Harnessing Machine Learning for Watermarking: A Survey on Robustness and Stealth in Image Protection
Kunal Routh, Priyanshu Mazumder, Amar Pa, Samit Karmakar, Soumik Kumar Kundu, Sutapa Ray, Pritom Adhikary, Bhaskar Roy · 2025
In recent years, the rapid growth of digital media has raised concerns over the protection of intellectual property rights. Image watermarking has emerged as a key technique for embedding copyright information into digital images. Traditional methods, such as spatial and frequency domain watermarking, often struggle with robustness against attacks like compression, noise, and resizing. However, machine learning (ML) has shown great promise in enhancing the effectiveness of watermarking systems by improving both robustness and imperceptibility. This paper provides a comprehensive review of the application of machine learning techniques in image watermarking, categorizing key approaches, evaluating their strengths and weaknesses, and discussing potential future directions.