Sharpen Image vs Image Resizer
Choose this if
the image is larger than it needs to be, which is what caused the softness in the first place
How they compare
| Criterion | Sharpen Image | Image Resizer |
|---|---|---|
| What it changes | Local contrast at edges. Pixel count is untouched | Pixel count. Local contrast is untouched |
| Can it recover lost detail | No. It amplifies what remains and invents nothing | No. Enlarging interpolates between existing pixels |
| Effect on noise | Amplifies it, worst in flat areas such as skies | Reduces it when downscaling, because noise averages outBetter here |
| Effect on file size | Usually increases it: more local detail compresses less well | Reduces it substantially. Halving both dimensions quarters the pixelsBetter here |
| Risk of overdoing it | High and visible: halos appear along every edge | Low. The result is simply smallerBetter here |
Which is best for you
Sharpen last, lightly, at the final size
Downscaling always softens an image slightly, because detail is being averaged away. A small amount of sharpening afterwards restores the apparent crispness. Keep it below about 100%: past that, the light side of every edge overshoots into a visible halo, and any sensor noise in flat areas is amplified along with the detail you wanted.
Resize first, and properly
Most softness in a browser-resized image comes from how it was resized rather than from the resizing itself. A browser asked to shrink an image by more than half in one step switches to a cheap filter that discards pixels. Stepping down in halves keeps every pixel contributing, and the result needs far less sharpening afterwards, sometimes none.
The recommendation
Resize first, then sharpen, never the other way round. Sharpening before a downscale means the artefacts you introduced get scaled along with everything else, and the softening undoes the work anyway. Try the resize on its own first before assuming you need to sharpen at all: a properly stepped-down image is much closer to sharp than one thrown down in a single draw, and every bit of sharpening you avoid is noise you have not amplified.
Frequently asked questions
My photo is blurry. Will sharpening fix it?
Not if it is genuinely blurred. Motion blur and missed focus destroy information, and sharpening only amplifies local contrast in what survives. It makes a blurry photo look like a blurry photo with harder edges. It works well on the slight softness left by downscaling and on lens softness, which are different problems.
Why does my resized image look worse than the original?
Usually because it was scaled down in one large step. Below about half size browsers switch to a fast filter that throws pixels away rather than averaging them. Resizing in halving steps preserves far more detail, which is what the resizer here does.