期刊名称：International Journal of Computer Trends and Technology
电子版ISSN：2231-2803
出版年度：2015
卷号：29
期号：1
页码：31-39
DOI：10.14445/22312803/IJCTT-V29P107
出版社：Seventh Sense Research Group
摘要：There are now large networks of CCTV cameras collecting great amounts of image data, many of which deploy PanTiltZoom (PTZ) controllable cameras. A multicamera and multisensor system has potential both for gaining improved imaging quality and for capturing more relevant details .Such a system can also cause overflow of information and confusion if data content is not analyzed in realtime. Video Analytics is the emerging technology where Computer Vision and Pattern Recognition techniques are used to filter and manage real time CCTV videos for security and intelligent monitoring. Background subtraction, object detection, object tracking, reidentification, and behavior analysis are the most important components for a Video Analytics system. The scientist has some of the cutting edge technologies in this area, which exploit recent statistical and differential geometric theories and adapt them to challenging tasks for example, individuating eye directions, tracking groups of people, reidentifying individuals in different days that take place in real case scenarios. Detecting and tracking human beings in a given scene represents one of the most important and challenging tasks in computer vision. We are interested to consider some of powerful methods in these issues for their implications in videosurveillance and driver assistance systems.
关键词：Multi object tracking; Multi objectdetection; video surveillance system.
