摘要
在过去的几年中,很多交通标志检测和分类方法被提出。本文综述了近年来交通标志检测与分类的一些有效方法。检测的主要目标是定位包含交通标志的感兴趣区域,检测方法大致分为三大类:基于颜色、基于形状和基于学习。分类方法主要分为两类:基于手工特征的学习方法和深度学习方法。为了便于参考,还将检测和分类公开数据集进行了总结。
In the past few years, many traffic sign detection and classification methods have been proposed. This paper summarizes some effective methods of traffic sign detection and classification in recent years. The main goal of detection is to locate the region of interest containing traffic signs. Detection methods are roughly divided into three categories: color based, shape based and learning based. Classification methods are mainly divided into two categories: manual feature-based learning method and deep learning method. For reference, different detection and classification methods and different data sets are summarized.
出处
《软件工程与应用》
2021年第3期348-353,共6页
Software Engineering and Applications
