摘要
在车辆模拟演示系统中,车辆的检测与跟踪是系统的核心任务,针对传统的车辆检测方法检测速度慢,检测准确率低等问题,设计并实现了一种基于Adaboost算法的车辆模型演示系统。通过对车辆的Haar特征构造Adaboost分类检测器,达到了准确分类的目的。实验结果表明,与其它车辆模拟演示系统相比,该系统具有更好的实时性与准确性。系统满足车辆模拟演示要求,检测准确率高达92.3%。
In the vehicle simulation demonstration system,detection and tracking is the core task of the system.A vehicle simulation demonstration based on Adaboost algorithm was designed and implemented in this paper to solve the problem of the low speed and the low detection rate of the traditional vehicle detection.The exact classification is implemented by constructed Adaboost classifier of vehicle Haar features.Experimental results show that,this system has better real-time and accuracy,compared with other vehicle simulation demonstration system.This system meets the requirements of vehicle simulation demonstration,and the detection rate is up to 92.3%.
出处
《计算机测量与控制》
CSCD
北大核心
2012年第1期250-252,共3页
Computer Measurement &Control
基金
国家自然基金项目(61071193)资助
