System and method for intelligent face recognition area network monitoring based on cloud technology

Abstract

The invention provides a system and a method for intelligent face recognition area network monitoring based on cloud technology. The system comprises a local server and a central server. The local server is connected with the central server through a network. The local server comprises a local face recognition system, a local people basic information data base, a local face information data base and a local people in-and-out information data base. The central server comprises a central face recognition system, an area complete people basic information data base, an area complete face information data base and an area complete people in-and-out information data base.

Description

technical field [0001] The invention relates to the field of intelligent and real-time area security monitoring, in particular to a cloud-based face recognition area network intelligent monitoring system and method. Background technique [0002] The traditional network video monitoring system uses cameras arranged at each monitoring point to obtain video images, and the images of each monitoring point are transmitted back to the central monitoring room through the Internet or video transmission lines, and are monitored by manual shifts on duty, or using storage media (tape disks) etc.) for future lookup or playback. This method has the following disadvantages: [0003] 1. Video images are non-structural data, which is difficult to analyze and count in real time. The objects that can be identified by human eye monitoring by memory are very limited, and one monitoring person monitors the images sent back by multiple cameras at the same time, negligence due to omission and fa...

AI Technical Summary

Problems solved by technology

[0003] 1. Video images are unstructured data, which is difficult to analyze and count in real time

Human eye monitoring relies on memory to identify very limited objects, and one monitor monitors the images sent back by multiple cameras at the same time, negligence due to omissions and fatigue is inevitable, let alone timely response

[0004] 2. The amount of multi-channel video image data is very large, occupying a huge amount of network transmission bandwidth resources, and the application range of the network video surveillance system is limited by the network transmission bandwidth

[0005] 3. The error rate is high and the stability is poor. The system’s recognition rate of suspicious objects will fluctuate with changes in clothing, surrounding environment, and status of monitoring personnel.

This system has the following disadvantages: it needs the active cooperation of the monitored object

Limited to one or a few monitoring points in a small area, the ability of regional network monitoring is lacking

Images