Spliced Image Forgery Detection Using Adaptive Over-Segmentation Combined With AKAZE, ORB, and SIFT Feature Descriptors

Abstract

The detection of digital image forgery is an essential component in the process of safeguarding the authenticity and integrity of visual data. Image forgery can be accomplished through a variety of tools. One of these techniques is called splicing, and it involves combining the contents of multiple images in order to create a composite image that has been forged. The identification of digital forgeries of this kind presents a significant challenge. One of the tried-and-true methods that is utilized in the process of forgery detection is called Adaptive Over Segmentation (AOS). Within the scope of this paper, we are integrating adaptive over-segmentation with effective feature extraction methods such as AKAZE, ORB, and SIFT. With the assistance of parameters like precision, recall, and F1 measures, the proposed method intends to enhance the outcomes in order to achieve the desired results.