Synthetic Image Identification using Deep Learning: A Review
IJSREM Journal · INTERANTIONAL JOURNAL OF SCIENTIFIC RESEARCH IN ENGINEERING AND MANAGEMENT · 2024
Recent developments in Artificial Intelligence (AI) have made it possible to create images of such high quality that people are unable to distinguish them from photographs taken in the real world. Given the vital importance of data authenticity and reliability, our system suggests a way to improve our capacity for computer-based AI image recognition. It is simple to imagine scenarios in which AI- generated images are exploited to create political unrest, fabricate acts of terrorism, and blackmail individuals. In this work, we have described a new deep learning-based method that can effectively distinguish AI-generated images from real images. Our method is capable of automatically detecting the authenticity of images present. Our system uses VGG-19, ResNet-50 and EfficientNet-B0 as the extractor of basic representation. We test our approach on a sizable dataset in order to replicate real-time events and improve the model’s performance on real-time data. Key Words: Synthetic Images, A.I. generated, Res-Net, Convolution Neural Network, CIFAKE, Explainable A.I.