PHOTO LAB

Image Duplicate Region Detector

Find repeated blocks for editing review; label evidence as similarity, not forgery proof.

Image proof

Load the worked example or open your own input.

Preparing local example…Local processing · Review before use

Start with the worked example. Edit your input or settings, inspect the result, and open Export for task files or a complete ZIP. Save project keeps editable text and settings; image/file projects need their original source files when reopened.

Record details

Using Image Duplicate Region Detector

Find repeated blocks for editing review; label evidence as similarity, not forgery proof.

  1. Select Try example to see a complete image duplicate region detector workflow. Open supported still PNG, JPEG or WebP images. Geometry and source markers can be edited in the workspace.
  2. Adjust the labelled controls or open Input/Geometry to change the evidence. Run workflow applies the current input; undo and redo restore text/settings decisions.
  3. Inspect Original and Result, using source markers where supported. Findings describe cases that need review; missing evidence remains unknown.
  4. Open Export for task artifacts, JSON/CSV results or a complete ZIP. Save project keeps input/settings; projects with local binary files need their original files to be reselected when reopened.

Example

The worked example for Image Duplicate Region Detector produces 30 evidence records and 2 task artifacts. Groups aligned blocks by a quantized 4×4 grayscale signature. Flat textures and natural repetition create false positives; this is editing evidence, never a forgery diagnosis.

Questions & answers

What does Image Duplicate Region Detector do?

Groups aligned blocks by a quantized 4×4 grayscale signature. Flat textures and natural repetition create false positives; this is editing evidence, never a forgery diagnosis.

What should I open or paste?

Open supported still PNG, JPEG or WebP images. Geometry and source markers can be edited in the workspace. The worked example defines the supported input fields. Edit its values, run the workflow and inspect the method notes before exporting. Each selected file must be nonempty and at most 25 MiB, with at most 30 files and 80 MiB total. Image jobs use up to 12 still frames, six million pixels per image and 18 million total pixels. Individual format inspectors may apply stricter limits.

What results can I export?

The Export view includes the task-specific artifacts listed for your actual result, a typed JSON report, a CSV table and a complete ZIP. Save project stores editable input and settings; selected binary sources are represented by fingerprints and must be reselected. Secret-option values are excluded from saved projects.

Does it upload files, run commands or change a live system?

Processing uses your supplied evidence in this browser. Exports create new images and leave original files unchanged. Public pages include the site’s approved advertising and optional consent-based analytics.

Help improve this tool

Report a problem or suggest an improvement

Describe the issue without pasting private tool input. Feedback goes to our admin inbox.

Find another tool · Read practical guides