A Lightweight Adaptive Encryption Scheme for Secure Real-Time Video Surveillance

Abstract

Real-time video surveillance systems deployed on resource-constrained Internet of Things (IoT) edge devices face a continuous trade-off between strict security requirements and real-time performance constraints. Standard cryptographic protocols often introduce high computational overhead and latency, making them unfeasible for streaming high-definition video in resource-limited environments. To address these challenges, this paper proposes a lightweight adaptive encryption scheme tailored specifically for secure real-time video surveillance. The proposed framework dynamically adjusts its cryptographic intensity based on real-time network conditions, frame regions of interest (ROI), and motion sensitivity. By selectively encrypting sensitive visual features while applying lightweight operations to non-critical regions, the system substantially reduces processing overhead and latency without compromising data confidentiality. Experimental evaluation demonstrates that the proposed approach achieves high encryption efficiency, robust defense against common cryptographic attacks, and minimal frame delay, making it an ideal candidate for low-power, real-time smart surveillance deployments.