Journal of Punjab Academy of Forensic Medicine & Toxicology
SCOPUS
  • Year: 2025
  • Volume: 25
  • Issue: 2

Visual Forensics of Deepfake Images: Manual Detection and Classification of Visual Artifacts

  • Author:
  • Anindita Malik1, Munimanda Sweshika2, M Pooja3, G Deepak Raj Rao4, Kiran Kumari5,*
  • Total Page Count: 5
  • Published Online: Apr 24, 2026
  • Page Number: 83 to 87

1PhD Research Scholar, Forensic Science Department, Lovely Professional University, Phagwara, Punjab, 144411

2Master of Forensic Science, Forensic Science Department, Lovely Professional University, Phagwara, Punjab, 144411

3Master of Forensic Science, Forensic Science Department, Lovely Professional University, Phagwara, Punjab, 144411

4Department of Cyber Security and Digital Forensics, National Forensic Sciences University, Delhi

5Assistant Professor, Forensic Science Department, Lovely Professional University, Phagwara, Punjab, 144411

*Corresponding Author: Dr. Kiran Kumari, Assistant Professor, Forensic Science Department, Lovely Professional University, Phagwara, Punjab, 144411, E-mail: malikraw91@gmail.com, Contact : +918708358690

Online Published on 24 April, 2026.

Abstract

The present work investigates the natural human ability to recognize and counter the negative impact of deepfake technology. Artificial Intelligence (AI) and machine learning play a crucial role in creating highly realistic synthetic media, known as deepfakes. Deepfakes have become a powerful tool for spreading misinformation and creating inappropriate content. Social media platforms are flooded with such synthetic media, raising serious concerns about the authenticity of digital information directly available to users. Moreover, there is a lack of knowledge and awareness among layman about the authenticity of such media.

To address this issue, manual analysis of hundreds of deepfake images, focusing on identifying visual anomalies without the aid of specialized software or tools, is performed.

Through this process, various visual markers are discovered, ranging from common to rare, that can help distinguish deepfake images.

These indicators are intended to contribute to public awareness and enhance digital media literacy, helping social media users recognize and question the authenticity of deepfake media.

This study highlights the importance of identifying visual anomalies and empowering the users to identify and safeguard themselves from deepfakes and its harmful impact.

Keywords

Deepfake, Exploitation, Artificial Intelligence, Digital Media, Social Media