Enhancing Data Hiding Techniques in Image Processing through AI-Driven Edge Computing
Zahid J Hussain · 2025
In the digital era, data security and confidentiality are of paramount importance, particularly in image-based communications. This study explores the integration of Artificial Intelligence (AI) with edge computing to enhance data hiding techniques in image processing. Traditional methods of steganography often face challenges related to detection resistance, data capacity, and computational efficiency. By leveraging the localized processing capabilities of edge computing and the adaptive learning features of AI, this research proposes a hybrid model that ensures real-time, secure, and intelligent data embedding. The model employs deep learning algorithms for identifying optimal embedding regions within images based on texture complexity and perceptual invisibility, thereby maximizing data payload while maintaining visual integrity. Additionally, edge devices are used to process and embed data at the source, significantly reducing latency and exposure to cyber threats. Experimental results demonstrate improved robustness against stainless attacks, enhanced embedding efficiency, and greater adaptability to dynamic image conditions. This approach not only bolsters data security but also aligns with the growing demand for decentralized and privacy-preserving computing in Internet of Things (IoT) and multimedia applications.