Can this fix a blurry photo?

It can make a slightly soft photo look considerably better. It cannot recover detail that was never captured — motion blur and genuine focus misses are not recoverable by any sharpening filter.

What is the threshold for?

It sets how different two areas must be before sharpening applies, so smooth regions like sky and skin are left alone. It is what stops the filter making noise crisper.

What settings should I start with?

Amount around 80%, radius around 1 pixel, threshold around 4. Adjust the threshold upwards if smooth areas start to look grainy.

Is my photo uploaded?

No. The sharpening runs in your browser using the canvas API, and the image is never sent anywhere.

Image Sharpener

Unsharp masking with a threshold control — so an out-of-focus photo gets crisper without the grain in the sky getting crisper too.

Runs in your browser — nothing is uploaded

Your image

Choose an image

Amount %

Radius px

Threshold

Your file never leaves this device. The whole operation runs in your browser — nothing is uploaded, queued on a server, or deleted later, because nothing was ever sent.

Threshold is what stops it sharpening noise

Sharpening increases local contrast, and it cannot tell an edge from grain — so on a high-ISO photo it makes the noise crisper too. The threshold sets how different two neighbouring pixels must be before the effect applies, which leaves smooth areas like sky and skin alone. It is the control most sharpeners do not give you.

Before and after

Choose an image to begin.

Why use it

Built to be genuinely useful

A threshold control

Sets how strong an edge must be before it is sharpened, which is what keeps noise out of smooth areas.

Drag to compare

The original and the sharpened version at the same scale, split by a slider you control.

Amount and radius separately

Radius decides the size of detail affected; amount decides how hard. They are not the same dial.

Nothing is uploaded

The processing runs on your device using the canvas, so the photo never leaves the browser.

How it works

Three steps

1
Choose your image.
2
Raise the amount until edges look defined, then the radius for larger detail.
3
Raise the threshold until smooth areas stop looking grainy, then download.

What sharpening actually does

Sharpening does not recover detail. It cannot — the detail is either in the file or it is not, and no filter can invent what the lens did not resolve. What it does is increase contrast along the edges that are already there, which the eye reads as sharper.

The classic technique is the unsharp mask, named after the darkroom process it copies. A blurred copy is subtracted from the original to isolate the edges, and that difference is added back amplified. Radius sets how blurred that copy is, which decides the size of the detail affected; amount sets how much of the difference is added back.

Because it is contrast amplification, it cannot distinguish an edge you want from grain you do not. That is not a flaw in a particular implementation — it is what the operation is — and it is why the threshold matters so much.

Why the threshold is the control that matters

A threshold sets a minimum: two neighbouring areas must differ by at least this much before the sharpening applies. A hard edge between a building and the sky easily clears it. The faint random variation across the sky itself does not, so it is left exactly as it was.

Without one, sharpening a photo taken in low light makes the noise sharper along with everything else, and the result looks worse than the blurry original — gritty rather than soft. This is the most common way sharpening is misused, and most online sharpeners give you no way to avoid it because they expose only a single strength slider.

A practical approach: set amount to around 80%, radius around 1, then raise the threshold until the smooth areas stop looking grainy. Somewhere between 3 and 10 works for most photographs; scanned documents can go higher.

Knowing when to stop

Over-sharpening announces itself as a pale outline on the dark side of an edge and a dark one on the light side — a halo. Once you have seen it you cannot stop seeing it, and it is far more obvious in print than on a screen.

Judge at 100%, not zoomed out. An image that looks perfect fitted to the window is frequently over-sharpened, because the browser's own downscaling hides the halos.

And sharpen last. If you are also going to resize, crop or compress, do those first — sharpening before a resize amplifies artefacts that the resize then scales up.

FAQ

Questions people ask

Related tools

More image tools

Available tools are linked; the rest are in development.

Background Remover Image Upscaler Image Enhancer Face Blur Image Background Changer AI Writer AI Summarizer AI Paraphraser AI Humanizer AI Grammar Checker AI Rewriter AI Translator

All Image Tools

Need a tool like this for your business?

We build internal tools, dashboards and automation that fit how your team works.

See Our Services