dc.contributor.advisor |
Klette, R |
en |
dc.contributor.author |
Rezaei, Mahdi |
en |
dc.date.accessioned |
2014-07-31T05:07:15Z |
en |
dc.date.issued |
2014 |
en |
dc.identifier.citation |
2014 |
en |
dc.identifier.uri |
http://hdl.handle.net/2292/22628 |
en |
dc.description.abstract |
Computer-vision based driver assistance is an emerging technology, in both automotive industry and academia. Despite the existence of some commercial safety systems such as night vision, adaptive cruise control, and lane departure warning systems, we are at the beginning of a long research pathway toward future generation of intelligent vehicles. Challenging lighting conditions in real-world driving scenarios, simultaneous monitoring of driver vigilance and the road hazards, ensuring that the system responds in real-time, and the legal requirements to comply a high degree of accuracy, are the main concerns for the developers of any advanced driver assistance system (ADAS). The research reviews relevant studies in the past decade as well as the state-of the- art in the field. This thesis proposes some algorithms, techniques, and methodologies to address the aforementioned challenges. The first part of the work focuses on monitoring driver vigilance including classification, detection and tracking of the driver’s facial features i.e., eye status, head pose, yawning detection, and head nodding. The second part of the research mainly contributes to methods for road perception and road hazard monitoring, by introducing novel algorithms for vehicle detection and distance estimation. In the third part of the research, we simultaneously analyse the driver’s attention (in-cabin data) and the road hazards (out-road data). We apply a data fusion approach on both data sources for measuring the overall risk of driving condition, to prevent or mitigate imminent crashes, and to assist a distracted driver in a timely and efficient manner. For each stage of the research we present and discuss our experimental results, supported by benchmarks on a comprehensive range of datasets. Some of the datasets have been created in the course of this research and made publicly available. The major outcomes of this research are published in 12 peer-reviewed papers including high-impact journal articles, lecture notes in computer science, ACM, and IEEE conference proceedings. |
en |
dc.publisher |
ResearchSpace@Auckland |
en |
dc.relation.ispartof |
PhD Thesis - University of Auckland |
en |
dc.relation.isreferencedby |
UoA99264734258502091 |
en |
dc.rights |
Items in ResearchSpace are protected by copyright, with all rights reserved, unless otherwise indicated. Previously published items are made available in accordance with the copyright policy of the publisher. |
en |
dc.rights.uri |
https://researchspace.auckland.ac.nz/docs/uoa-docs/rights.htm |
en |
dc.rights.uri |
http://creativecommons.org/licenses/by-nc-sa/3.0/nz/ |
en |
dc.title |
Computer Vision for Road Safety: A System for Simultaneous Monitoring of Driver Behaviour and Road Hazards |
en |
dc.type |
Thesis |
en |
thesis.degree.discipline |
Computer Science |
en |
thesis.degree.grantor |
The University of Auckland |
en |
thesis.degree.level |
Doctoral |
en |
thesis.degree.name |
PhD |
en |
dc.rights.holder |
Copyright: The Author |
en |
pubs.author-url |
http://hdl.handle.net/2292/22628 |
en |
dc.rights.accessrights |
http://purl.org/eprint/accessRights/OpenAccess |
en |
pubs.elements-id |
448276 |
en |
pubs.record-created-at-source-date |
2014-07-31 |
en |
dc.identifier.wikidata |
Q111963636 |
|