cat > README.md << 'EOF'
Media Integrity & Tamper Detection for Images and Videos
Hemlock is a tool that helps you prove whether an image or video has been changed after it was created.
If a file is modified — even slightly — Hemlock will detect it, explain it, and show where it happened.
Images and videos can be:
- edited
- cropped
- re-encoded
- partially changed
- or maliciously manipulated
After that, it becomes very hard to answer a simple question:
“Is this file still the original one?”
Hemlock answers that question reliably.
- Detect any change in an image or video
- Catch changes as small as one pixel or one byte
- Verify files using public-key cryptography
- Show which video frame was modified
- Generate a JSON verification report
- Produce a visual image of the tampered video frame
Hemlock creates a strong digital fingerprint of the image and signs it. If even one pixel changes, verification fails.
A video is treated as many images (frames). Hemlock fingerprints each frame, links them together, and signs the result. If any frame is changed, removed, or reordered, verification fails.
- Private key → used only while signing
- Public key → used during verification
This guarantees:
- authenticity
- integrity
- tamper detection
Anyone can verify the file, but no one can fake a valid signature.
- C — fast image processing
- Python — video frame handling
- ECDSA cryptography — signing & verification
- JSON reports — machine-readable results
- Media authenticity checks
- Digital forensics
- ML dataset verification
- Academic and security research
Hemlock detects and explains any tampering in images or videos using cryptographic verification. EOF