Cognitive Cyber Crimes in the Era of Artificial Intelligence
by Rajesh Kumar Chakrawarti, Romil Rawat, Kriti Bhaswar Singh, A. Samson Arun Raj, Abhishek Singh, Hitesh Rawat, Anjali Rawat
21AI-Powered Image Morphing and Watermarking Techniques for Threat Attribution
Chandrapal Singh Dangi1*, Abhishek Singh Rathore2, Kamal Borana3 and Sachin Chirgaiya4
1Department of Information Technology, Manipal University, Jaipur, India
2Department of AI and DS, SVIIT, Shri Vaishnav Vidyapeeth Vishwavidyalaya, Indore, India
3Department of AI and DS, SVIIT, SVVV, Indore, India
4Department of Artificial Intelligence and Data Science (AI&DS), SVIIT, SVVV, Indore, India
Abstract
This study presents an artificial intelligence (AI)–powered framework integrating advanced image morphing and robust watermarking techniques for effective threat attribution in digital media. Leveraging the widely recognized CASIA Image Tampering Detection Evaluation Dataset v2.0, which contains over 12,000 authentic and tampered images, our approach uses a novel hybrid DeepMorph-WaterNet model combining convolutional autoencoders with a resilient spread-spectrum watermarking scheme. The method ensures imperceptible morphing while embedding traceable watermarks resistant to common attacks such as JPEG compression, cropping, and geometric transformations. Performance evaluation using peak signal-to-noise ratio (PSNR), Structural Similarity Index Measure (SSIM), and bit error rate reveals an average PSNR of 38.6 dB, SSIM of 0.97, and a watermark robustness of 98.4% accuracy under multiple adversarial conditions. These results demonstrate superior fidelity and attribution reliability compared to baseline ...
Become an O’Reilly member and get unlimited access to this title plus top books and audiobooks from O’Reilly and nearly 200 top publishers, thousands of courses curated by job role, 150+ live events each month,
and much more.
Read now
Unlock full access