Deep Fake in picture using Convolutional Neural Network

Deepfake is a system that combines fake pictures and videos with deep learning. Deep learning is the source of Deepfake. The unethical practice of creating falsified photographs and movies is now possible because of neoteric advances in the fields of Artificial Intelligence and Machine Learning. Today, it is very easy to create photo simulative images using generative adversarial networks. These fake images and videos are widely available on the internet and social media. It is difficult to tell which of them is real or not. These images are typically taken with the goal of stirring up social disturbance, political turmoil, or distributing false information among the general public. Viewers will easily comprehend these images because they will look to be real. Deepfake is rapidly harming individuals, communities, companies, security, religion, and democracy, according to recent studies. These videos and photographs are of astounding quality, and they have a huge social media reach. It has far-reaching effects that are destructive beyond comprehension. An extensive overview of the different deep-fake methods is provided in this publication. So, the objective of this paper is to identify these fake images using a conventional neural network. In order to address this issue, we train a model for particular datasets, produce deep fakes, and then use that model to try to identify the deep fake. An authentic and false image is required for the training process in order to train the model.