Sharpen a Blurry Image — Free, No Signup
Sharpen a soft image with a strength control you set yourself. Worth being upfront: this recovers definition from photos that are slightly soft. It does not unblur a badly blurred photo, and nothing else does either.
How to do it
- Drop the soft or blurry photo onto the upload box.
- Choose sharpen, then raise the strength gradually.
- Stop before halos appear around edges — that is the sign you have gone too far. Motion blur and badly out-of-focus shots cannot be fully recovered.
- Download the sharpened image.
What sharpening actually does
Sharpening finds edges — places where brightness changes quickly — and exaggerates the change by darkening the dark side and brightening the light side. Your visual system reads that increased local contrast as crispness. Nothing is added to the image; existing structure is emphasised. That is why it works so well on a photo that is nearly right and fails completely on one that is genuinely blurred. If the edge was never recorded, there is no contrast to exaggerate, and the algorithm has nothing to work with.
Which blur can be helped
Slight softness from a marginally missed focus, from diffraction at a small aperture, from shooting through glass, or from an aggressive downscale responds well — the detail is present but low in contrast. Heavy motion blur does not, because the information has been smeared across many pixels and averaged together, and averaging is not reversible. Out-of-focus blur beyond a small amount does not either. Neither does the softness in a heavily compressed thumbnail, where the detail was discarded by the encoder rather than the lens.
Halos are the sign you went too far
Push the strength up and bright outlines appear along high-contrast boundaries — a white fringe against a dark roofline, a glow around text. That is the algorithm exaggerating an edge past the point of plausibility, and once it is visible the image looks processed rather than sharp. Sharpening also amplifies noise, since noise is high-frequency detail and the filter cannot tell it apart from real texture. On a grainy photo, denoise first with the enhancer, then sharpen, or you will end up with very crisp grain.
How to use the strength control
Start at the default of 1.5 and judge the result at full size rather than in a thumbnail, because sharpening artefacts are invisible when scaled down and obvious at 100 percent. If the photo is destined for the web at small dimensions you can push harder than you would for print, since the viewer will never see it large. If you cannot find a setting that helps without introducing halos, the source is blurred rather than soft, and that is a shooting problem no filter fixes.
Limits
Input up to 5 MB with the long edge capped at 2000 pixels, which keeps the filter fast in a browser tab. Output downloads without a watermark. Everything runs on your own device, so nothing is uploaded — useful when the blurry photo happens to be a document or something private you would rather not hand to a website.
Frequently asked questions
Can this unblur a photo?
It sharpens a slightly soft photo by boosting edge contrast. It cannot reverse heavy motion blur or genuine out-of-focus blur, because that information was averaged away rather than merely dimmed.
Why does my sharpened photo look worse?
Either the strength is too high, which produces bright halos along edges, or the image is noisy and the filter has sharpened the noise along with the detail.
Should I denoise before sharpening?
Yes, on any grainy photo. Sharpening cannot distinguish noise from texture, so sharpening first gives you very crisp grain.
What strength should I use?
Start at 1.5 and judge at full size, not in a thumbnail. Web-bound images tolerate more than print-bound ones because nobody sees them large.
Will it help a blurry photo from a security camera?
Almost certainly not. That kind of footage is usually both low-resolution and motion-blurred, which is the combination sharpening cannot address.
What is the maximum image size?
5 MB, with the long edge scaled to 2000 pixels before processing to keep the filter responsive in a browser tab.