摘要
为提高智能视频监控中目标检测与跟踪的实时性、稳定性以及设计灵活性,基于现场可编程门阵列技术,设计一个应用于静态背景下的单目标跟踪系统,并设定报警功能。通过帧间差分和形态学方法减少环境变化对系统的影响,利用多点定位算法实现目标跟踪。实验结果表明,该系统稳定且处理速度快,当目标在摄像头视野内突然停止时也可进行定位与跟踪,具有较好的鲁棒性。
In order to improve real-time,stability and design flexibility of target detection and tracking in intelligent video surveillance,a real-time single target tracking system with alarming function applied to static background based on Field Programmable Gate Array（FPGA） is designed.Frame difference combined with morphology is used to reduce the influence of change of environment on the system,and multi-point positioning location algorithm is used to achieve the purpose of target tracking.Simulation results show that the system runs quickly and stably,the target within the camera view can be real-time tracked when the target is not moving,and the system is robust.
出处
《计算机工程》
CAS
CSCD
2012年第5期243-246,共4页
Computer Engineering
基金
广东省教育部产学研结合基金资助项目(2010B080703025)
关键词
视频监控
现场可编程门阵列
目标检测
目标跟踪
帧间差分
形态学
video surveillance
Field Programmable Gate Array（FPGA）
target detection
target tracking
inter-frame difference
morphology
