Research of Moving Target Dection Technology in Intelligent Video Surveillance System

Moving object detection is one of the most challenging tasks in intelligent video surveillance system. It aims to detect targets with a camera instead of human eyes and assist people to complete monitoring and control tasks. Moving target detection, classification, tracking, behavior understanding and description are included in intelligent video surveillance, but moving target detection is the key step. In this paper, according to the background is static or dynamic, the existing major moving target detection algorithms are divided into two categories. A detailed introduction is presented. Some methods such as frame subtraction, background subtraction and optical flow method are included in static background. Furthermore, other algorithms are also presented in dynamic scene, and finally the prospects of moving target detection are reviewed.