Digital forensics is the field of study that deals with determining and verifying the integrity of digital data. Photo morphing detection is a popular technique used in digital forensics. The use of fake photos in journalism, social media, and the judicial system has increased due to digital media being more accessible and simpler to work with, as well as the increased accessibility of image editing tools. As such, there is an increasing demand for genuine tools to detect instances of picture modification. Combining Error Level Analysis (ELA) with deep learning presents a guaranteed new strategy for identifying image forgery. The designed method, “Image Forgery Detection Based on ELA and Deep Learning,” aims to achieve this by providing a dependable and effective method for detecting image forgeries. By combining Error Level Analysis (ELA) with deep learning, we strive to develop a system that effectively and precisely detects photo forgeries. This model is evaluated using the “CASIA2” Image Forgery Detection dataset. The dataset used in the proposed model is available on Kaggle, a central repository for academic datasets. This dataset consists of 7492 authentic and 5124 manipulated photographs, with an accuracy rate of at least 90%. words.