dc.contributor.author |
Nandasena, Kalaru |
|
dc.date.accessioned |
2025-06-11T03:30:09Z |
|
dc.date.available |
2025-06-11T03:30:09Z |
|
dc.date.issued |
2024 |
|
dc.identifier.citation |
Nandasena, Kalaru (2024) Enhancing Image Authenticity and Integrity in Social Media: Machine Learning-Driven Deepfake Detection and Metadata Watermarking for Image Authenticity Enhancement. BSc. Dissertation, Informatics Institute of Technology |
en_US |
dc.identifier.issn |
2019409 |
|
dc.identifier.uri |
http://dlib.iit.ac.lk/xmlui/handle/123456789/2490 |
|
dc.description.abstract |
"The authenticity dilemma in social media photography is being addressed by a ground-breaking
solution. The project uses machine learning to analyze submitted photographs in real-time,
identifying true captures and modified deepfakes. Once authentic, discrete watermarks are added
to enhance integrity without compromising visual quality. The embedded watermark acts as a
permanent sign of authenticity, and if targeted for deepfake modification, the watermark is subtly
disturbed for quick identification. This approach counteracts misleading material on social media
sites, allowing users and platforms to navigate digital visuals with greater confidence. Deepfake
technology, a sophisticated tool for creating fake images and movies, has led to a trust crisis in
social media." |
en_US |
dc.language.iso |
en |
en_US |
dc.subject |
Machine Learning |
en_US |
dc.subject |
Deepfake |
en_US |
dc.subject |
Image Enhance |
en_US |
dc.title |
Enhancing Image Authenticity and Integrity in Social Media: Machine Learning-Driven Deepfake Detection and Metadata Watermarking for Image Authenticity Enhancement |
en_US |
dc.type |
Thesis |
en_US |