Built-in intelligent monitoring system for automatically detecting moving target

Abstract

The invention discloses an embedded intelligent monitoring system for automatically detecting moving target, including a video input device, a visual analysis system and a network input and output device. In order to perform high precision real-time detection of moving targets, the system adopts a method of using a real-time visual analysis algorithm to detect moving targets in real-time, and the method is divided into a training phase and a testing phase; the system will divide the obtained video data into a stable region and a dynamic region in the chrominance brightness spaces, adopting a simple and fast self-adapting single-Gaussian model to the stable region and adopting a computationally complex and effective non-parametric model to the dynamic region. The system has advantages of simple structure, high precision, real-time, strong expandability and multiplex network access function; and can effectively realization real time detecting function of moving target.

Description

technical field [0001] The invention belongs to the field of electronic equipment and relates to an embedded intelligent monitoring system for automatically detecting moving targets. Background technique [0002] With the rapid growth of video surveillance demand and the increasing scale of surveillance, manual surveillance is far from meeting the surveillance requirements, and the "intelligence" of video surveillance systems has become more and more urgent. [0003] The continuous expansion of the monitoring scale and the increasing number of visual analysis tasks have made the centralized visual analysis processing mode based on the PC central server more and more prominent limitations-computer resources and network communication bandwidth are seriously insufficient, and it is increasingly necessary to The front end of the camera performs real-time and efficient local processing on the surveillance video images, and only transmits the relevant alarm status and image data t...

AI Technical Summary

Problems solved by technology

In order to effectively detect moving targets, the single Gaussian model adopted by S.J.Mc, S.Jabri et al. in the document "Tracking Groups of People", and C.Stauffer and W.E.L adopted in the document "Adaptive background mixture models for real-time tracking" The mixed Gaussian model of A.Elgammal, D.Harwood and others in the literature "Non-parametric Model for Background" and their improved models have been proposed, these modeling methods in some specific Better results can be achieved in the environment, but in the face of more complex scenes, either the effect will become worse, or the effect will be better but the amount of calculation and resource requirements are large, and it is difficult to meet real-time requirements. Its limitations in embedded intelligent vision analysis Sex is very obvious

Images