Prove an image is real.
TrueImage analyses photos with CNNs to detect manipulation, then anchors the verification result on a blockchain so newsrooms, courts and investigators can trust it.
The problem
Manipulated images spread faster than the truth.
Editing tools make forgeries easy to create and hard to spot. Older detection techniques struggle with accuracy and scale, and their results are hard to prove later.
accuracy of our forgery detection model.
recall: most forged images are caught.
F1-score, with 0.92 precision.
Source: the founding research paper (AICCONF, 2024).
How it works
How it works
A clear path from data to decision.
Upload
An image is submitted for verification.
Analyse
A CNN checks for signs of manipulation.
Record
The result and image fingerprint are written to a blockchain.
Share
A verifiable report is shared with editors or legal teams.
Product
What makes it different.
Every feature is grounded in the founding research.
CNN detection
Trained on authentic and forged images made with various manipulation techniques.
Blockchain proof
Tamper-proof record of every verification.
High accuracy
0.95 accuracy and 0.93 F1-score.
Evidence-ready
Reports designed for journalism and legal processes.
Who it's for
Who it's for.
Newsrooms & fact-checkers
Verify images before publishing.
Legal & forensic teams
Trustworthy evidence for cases.
Insurers
Detect manipulated claim photos.
The research
Built on peer-reviewed research.
- Merges convolutional neural networks with blockchain technology for image forgery detection.
- Trained on a curated dataset of authentic and forged images.
- Accuracy 0.95, precision 0.92, recall 0.94, F1 0.93.
- Addresses accuracy and scalability limits of earlier techniques.
Khalaf, L. I., et al. (2024). Image Forgery Detection using Convolutional Neural Networks and Blockchain Technology. Cognitive Models and Artificial Intelligence Conference (AICCONF), 316–321. ACM.
Team
The team behind it.
Founded in 2024 by a student team from two faculties.
Faculty of Science (Forensic Evidence)
Forensic methodology and evidence standards.
Faculty of Engineering Technologies
Deep learning and blockchain engineering.
FAQ
Questions we hear often.
What kinds of edits can it detect?
It is trained on multiple manipulation techniques such as splicing and copy-move.
Why blockchain?
It makes verification results tamper-proof and independently checkable.
Is this a finished commercial product?
It is a research-based student startup moving from validated prototype toward pilots with real users.
Can we partner or run a pilot?
Yes. We are looking for organisations willing to test the solution in real conditions and share feedback.
Bring TrueImage to your organisation.
We are onboarding pilot partners now. Read the research and get in touch.