University of Technology Sydney
Exploring pedestrian behavior through deep learning methods in smart cities
Abstract
dc:description.abstractPedestrian detection, pedestrian behavior recognition, and pedestrian trajectory prediction are three integral components that play crucial roles in ensuring road safety and optimizing traffic management within smart city infrastructures. Precise pedestrian detection can effectively prevent traffic accidents and ensure pedestrian safety. Behavior recognition allows for the monitoring and analysis of pedestrian dynamics, providing real-time risk alerts. Trajectory prediction, on the other hand, enables the anticipation of pedestrians’ future movement paths, assisting autonomous driving systems in decision-making. However, these tasks face numerous challenges in complex real-world environments, such as variations in lighting, the complexity of pedestrian behaviors, and the uncertainty of trajectory predictions, which continue to pose challenges to the overall efficacy and safety of intelligent transportation systems. This research introduces improved deep-learning approaches to address these challenges, contributing to safer and more efficient monitoring environments in smart cities.
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Zhao, Rui
Rights
dc:rights.accessrights- Statement dc:rights.accessrights
-
- open_access
- info:eu-repo/semantics/embargoedAccess
- The author owns the copyright in this thesis including all reproduction and reuse rights for the work. The work may not be altered without the permission of the copyright owner. Attribution is essential when quoting or paraphrasing from this thesis.
- © 2024 Rui Zhao
- au.edu.uts.lib/nph
- Language dc:language.iso
- en_US
Identifiers
dc:identifier.*- Handle dc:identifier.uri
- http://hdl.handle.net/10453/186183