Abstract
Keeping road network databases up-to-date is crucial to Geographical Information System (GIS) applications such as vehicle navigation. The vector road centerlines extracted from satellite images or in-car Global Positioning System (GPS) devices are likely to be inaccurate due to costly and labour intensive or long updating circle. The GPS data crowdsourced through smartphones provides an emerging source for refining road map due to its rich spatio-temporal coverage and reasonable level of accuracy. This thesis introduces an optimized methodology to automatically generate road network data from smartphone GPS data without using any reference maps. The horizontal accuracy of the extracted road centerlines, measured as a root mean square of 1.424 m and 1.252 m for curved and straight road segments respectively, is better than that of some existing road datasets. The outcome of this research will provide a new way of generating a more accurate and up-to-date road network databases.
Highlights
Herman (2002) summarized the benefits of using bidirectional road centerline such as: better matching with other data layers; providing a truer presentation of the highway network, and analyzing the road information more effectively
Due to the 2-Gigabyte memory limitation to running 32-bit PythonWin on the 64-bit Microsoft Windows 7 operating system (OS), smoothed Global Positioning System (GPS) data in the 11th tile cannot be completely processed by the later algorithm of representative point extraction, even though a patching application23 is utilized to allow the OS addressing up to 4-Gigabyte of Random Access Memory (RAM)
This new approach entails a fast and inexpensive way of updating existing road maps and refining road maps with near real-time changes
Summary
Research Objectives The ultimate goal of this thesis is to develop a cost-effective road network data extraction methodology based on GIS and crowdsourcing GPS data from smartphone users. In order to minimize the negative effects of biased GPS measurements collected by crowd-sourced smartphone users and ensure the quality of extracted road centerlines, the objectives of this research are summarized as:
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have