 This paper proposes an automated method to identify deep fake images using machine learning techniques. It first analyzes the image to determine whether it has been modified and then extracts deep features from the image before classifying them using support vector machines and k-nearest neighbors. The proposed method achieved an accuracy of 89.5%, demonstrating its effectiveness in identifying deep fake images. This article was authored by Rimsor Rafiq, Ramagantassi, Rashida Meen, and others.