摘要
为了突出交通标志的空间结构分布特征,提出一种基于极坐标分区局部二值模式的特征提取算法.对国际公开交通标志识别数据库GTSRB的43类交通路标中共计12 630张测试图像进行仿真实验,实验结果证明了该算法是有效的.
In order to highlight the spatial structure characteristics of traffic signs,a feature extraction algorithm based on local two valued model of polar coordinates was proposed.Simulation experiments were conducted on a total of12630test images of43types of traffic signs of GTSRB,an international public traffic sign identification database.The experimental results show that the algorithm is effective.
作者
韩伟
刘磊
马伟东
杜科
HAN Wei;LIU Lei;MA Weidong;DU Ke(Henan Electric Power Research Institute,Zhengzhou 450052;Ponovo Power Co. LTD,Beijing 100176 China)
出处
《湘潭大学自然科学学报》
北大核心
2017年第4期92-94,共3页
Natural Science Journal of Xiangtan University
关键词
交通标志识别
局部二值模式
车载视觉
traffic sign recognition
local binary pattern
vehicle vision
