基于二次帧差背景提取的车辆追踪方法  

Vehicle Tracking Based on Twice Frame-Difference for Background Extraction

作　　者：陈泓屺[1] 章云[1] 刘国英[1] 

机构地区：[1]广东工业大学自动化学院,广东广州510006

出　　处：《广东工业大学学报》2013年第1期55-60,共6页Journal of Guangdong University of Technology

摘　　要：针对智能交通系统对交通参数获取的实时性需求,利用背景的短时不变性,并结合帧差分法与背景差分法,提出一种运动车辆的追踪方法.该方法采用二次帧差法对背景快速提取,再对采用背景差分法获得的差分图像进行二值化,同时,为提高识别率对二值化图像进行了形态学操作,得到了较为完整的连通域.最后以帧间连通域的距离为依据,进行车辆追踪.实验结果表明,该方法具有良好的实时性和精确度以及鲁棒性,能够满足智能交通系统的实时需求.To satisfy the real-time demand for the traffic parameters in the Intelligent Transportation Sys- tem, it proposed a new method by combining the frame-difference method with the background-difference method for vehicle tracking with the invariance property of the background in a short time. This method was used to complete a rapid background extraction by the method of twice frame-difference, and the bi- nary difference image was obtained by the method of background-difference. Morphology operation was introduced to handle the binary image, which can improve the recognition rate to get a relatively complete connected domain. At last, vehicles were tracked, based on the distance between the connected domain of the close frames. Experimental results show that the scheme has good real-time and accuracy perform- ance as well as robustness to meet the demand in the Intelligent Transportation System.

关 键 词：车辆追踪 背景提取 帧差分法 二次帧差背景提取法 背景差分法 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]