摘要
车辆检测技术是智能汽车领域重要的研究方向,广泛应用于汽车安全辅助驾驶系统中,准确、快速地对道路前方车辆检测可以很好地保障车辆的安全行驶。论文主要研究了基于Adaboost机器学习的车辆检测算法,为提高检测精度,提出了一种多区域多分类器车辆检测方法,根据车辆特征在不同角度的差异性,把待检测的车辆分为左斜侧车辆、前向车辆、右斜侧车辆,然后选择不同的特征分别训练级联分类器进行检测,实验结果表明论文提出的方法检测速度、准确性都有了明显的提高。
Vehicle detection technology is an important research direction in the field of intelligent automobile.It is widely used in vehicle safety auxiliary driving system.Accurate and fast vehicle detection in front of the road can ensure the safe driving of vehicles.In this paper,the vehicle detection algorithm based on Adaboost machine learning is mainly studied.In order to improve the detection accuracy,a multi-area and multi-classifier vehicle detection method is proposed.According to the difference of vehicle characteristics in different angles,the vehicles to be detected are divided into left oblique vehicle,forward vehicle and right oblique vehicle,and then different features are selected to train cascade classifiers for detection.The experimental results show that the proposed method can detect the speed of the vehicle.The accuracy has been significantly improved.
作者
李春海
赵波
涂超
LI Chunhai;ZHAO Bo;TU Chao(Shanghai University of Engineering and Technology,Shanghai 201600)
出处
《计算机与数字工程》
2021年第10期2122-2125,2167,共5页
Computer & Digital Engineering
关键词
车辆检测
机器学习
多区域
分类器
vehicle detection
machine learning
multi-region
classifiers
