Discovering Tampered Image in Social Media Using ELA and Deep Learning

In the era of social media, we have access to millions of images. Nowadays with the rise of many advanced photo editing software finding a tampered image online is a very common situation. Most of the time an image is tampered for fun, but there are scenarios where an image is tampered with malicious intent and can cause harm to society. Digital image forensics is having a tough time dealing with tampered images due to the advancement of technology. Here, in our approach, we combined error level analysis (ELA) with a convolutional neural network (CNN) to classify whether an image is authentic or not. Our experiment has yielded a validation accuracy of 96.18% after 24 epochs. © 2022, The Author(s), under exclusive licence to Springer Nature Singapore Pte Ltd.