摘要
对于视频监控图像中运动目标位置的检测,现有的定位跟踪系统主要采用双目或结构光立体视觉技术,存在系统设计复杂、检测速度慢等缺点.有鉴于此,文中将地面约束引入到单目视觉监控中,提出了一种单目序列图像运动目标跟踪方法.该方法利用摄像机安装信息和几何成像原理,结合非线性补偿,推导出单目不对称非线性成像的地面运动目标实际位置计算公式.通过搭建室内监控测试平台,在序列图像中进行了寻找步行者足印位置的实验,结果验证了文中方法的有效性.
For the location detection of a moving target in surveillance videos,the existing locating and tracking systems mainly based on binocular or structured-light stereo vision techniques are of high system design complexity and low speed.In order to remedy these drawbacks,a new method of moving target tracking based on monocular sequence images is proposed by introducing a ground constraint.In this method,the formulas to estimate the position of moving target in asymmetric nonlinear imaging are derived according to camera＇s installation information and geometrical imaging principle together with nonlinear compensation.By setting up an indoor monitoring and test platform,an experiment looking for a walker＇s foot position in an image sequence is carried out,and the results verifies the effectiveness of the proposed method.
出处
《华南理工大学学报（自然科学版）》
EI
CAS
CSCD
北大核心
2012年第3期94-99,共6页
Journal of South China University of Technology(Natural Science Edition)
基金
NSFC-广东省自然科学联合基金资助项目(U0735004)
华南理工大学中央高校基本科研业务费专项资金资助项目(X2dXD2116370)
关键词
机器视觉
定位跟踪
目标检测
不对称投影
machine vision
location tracking
object detection
asymmetric projection
