Detecting Manipulated Digital Entities Through Real-World Anchors

Document Type

Article

Publication Date

1-1-2025

Abstract

As the digital and physical worlds become increasingly intertwined, manipulating digital entities, such as avatars, digital twins, and virtual environments, poses significant challenges for forensic analysis. Traditional digital forensics often lacks the mechanisms to detect sophisticated alterations within these virtual constructs. This paper presents a novel framework called DeepAnchor that establishes real-world anchors for effective forensic analysis in digital domains. DeepAnchor adopts a Feature-Integrated and Attention-Enhanced digital watermarking mechanism based on GAN (FIAE-GAN). By creating verifiable links between digital entities and their physical counterparts, DeepAnchor facilitates the detection of manipulations that might otherwise go unnoticed. Through a series of case studies, we demonstrate FIAE-GAN’s ability to detect and analyze manipulated digital entities accurately. DeepAnchor lays the foundation for future forensic strategies to preserve authenticity and trust in an era of increasingly accessible and widespread digital manipulation.

Publication Title

Lecture Notes on Data Engineering and Communications Technologies

Link to Full Text

Share

COinS