摘要
由于现有行人路网的基础数据中缺失属性信息,造成行人导航无法满足个性化需求,影响了行人出行的舒适度与安全性。将路网分为平路与坡路两类,提出了一种基于智能手机的行人路网坡度属性检测方法。首先,基于智能手机传感器数据,采用机器学习算法实现对坡度属性的检测。其次,通过数据融合与地图匹配,得到带有属性信息的行人路网数据。最后,对该数据投票与修正并进行可视化。方法最终达到了97.3%的属性检测精度,表明了本文方法的有效性。所得路网数据可为个性化导航提供数据基础。
Due to the lack of attribute information in the existing basic data of pedestrian network,pedestrian navigation can’t meet the personalized needs,which affects travel comfort and safety.In this paper,the road network is divided into flat road and slope road.In addition,a method of detecting slope attribute of pedestrian road network based on smartphone is proposed.Firstly,machine learning algorithm is used to detect the slope attribute based on the sensor data.Secondly,,the pedestrian network data with attribute information is obtained through data fusion and map matching.Finally,the data is processed with voting and correction and the results are visualized.The detection accuracy achieves97.3%,which proves the effectiveness of the proposed method.What’s more,the obtained road network data can provide data basis for personalized navigation.
作者
雷霞
周宝定
LEI Xia;ZHOU Baoding(Zhuhai Institute of Urban Planning&Design,Zhuhai 519000,China;College of Civil and Transportation Engineering,Shenzhen University,Shenzhen 518060,China;Institute of Urban Smart Transportation&Safety Maintenance,Shenzhen University,Shenzhen 518060,China;Guangdong Key Laboratory of Urban Informatics,Shenzhen University,Shenzhen 518060,China)
出处
《测绘地理信息》
CSCD
2022年第6期65-69,共5页
Journal of Geomatics
基金
国家自然科学基金(41701519)
深圳市科技计划项目(JCYJ20180305125058727)
广东省自然科学基金面上项目(2019A1515011910)。
关键词
行人路网
属性检测
机器学习
智能手机
pedestrian network
attribute detection
machine learning
smartphone
