Sharpness Analyzer

What it does

Scores how sharp or in-focus a photo is, using a standard edge-detection technique (Laplacian variance): the more fine detail and edge contrast the image has, the higher the score. A photo that's out of focus, motion-blurred, or heavily softened by compression will score low.

What you'll need

• A photo (JPEG, PNG, WebP, etc.) from your device.

Steps

1. Tap "Upload a photo" and choose an image.
2. The score, verdict and raw variance number appear automatically — there's no separate "analyze" step.
3. Upload a different photo any time to compare.

Reading the result

The score runs 0-100 and is relative, not an absolute lab measurement — it's most useful for comparing a few shots of the same scene to pick the sharpest one, or for a fast sanity check before you post. “Soft” means the image likely has focus or motion blur; “Sharp” means strong edge detail throughout.

Tips

• Zoomed-in crops of an already-sharp photo can still score high — this measures the pixels it's given, not the original capture.
• A shallow depth-of-field portrait can score lower even when correctly focused, since large out-of-focus areas reduce the overall edge detail.

What is a sharpness score?

The sharpness score is a relative measure of how much fine detail and edge contrast a photo contains, calculated with a standard computer-vision technique called Laplacian variance. The tool scans the image for abrupt brightness changes — edges — and measures how much variation there is. A photo full of crisp, well-defined edges scores high; one that's blurred, soft, or heavily smoothed scores low.

Why photographers use it

Comparing a handful of near-identical shots — several frames from a burst, multiple takes of the same pose or product — is tedious by eye at 100% zoom, one at a time; this picks the sharpest one instantly. It also works as a quick sanity check before posting or sending a photo, catching an accidentally soft or motion-blurred shot before it goes out.

Common mistakes

• Comparing scores across unrelated photos. the score is relative and scene-dependent — a busy, detailed scene naturally scores higher than a smooth, minimal one. It's only meaningful when comparing variations of the same or a similar shot, not as an absolute quality ranking across different photos.
• Reading a low score on a shallow-depth-of-field portrait as a focus mistake. large, deliberately out-of-focus background areas reduce the overall edge detail across the frame even when the subject itself is tack-sharp. A correctly focused portrait can score lower than a busier, deep-focus shot.
• Uploading a compressed or resized copy and judging the original by it. this tool measures the pixels it's actually given. A photo re-saved at low quality or downscaled before upload will score lower than the original file, regardless of how sharp the original capture was.
• Assuming a high score means the intended subject is in focus. the score reflects overall edge detail across the whole frame, not specifically where the sharpest point sits. A photo can score well because a busy background is detailed, while the actual subject is soft.
• Trusting the number without looking at the photo. this is a fast triage tool, not a replacement for actually viewing the image. Use it to narrow down candidates from a batch, then confirm the winner by eye before making a final choice.

Example settings

FAQ

Is 0-100 an absolute quality scale?

No — it's relative. It's most useful for comparing several shots of the same or a similar scene, not as a universal quality benchmark across different photos.

Why did a photo I think is sharp score lower than expected?

Likely a shallow depth of field reducing overall edge detail even with a sharp subject, or the file was compressed or resized before you uploaded it — both lower the score without meaning the original capture was actually soft.

Is my photo sent anywhere when I upload it here?

No — this runs entirely in your browser. Nothing is uploaded to a server for this tool.

Can I check a whole batch of photos at once?

Not currently — upload and check one at a time, comparing the scores yourself. It's a fast per-photo check rather than an automated batch culling tool.

What counts as a "good" score?

There's no fixed pass/fail threshold. The "Soft" label flags a likely focus or motion issue, but always weigh the number against what you can actually see in the photo — especially for intentionally shallow-depth-of-field shots.