University of Twente Student Theses
Image Tampering Detection in Social Media
Qureshi, Faizan Mazhar (2024) Image Tampering Detection in Social Media.
|
PDF
1MB |
| Abstract: | ’A picture is worth a thousand words’ is a common phrase used worldwide, but in today’s era, these pictures can be edited, causing the truth to change. Manipulated images, including fabricated news, and merged images are widespread on platforms such as Facebook and Twitter, contribut- ing to the spread of false and harmful content. Public opinion can easily be swayed via this method and can result in uninformed decisions. Thus, exploring ways to detect image tampering is critical for the well-being of social media users. Current research primarily focuses on the detection of tampered images, but social, moral, and psychological issues caused by im- age tampering in social media are not widely addressed. Moreover, research lacks in providing a complete solution to this rising problem of widespread tampered images over social media. It also mostly focuses only on single and common techniques like Error Level Analysis (ELA) and Double JPEG Compression. In this research, we evaluate the problem of the spread of tam- pered images on social media and its consequences. We develop a solution by utilizing some existing advanced image analysis techniques, such as Discrete Cosine Transformation (DCT) coefficient, Histogram Value Analysis, Noise Variance Inconsistencies, and ELA. These techniques have not been exten- sively explored in the field of detecting manipulated images. We integrated these techniques by training machine learning models and incorporating them into our developed web application. The solution was developed using a combination of Python, JavaScript, and HTML. The results help users validate image authenticity on social media platforms by directly uploading or providing the image link in the web application, reducing tampered image spread. |
| Item Type: | Essay (Bachelor) |
| Faculty: | EEMCS: Electrical Engineering, Mathematics and Computer Science |
| Subject: | 54 computer science |
| Programme: | Computer Science BSc (56964) |
| Link to this item: | https://purl.utwente.nl/essays/98167 |
| Export this item as: | BibTeX EndNote HTML Citation Reference Manager |
Show download statistics for this publication
Show download statistics for this publicationRepository Staff Only: item control page