Paper Info
Title
A Hybrid Method to Incrementally Extract Road Networks Using Spatio-Temporal Trajectory Data
Abstract
With the rapid development of urban traffic, accurate and up-to-date road maps are in crucial demand for daily human life and urban traffic control. Recently, with the emergence of crowdsourced mapping, a surge in academic attention has been paid to generating road networks from spatio-temporal trajectory data. However, most existing methods do not explore changing road patterns contained in multi-temporal trajectory data and it is still difficult to satisfy the precision and efficiency demands of road information extraction. Hence, in this paper, we propose a hybrid method to incrementally extract urban road networks from spatio-temporal trajectory data. First, raw trajectory data were partitioned into K time slices and were used to initialize K-temporal road networks by a mathematical morphology method. Then, the K-temporal road networks were adjusted according to a gravitation force model so as to amend their geometric inconsistencies. Finally, road networks were geometrically delineated using the k-segment fitting algorithm, and the associated road attributes (e.g., road width and driving rule) were inferred. Several case studies were examined to demonstrate that our method can effectively improve the efficiency and precision of road extraction and can make a significant attempt to mine the incremental change patterns in road networks from spatio-temporal trajectory data to help with road map renewal.
Year
DOI
Venue
2020
10.3390/ijgi9040186
ISPRS INTERNATIONAL JOURNAL OF GEO-INFORMATION
Keywords
DocType
Volume
road network extraction,changing pattern mining,K-temporal trajectory partition,spatial adjustment,k-segment curve fitting
Journal
9
Issue
Citations 
PageRank 
4
0
0.34
References 
Authors
0
8
Name
Order
Citations
PageRank
Yunfei Zhang100.34
Zexu Zhang200.34
Jincai Huang35416.88
Tingting She400.34
Min Deng55123.80
Hongchao Fan6177.44
Peng Xu700.34
Xingshen Deng800.34