Work

Real-Time Video Analytics: An Edge-Cloud Architecture Approach

Year: 2025
Type: article
Abstract: This paper offers a novel edge-cloud architecture for real-time video analysis. This design aims to solve problems connected to the use of bandwidth, latency, and limited computational resources. By integrating the benefits of cloud architecture and edge computing—which are complementary to one another—our method builds an efficient distributed processing system. The proposed system dynamically allocates computing responsibilities between edge devices and cloud servers. The network's features, the processing needs, and the resource availability drive this distribution. Though it keeps a high degree of accuracy in video analysis tasks, our approach cuts end-to-end latency by 62% when compared to alternatives depending only on cloud computing, according to the findings of our trials. The method reduces bandwidth use by 78% and attains 94% accuracy in cases involving object tracking and detection. By providing a scalable, efficient, and privacy-preserving solution, our work speeds up the creation of real-time video analytics. Surveillance, autonomous systems, and smart cities are just a few of the many sectors this approach fits.
Language: English
Open Access status: diamond
OpenAlex ID: W7219339423
FWCI:
Cites: 21
Cited by:
Related to: 10