Date of Award
Open Access Dissertation
Civil and Environmental Engineering
Increased vulnerability of water systems to extreme events and climate change is among the profound challenges facing the management of water resource systems around the world. Extreme events, including droughts, floods, and natural hazards have become more frequent and intensive, particularly in coastal regions. Floods, for instance, caused tens of billions of US dollars losses and put the lives of thousands in danger, globally. To cope with the adverse consequences of floods, a wide range of structural, non-structural, and emergency measures are studied and deployed by flood management sectors. Various flood simulation, mapping, and forecast systems have been developed to predict flood events, warn the public and inform decision-makers to react accordingly for making better decisions to protect lives and property. Development and level of accuracy of these systems, however, rely heavily on the availability and quality of temporal and spatial data received from ground-based gauge sensing, remote sensing, and more recently crowdsourcing. While all these data sources provide useful information, they have their limitations, such as the small spatial scale of in-situ gauging or satellites' long revisit period. Recent advancements in Artificial Intelligence provide a unique opportunity to gather and analyze complementary flood information from new and unconventional sources of data and accelerate flood modeling. This research aims to introduce a vision-based framework using surveillance imageries to enhance the monitoring, modeling, and management of water resources, using Computer Vision and Deep Learning techniques. The proposed framework interprets the visual features of the captured images into water-related numerical parameters, such as water level and inundation area. Such a framework will support flood modeling by providing real-time input information for data assimilation and inform decision-makers and first responders to undertake appropriate actions and adaptation strategies facing flood risk.
Erfani, S. M.(2023). Developing a Vision-Based Framework for Measuring and Monitoring Water Resource Systems Using Computer Vision and Deep Learning Techniques. (Doctoral dissertation). Retrieved from https://scholarcommons.sc.edu/etd/7444