stegoveritas¶
stegoVeritas brute-forces common image hiding tricks. Aperi'Solve uses its image transforms: edge and blur filters, single colour planes, sharpness extremes, auto-contrast, equalisation, inversion and solarisation. It is a good catch-all when you suspect visual steganography but no specialized tool has hit yet.
What Aperi'Solve runs¶
Only on deep analysis, for every image. Aperi'Solve does not run the
stegoveritas command: it calls StegoVeritas' transform functions directly,
one after the other, in an isolated environment. That is deliberate:
- No bit planes. The decomposer already renders every Red, Green, Blue and Alpha plane, so they are not computed twice.
- No LSB brute force, carving, steghide or EXIF. Other analyzers (zsteg, foremost, steghide, exiftool) do these better.
- Palette images are converted to RGB first, so they get transforms too.
- Sizes above the 64 megapixel limit are refused, with an error, before the image is decoded.
The results are one Transforms gallery. Animated GIFs and APNGs are transformed on their first frame; their frames have their own stegoveritas frames result.
If Pillow refuses a transform for the image mode (for instance inversion on an image with transparency), that transform is skipped and the note on the result lists it.
Reading the output¶
Each image is one filter applied to the whole input: look for text, QR codes or shapes that only appear under one specific transform.
Running it manually¶
Aperi'Solve's subset is easy to reproduce with the StegoVeritas CLI, which also writes the bit planes:
$ stegoveritas -imageTransform image.png
$ ls results/
Installing locally¶
$ pip install stegoveritas
$ stegoveritas_install_deps
Common CTF patterns¶
- A message readable only after inversion or a specific colour filter.
- Text that appears in the sharpness-enhanced or edge-detected variant.