摘要
车流量检测是城市智能交通的关键技术之一.针对目前视频检测算法复杂度高、检测准确率不高的问题,提出了一种快速车流量检测算法.该算法通过划定进入和离开检测线,并结合背景差分方法,将目标的面上检测,转化为线上检测,避免了对目标复杂的跟踪过程及其带来的检测误差,有效降低了算法的复杂度,提高了检测的准确率.实验表明,该算法能够快速分车道检测车辆,计算复杂度低,检测准确率高,且具有车辆速度测量能力,能够为智能交通系统提供必要的支持.
Vehicle flow detection is one of the key techniques in intelligent transportation system. The present video traffic detection algorithm design is complex and its detection probability is low. A fast traffic detection algorithm is proposed. The algorithm defines the video detection line, combins the background subtraction method, and converts the face detection to line detection, which can avoid the complicated process of target tracking and its de- tection error and reduce the complexity of the algorithm effectively, while maintaining a high detection rate. Experi- ments show that this algorithm can quickly detect vehicles in lane, and has the advantages of less resource-inten- sive, high accuracy, strong robustness, and high real-time process. The new method can provide necessary support for the intelligent transportation system.
出处
《哈尔滨理工大学学报》
CAS
北大核心
2016年第4期19-24,共6页
Journal of Harbin University of Science and Technology
基金
中央高校基本科研业务费专项资金(3132013054)
关键词
智能交通系统
车流量检测
背景差分
检测线
intelligent transportation system
vehicle flow detection
background subtraction
detection line
