摘要
行道线检测是主动安全和视觉导航技术中的一个重要研究课题.在总结前人检测算法的基础上,设计了基于Adaboost算法的行道线检测方法.Adaboost算法作为一种新型的机器学习算法,可以在比随机预测略好的弱分类器基础上构建高精度的强分类器.该算法简单可靠、学习效率高,较好地解决了实时检测系统中速度和精度的矛盾.实验结果表明该方法有较好的检测效果.
As a new machine learning algorithm, Adaboost algorithm could construct a highly accurate classifier by combiining many weak classifiers that just are slightly better than random prediction. The algorithm is simple and reliable, and has high learning efficiency. For real-time object detection, it solves the contradiction between speed and precision relatively well. Lane detection is an important topic in active safety and visual navigation technology. In this paper, based on the analysis of existing lane detection methods, a new method based on Adaboost algorithm is designed. The experiments show that the new method can get better result.
出处
《江南大学学报(自然科学版)》
CAS
2007年第6期887-890,共4页
Joural of Jiangnan University (Natural Science Edition)