论文部分内容阅读
智能交通是智慧城市的重要组成部分,面对复杂多变的道路背景,如何能够快速检测、跟踪道路监控影像中的动态目标,是智能交通建设的关键技术难点。本文根据道路监控视频特点,提出了采用道路约束条件与颜色特征集相结合的动态目标跟踪方法,以道路约束条件确定运动目标搜索区域,利用HSV颜色特征集进行特征匹配,然后基于IIR滤波背景法对背景影像进行更新及动态目标的检测,并根据道路约束条件与颜色特征相结合跟踪方法实现对动态目标的跟踪及动作预测。试验结果表明该方法可准确对运动目标进行检测与跟踪,且对慢速运动目标也具有较好的响应能力,实现了对道路动态目标的实时检测与跟踪。
Intelligent transportation is an important part of a smart city. Facing the complex and ever-changing road background, how to quickly detect and track the dynamic targets in road surveillance images is a key technical challenge in ITS. According to the characteristics of road surveillance video, this paper proposes a dynamic target tracking method based on the combination of road constraint and color feature set, locates the moving target search area with road constraints and uses HSV color feature set to perform feature matching. Then, based on IIR filtered background method The background image is updated and the dynamic target is detected. The dynamic target tracking and motion prediction are realized according to the combination of the road constraint and the color feature. Experimental results show that this method can accurately detect and track moving targets, and also has good response ability to slow moving targets, which realizes real-time detection and tracking of dynamic targets.