Video Steganography for Cybersecurity Applications: A Systematic Review of Classical, Hybrid, and AI-Driven Techniques

Abstract

The rapid growth of digital communication, cloud computing, Internet of Things (IoT), and intelligent surveillance systems has increased the demand for secure and covert information exchange. Video steganography has emerged as an effective information-hiding technique that enables confidential data trans-mission while concealing the existence of communication. Compared with im-age-based approaches, video steganography offers higher embedding capacity and improved imperceptibility due to the availability of multiple frames and temporal redundancy.
This paper presents a systematic review of video steganography techniques for cybersecurity applications. Existing approaches are categorized into classical spatial-domain methods, transform-domain techniques, hybrid models, and re-cent Artificial Intelligence (AI)-driven approaches. Major techniques including Least Significant Bit (LSB), Discrete Cosine Transform (DCT), Discrete Wave-let Transform (DWT), Random Pixel Selection (RPS), Huffman coding-based embedding, and deep learning-based methods are analyzed and compared. The review also examines commonly used benchmark datasets and performance metrics such as Peak Signal-to-Noise Ratio (PSNR), Mean Square Error (MSE), Root Mean Square Error (RMSE), embedding capacity, and robustness.
Furthermore, the study discusses applications in cybersecurity, healthcare, military communication, cloud environments, and smart surveillance systems. Key challenges and emerging research directions, including AI-assisted adaptive embedding, blockchain-enabled security, and quantum-resilient steganography, are highlighted. The review indicates that hybrid and AI-driven techniques provide improved security and robustness, making them promising solutions for next-generation secure multimedia communication systems.