基于自适应背景模型运动目标检测  

Moving Object Detection Based on Adaptive Background Model

作　　者：王陈阳[1] 周明全[1] 耿国华[1] 

机构地区：[1]西北大学可视化研究所,陕西西安710127

出　　处：《计算机技术与发展》2007年第4期21-23,26,共4页Computer Technology and Development

摘　　要：随着城市化速度的加快,机动车日益普及,人们在享受机动车所带来的巨大便利的同时,也面临着交通拥挤的困扰。随着计算机硬件技术和计算机视觉技术的发展,基于计算机视觉的交通监控系统成为可能。从一个交通视频序列中识别出运动物体是许多交通监控系统应用系统的重要任务,针对该问题,提出了一种建立在对视频序列中的整个背景情景的统计描述基础上的运动目标的检测的有效方法,该方法能够适应变化的背景,具有较强的鲁棒性和较好的实时性。With the acceleration of urbanization, automobile becomes more and more popular and people are enjoying the convenience that offer; however, people are trapped in the bewilderment of traffic congestion. Along with the computer hardware technology and the computer vision technology development, a computer vision - based traffic monitoring system has become possible. The main task for most a computer vision - based traffic monitoring application system is to identify the moving objects form a video sequence. To solve this problem, authors present a new statistics description of the whole background scene based on the video sequence, establish an effective algorithm for moving object detection. This method can adapt to the changing background with high robusmess and excellent real time performance.

关 键 词：背景差分 背景提取 背景更新 自适应背景模型 运动目标检测 

分 类 号：TP391.3[自动化与计算机技术—计算机应用技术]