01
Pick a picture.
A PNG or JPEG. Larger images carry a longer message. Everything stays in the browser.
Drop an image, or click to browse
PNG · JPEG · up to ~4096 px
Vol. 01 · A field guide to invisible marks
Hide a message inside an image, recover it without the original.
An in-browser studio for hiding and recovering invisible traces inside images. Uses the DWT · DCT · SVD algorithm executed in your tab via Pyodide. Your pixels never leave the device.
Section 01
The studio.
Pick a picture, choose a message (text or image), embed. The Python package handles the rest.
02
Write the message.
A short text or a small image. The library encodes it as a bit array and embeds it across all three colour channels.
UTF-8 · up to 200 chars
22
Drop a signature image
PNG · grayscale or RGB
Advanced options
Optional. Two separate keys: one for the message content, one for the image-block scramble. Change them to test password-protected recovery.
03
Render & save.
The engine hides your message in your browser. No upload, no server, no waiting room.
Section 02
Recover the message.
Drop a marked image. If you used the studio above to hide the message, the trace shape is pre-filled; change it only if the mark came from somewhere else.
Drop marked image
No file selected
Recovered text
—
Recovered image
Colophon
A small credit, a real link.
The engine is the DWT · DCT · SVD algorithm,
executed in your tab via Pyodide with
opencv-python 4.11, numpy 2.x,
pywavelets and Pillow.
First load takes a few seconds while Pyodide boots and pulls ~30 MB of compiled wheels from the CDN. After that, every hide / recover is local. No analytics. No cookies. No server.
- EngineMystiMark · DWT-DCT-SVD
- RuntimePyodide 0.27
- StackPython · opencv · numpy · pywavelets
- ThemeLight / Cream / Ink / Oxblood
- StatusStudio is open
- VersionV1.0