Design of security platform based on cloud edge combined with machine vision

Abstract

Aiming at the problems of high delay, high bandwidth and low scalability of traditional security system based on deep learning, an extensible security platform design with edge side real-time reasoning is proposed to save bandwidth and computing resources. The design is based on smart chip technology and lightweight deep learning framework, design the fast inference and edge cloud training, intelligence analysis, management, system architecture, using the image optimization technology and large data distributed storage technology, to realize mass storage, video and image data effectively and efficiently design based on micro service business architecture, Service requirements such as monitoring scale expansion and personalized analysis are met. Experimental results show that the platform can significantly reduce intelligent reasoning delay, bandwidth and resource cost, and has scalability and practicability suitable for promotiont.