Sharpen Image

What is Sharpen Image?

Sharpening does not actually add information to a photo, it makes the information that is already there more visible. The technique, called unsharp masking, was originally a darkroom process: a blurred negative was sandwiched with the original to create an outline halo that increased the apparent crispness of edges. Digital sharpening does the same thing mathematically, the algorithm identifies edges using a convolution kernel, then locally boosts contrast across them. This tool uses a 3x3 sharpening kernel with adjustable strength. At low settings (10-30%) it recovers crispness lost to JPG compression or a slightly soft lens. At medium settings (40-60%) it gives a deliberately crisp, web-ready look. At high settings (70%+) it starts producing visible halo artefacts, useful for stylised graphic work but distracting for natural photography.

Why use this tool?

Processing is purely client-side via Canvas convolution. The 3x3 unsharp kernel runs in a single pass, so even a 20-megapixel photo sharpens in roughly a second on a mid-range phone. The strength slider gives you a live preview, you can dial in the right amount without re-uploading. Output preserves the input's format, dimensions and any embedded metadata. No file ever uploads, no account is required, and there is no watermark on the result.

Common use cases

The most common case is a photo that looks fine on the camera back but soft on a larger screen. Phone cameras compute aggressive in-sensor sharpening at small preview sizes, when you scale up the same image to a laptop monitor the softness becomes visible. A light sharpening pass restores the crispness you remember seeing. Other workflows: preparing images for print, where the ink-spread on paper rounds off edge contrast, so a moderate pre-print sharpening pass compensates; recovering detail from a slightly out-of-focus macro shot; emphasising fine texture in product photography for an e-commerce listing. Note the limit, sharpening cannot rescue a heavily blurred photo, the information simply is not there. For true detail recovery on low-resolution images, use our Upscale Image tool, which uses a Real-ESRGAN neural network to reconstruct plausible texture at 4x.

How to use Sharpen Image

1. 1Drop a JPG, PNG, WebP or BMP image into the upload area
2. 2Drag the strength slider to the desired sharpening level
3. 3Watch the live preview, look for crispness without halo artefacts
4. 4Download the sharpened image in the same format as the original

Frequently asked questions

Can sharpening fix a blurry photo?

Only slightly soft photos. Sharpening works by enhancing the edges that are already in the image, if the image is heavily blurred those edges are gone and there is nothing to enhance. For low-resolution images that need genuine detail reconstruction, our Upscale Image tool uses a Real-ESRGAN AI model to generate new realistic texture at 4x the original resolution.

How do I know I have applied too much sharpening?

Look at edges between dark and light areas, such as text on a white background or a person's hair against the sky. If you see a bright halo on the light side of the edge and a dark halo on the dark side, sharpening has gone too far. Dial back until the halos disappear.

Does sharpening reduce file size?

Slightly, in JPG and WebP, because increased local contrast often compresses more efficiently. The change is usually 5% or less. In PNG, lossless compression typically produces a marginally larger file because added detail breaks runs of identical pixels.

Should I sharpen before or after resizing?

After. Resizing softens edges, so sharpening after resize compensates exactly for the loss. If you sharpen first and then resize, the resize step undoes part of the sharpening. The order is: crop, then resize (Resize Image), then sharpen, then export.

Will sharpening introduce noise?

It amplifies noise that is already in the image, because noise looks like high-frequency detail to the sharpening kernel. If the source is a low-light photo with visible grain, apply a light noise reduction first (our Adjust Colors tool offers desaturation that helps) or use a low sharpening strength of 15-25%.

Does it work on screenshots and graphics?

Yes, and screenshots often benefit the most. Anti-aliased text and UI edges become crisper after a 20-30% sharpening pass, which is helpful when reusing screenshots in documentation or social media.

1. 1For web use, 25-40% strength is usually right. For print use, push to 50-70% because ink spreads on paper and absorbs some of the apparent sharpness.
2. 2If you intend to compress a JPG after sharpening (our Compress Image tool), sharpen slightly more than feels right, because JPG compression softens edges again. The two operations balance out.
3. 3Sharpen a copy, not the original. Sharpening is destructive, the moment you save the result, the unsharpened version is the only fallback if you go too far.

How does it compare?

Photoshop's Unsharp Mask filter does the same operation with more controls (radius and threshold). This tool exposes the most-used control - strength - and runs in your browser with no install and no subscription.

Related tools

More free tools. No signup, no watermarks.