Image Tampering Detection
Sandip Shinde, Apurv Waghmare, Anushka Varpe, Durva Ajgaonkar, Arya Alurkar · 2025
This paper proposes an image forgery detection technique combining the use of Convolutional Neural Networks (CNNs) with Error Level Analysis (ELA) for efficiently distinguishing tampered images from those that are original. ELA identifies potential regions of manipulation based on the discrepancy of the compression artifacts, whereas the VGG16based CNN classifier distinguishes between original and tampered images. The proposed model was assessed using the CASIA2 dataset, thus illustrating its effectiveness in accurately classifying images. This research contributes to the field of digital forensics, providing an advanced method for ensuring the integrity and authenticity of digital content, which is crucial for security and verification in various applications.