计算机技术与发展2016，Vol.26Issue(5)：165-169,5.DOI:10.3969/j.issn.1673-629X.2016.05.036
基于Gaussian模型及Kalman滤波的车辆跟踪方法
Research on Vehicle Tracking Based on Gaussian Model and Kalman Filter
摘要
Abstract
In recent years,with the increase of motor vehicles,major"Du City" start to appear. The variety of traffic problems are increas-ing,thus making the rapid development of intelligent transport systems is imminent. Based on research of traditional tracking methods for vehicles,a tracking vehicles algorithm is proposed based on Gaussian model and Kalman filter. Through in-depth analysis of complex is-sues on external environment and self-conversion,the foreground is retrieved by using the background subtraction method. The mixture Gaussian model is adopted to model the adaptive background subtraction,and real-time updating is done to eliminate the interference of noise and fake target. In view of the target properties,in order to ensure tracking effect,through the establishment of the Kalman filtering prediction model for target vehicle,the stable tracking of targets is carried out through eliminating noise disturbance by using the uniformi-ty and continuity of characteristics of target parameters,and get the accurate traffic statistics. Experiments show that the method has good real-time and tracking performance and meet the needs for real-time monitoring.关键词
混合高斯模型/Kalman滤波/边缘特征/车辆跟踪Key words
mixture Gaussian model/Kalman filter/edge feature/vehicle tracking分类
信息技术与安全科学引用本文复制引用
丁晓娜..基于Gaussian模型及Kalman滤波的车辆跟踪方法[J].计算机技术与发展,2016,26(5):165-169,5.基金项目
陕西省教育专项科研计划项目(14JK1341) （14JK1341）
