CNN-Based Detection of Tampered Digital Images

This project presents a novel approach for detecting tampered digital images using Convolutional Neural Networks (CNN). Our methodology leverages the power of deep learning to analyse image content and identify alterations introduced through manipulation. The key innovation lies in the integration of CNN architectures, enabling robust and accurate detection of tampering in various image types. Additionally, the project incorporates Error Level Analysis (ELA) as a complementary feature to enhance the detection accuracy. ELA helps identify areas of an image with inconsistent compression levels, aiding in the identification of potential tampering artifacts. The synergy between CNN-based analysis and ELA creates a comprehensive solution for the reliable detection of tampered digital images. This approach demonstrates promising results in differentiating between authentic and manipulated images, making it a valuable tool for forensic image analysis, and ensuring the integrity of digital content.