400.