Loading...
Development of fixed-site photogrammetric applications and optimization for slope hazard monitoring
Gray, Brian Z.
Gray, Brian Z.
Citations
Altmetric:
Advisor
Editor
Date
Date Issued
2021
Date Submitted
Collections
Research Projects
Organizational Units
Journal Issue
Embargo Expires
Abstract
Geologic hazards present a growing concern for the safety of humans and the development of infrastructure. There are a variety of remote monitoring methods utilized to monitor these hazards for risk analysis and mitigation purposes. Of them, fixed-site photogrammetry is a relatively new and, therefore, underutilized remote monitoring method in geologic contexts that offers time efficient, low-cost, high temporal resolution data collection with fit-to-purpose accuracy. The development of applications and practices surrounding fixed-site photogrammetry is thus valuable for the advancement of hazard monitoring more generally.
The full capabilities of the high temporal resolution of fixed-site photogrammetric monitoring are currently unexplored in geologic contexts. Other industries have made many advances in utilizing 3D data and photogrammetry, coupling the data with other advances in machine learning, numerical modelling, and statistical shape modelling. Slope hazard management can greatly be assisted via the adoption of these emerging applications in a geologic engineering context, and potential avenues for adoption are evaluated in this thesis. In particular, a proof-of-concept methodology for the prediction of rockfall event volumes was developed, with demonstrated potential to predict volumes given limited surficial information, such as that available prior to failure.
Research into best practices in the design of fixed-site photogrammetric sites remains limited. Utilizing a combination of theoretical fundamentals and empirical data collected from Red Rocks Park, Colorado, US., this work has developed a framework to enable performance predictions of fixed-site photogrammetry, which can be useful both in the design of a fixed-site system and in comparative analyses between remote monitoring methods. Empirical trends correlating ground pixel size and model accuracy were identified. Comparisons to an active, fixed-site photogrammetric system were made to evaluate the performance of the framework. Overall, the approach performs adequately for initial performance prediction and for use in ‘go/no-go’ analysis.
Associated Publications
Rights
Copyright of the original work is retained by the author.