Author(s) :

Kodam Swathi.

Volume/Issue :

Abstract :

With the exponential growth in video data across surveillance, digital communication, social media, and cloud-based storage, securing visual content has become a critical challenge. According to recent reports, over 80% of global internet traffic consists of video, making it a major target for unauthorized access, tampering, and data breaches. Traditional video security methods rely heavily on full-stream encryption, which is computationally expensive and often unsuitable for lightweight or real-time applications. In many manual systems, frame-level security is either ignored or implemented using weak methods such as basic LSB (Least Significant Bit) manipulation without any noise handling or robust decryption validation. These manual techniques, while simple, are highly vulnerable to attacks, offer no resistance to compression or noise, and cannot be reversed accurately in practical transmission scenarios. This gap has motivated the need for a secure, lightweight, and frame-level encryption method that can preserve content integrity while ensuring confidentiality. The objective of this research is to evaluate an existing LSB-based decryption model and propose a novel scrambling-based encryption technique for video frames. The existing model modifies the least significant bit of each pixel to embed data, but lacks robustness when exposed to distortions or data loss. The proposed system introduces a deterministic scrambling mechanism where pixel positions are randomized using a pseudorandom index generated from a fixed seed, allowing for precise encryption and decryption. Additionally, controlled probabilistic noise is injected during preprocessing to simulate real-world transmission environments, making the evaluation more rigorous. Comparative analysis based on PSNR, MSE, and SSIM shows that the proposed method significantly improves visual obfuscation while preserving decryptability. This research not only enhances video content protection at the frame level but also paves the way for secure deployment in resource-constrained environments such as mobile devices, smart surveillance, and low-power IoT platforms.

No. of Downloads :

0