AQID™ · working prototype

image
provenance

Find every copy, crop and re-edit of your images across the web, from a lightweight ~34kb fingerprint, not a costly model.

One image, followed

One photo. Four sites. Three days.

On Tuesday, February 10, 2026, ESA published this image. Within three days it had run on three more sites, five times in a new crop. AQID linked all nine copies back to the original by blanket crawling each site, generating a lightweight fingerprint for every image found, and comparing each fingerprint for matches.

Hubble captures light show around rapidly dying star: the original image, as published by ESA
Outlined: where each cropped copy sits in the original. Image: ESA, CC BY 4.0 IGO.
  1. Feb 10, 2026 ESA The original
  2. Feb 10, 2026 NASA Same image, and in 19 more articles
  3. Feb 11, 2026 Space.com Four crops
  4. Feb 13, 2026 ScienceDaily Cropped
Talk about licensing →

Measured

How often it’s right

100%of links correct on hand-labeled groups, with no links between unrelated images
96%of hand-labeled same-source pairs found
98.4%of 1,280 edited copies, made after the index was built, traced to their source
Each edit applied to 40 crawled images, then searched for. Matcher v24, November 16, 2026.
Thumbnail, re-encode, format change40 of 40
70% crop with blur, noise or heavy JPEG40 of 40
70% crop, flipped upside down40 of 40
70% crop with a gamma or contrast change39 of 40
80% crop under a watermark39 of 40
Crop, mirror, rotate 90°, grayscale and JPEG 50, all at once39 of 40
Rotated 20° or 45°, then cropped to fit37 and 39 of 40
Corner crop keeping 40% of the frame30 of 40

Your turn

Try to lose it

Pick an image and do your worst. Crop it, turn it, mirror it, drain the color, crush the JPEG. Your edit becomes an AQID and is searched against all 2,610 crawled images, frozen in our index.

1. Pick an image

2. Change it any way you like

3. Your version

This exact file is what gets searched

AQID digest ········

 

Followed

Thirteen more trails

Fourteen images ran on three or more sites. Each stop is dated by the article it ran in, and links to it.

Yes, but

Fair questions

Is this a perceptual hash?

No. There is no image-to-image comparison, no perceptual hash, no trained model and no inference. Each image is read once to derive its AQID, about 34 KB. After that only the AQID is used, and the same two AQIDs always give the same answer.

What edits does it survive?

Re-encodes, thumbnails, crops down to 70% of the frame, flips, watermarks, rotations, and all of them stacked at once: 37 to 40 times in 40. The hardest case we tested, a corner crop keeping 40% of the frame, came back 30 times in 40.

Will it confuse two photos that only look alike?

It keeps them apart. Every pair of AQIDs is classed as the same image, a derivative, a similar scene, or unrelated, and on hand-labeled groups it made no links between unrelated images.

What was it tested on?

476 public-domain NASA articles and every image in them, then ESA, Space.com, Phys.org, ScienceDaily, Universe Today, MIT News and more: 1,411 articles and 2,610 images across 11 sites. Then 1,280 fresh edited copies, made after the index was built, to see whether it could find them.

Can we license it?

AQID is a working prototype of the content identity disclosed in PCT/US26/28630, filed May 18, 2026. Any license is by separate written agreement; nothing on this page grants rights.

In Production AQID™ is in production on sidenote.news, which offers hyper-local news and community stories in Thetford, VT. Across 1,145 images, it found 27 groups of duplicates that an editor confirmed, removing 30 duplicate copies and about 10 MB of storage. pHash, at a typical threshold, would have caught 17 of those groups. AQID never linked unrelated images.

Rights follow the picture, not the file

From de-duplication to rights management — if your business depends on knowing where your images end up, let’s talk.

Talk about licensing →
Nick Clark, inventor Or write to me directly: licensing [at] qu3ry [dot] net

Disclosure scope

AQID implements the content identity disclosed in PCT/US26/28630, Structural Content Identity and Rights-Grade Admissibility for Digital Artifacts, filed May 18, 2026. See also Content Anchoring. This page reports behavior and measured results only. No rights are granted by this page; any license requires a separate written agreement.

NASA imagery is in the public domain. ESA imagery is licensed CC BY 4.0 IGO. Images on other sites belong to their publishers and appear as small thumbnails linking to the article they ran in.