基于监控视频的运动车辆检测优化方法  

Optimization method of moving vehicle detection based on traffic video

作　　者：王琼[1] 胡晓源[1] 吕晓猛[1] 陆刚[1] 

机构地区：[1]常熟理工学院信息化办公室,江苏常熟215500

出　　处：《计算机工程与设计》2010年第24期5359-5362,共4页Computer Engineering and Design

摘　　要：为满足实时性处理需要,提出了一种基于监控视频的运动车辆检测优化方法。运用自适应ROI(regionofinterest)提取算法,在获取可能出现运动车辆的区域后,基于帧间差分法与分块处理思想,提出了一种改进的背景提取算法,有效地提取运动目标区域。对提取的多目标运动区域进行分离,分别提取可能是车辆的区域后,提出了一种简单、快捷的阴影去除算法,有效地去除阴影,获得准确的运动目标区域。实验结果表明,该方法速度快、准确率高,能很好地满足实时性要求。In order to meet real-time processing requirements,the optimization method of moving vehicle detection based on traffic video is given.It uses the adaptive ROI（region of interest） extraction algorithm to the pick up possible regions and based on the theories of inter-frame difference and block,a new improved method is put forward to pick up background for obtaining target region in moving.After separating each target and picking up the regions of the vehicles in moving,an algorithm that is simple,efficient is put forward to remove the shadow and can obtain the accurate region of moving target.The experimental results show that the fast and accurate method can well meet the real-time requirements.

关 键 词：ROI提取 车辆检测 多目标分离 阴影去除 区域填充 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]