CONVOLUTIONAL NEURAL NETWORKS FOR REAL VS. SYNTHETIC IMAGE DETECTION USING CIFAKE

Recent advances in synthetic data have enabled the generation of images with such high quality that human beingscannot distinguish the difference between real-life photographs and Artificial Intelligence (AI) generated images.Given the critical necessity of data reliability and authentication, this article proposes to enhance our ability torecognize AI-generated images through computer vision. Initially, a synthetic dataset is generated that mirrors theten classes of the already available CIFAR-10 dataset with latent diffusion, providing a contrasting set of imagesfor comparison to real photographs. The model is capable of generating complex visual attributes, such asphotorealistic reflections in water. The two sets of data present as a binary classification problem with regard towhether the photograph is real or generated by AI. This study then proposes the use of a Convolutional NeuralNetwork (CNN) to classify the images into two categories; Real or Fake. Following hyperparameter tuning andthe training of 36 individual network topologies, the optimal approach could correctly classify the images with92.98% accuracy. Finally, this study implements explainable AI via Gradient Class Activation Mapping to explorewhich features within the images are useful for classification. Interpretation reveals interesting concepts withinthe image, in particular, noting that the actual entity itself does not hold useful information for classification;instead, the model focuses on small visual imperfections in the background of the images. The complete datasetengineered for this study, referred to as the CIFAKE dataset, is made publicly available to the research communityfor future work.