高饱和交叉口背景提取与更新算法  

An Algorithm of Background Extraction and Update for Saturated Intersection

作　　者：李熙莹[1] 罗长政[1] 王镇波[1] 罗东华[1] 

机构地区：[1]中山大学工学院//广东省智能交通系统重点实验室,广东广州510006

出　　处：《中山大学学报（自然科学版）》2012年第3期23-28,共6页Acta Scientiarum Naturalium Universitatis Sunyatseni

基　　金：国家自然科学基金资助项目(51078362)

摘　　要：为了满足在高饱和城市交叉路口采用背景差法实现车辆检测的需求,设计了一种基于背景评估的背景提取与更新算法。算法针对交叉路口的车流特点,提出了一种基于检测结果反馈的背景评估机制,并融合中值滤波与帧间差分算法,将含噪声的初始背景不断纠错更新,向真实背景逼近,最终提取出一个可用的背景图。实验结果表明:对于一个信号周期内不能放完所有排队车辆的高饱和交叉路口,算法具有较快的收敛速度,且算法复杂度满足实际工程应用要求。A method of background extraction and update for vehicle detection is proposed, which is special for the saturated intersection. According to the characteristics of traffic flow in intersections, a back- ground evaluation mechanism is established by using the feedback of vehicle detection result. Then the true background image is updated from the noisy initial background iteratively combining evaluation mechanism, Median Filter and Inter-frame Difference algorithm. The experiments show that the algorithm has a faster convergence rate in saturated intersection, and satisfy complex engineering needs.

关 键 词：背景提取 高饱和交叉口 中值滤波 帧间差分法 

分 类 号：TP391.4[自动化与计算机技术—计算机应用技术]