摘要
为研究驾驶员视觉特征与驾驶意图的关联性进行了自然驾驶试验,采集了场景视频、驾驶员运动状态和转向盘转角信号,构建了包含若干直行、左换道、右换道片段的数据集,开发了基于机器视觉的视线方向及头部姿态估计算法,系统地提取了驾驶员头、面部特征,并使用统计学方法获得了各特征与驾驶意图的关联性。
In order to study the correlation between driver′s visual characteristics and driving intentions,natural driving tests on open roads and semi-closed roads were carried out.Scene video,driver′s movement status,and steering wheel angle signals were captured.A data set consisting of several straight-through segments,left change segments,and right change segments was created.Line-of-sight direction and head posture estimation algorithm based on machine vision were developed.The visual features of the driver′s head and face were systematically extracted from the dataset,and the correlation between each feature and driving intention was obtained using statistical methods.
作者
张立军
唐鑫
孟德建
Zhang Lijun;Tang Xin;Meng Dejian(Tongji University,Shanghai 201804)
出处
《汽车技术》
CSCD
北大核心
2019年第2期14-20,共7页
Automobile Technology
基金
国家重点研发计划课题(2016YFB0100901)
关键词
驾驶意图
机器视觉
特征提取
关联性
Driving Intention
Machine vision
Feature extraction
Correlation
