Supervisor: Jay Yang
The utilization of social determinants of health (SDOH) has enhanced the understanding of cancer disparities, yet the integration of urban-environmental characteristics at the neighborhood level remains partial. Satellite imagery, as a readily available data source, facilitates the observation and quantification of urban geographical contexts and their relation to environmental, socio-demographic, and health processes. Traditional deep learning (DL) methods, particularly Convolutional Neural Networks (CNNs), have been employed to analyze high-resolution images, focusing predominantly on local contextual information.
Team member: Zongrong Li, Qiluo Li, Haiyang Li
This project presents a detailed analysis of the 2023 wildfires in Hawaii, particularly focusing on Maui Island, with the goal of developing a cost-effective and efficient wildfire assessment model.
Methodology: three-tiered approach, initiating with broad-scale fire detection using high-frequency FIRMS data at approximately 1-kilometer resolution. It then progresses to a more refined analysis using high-resolution Sentinel 2 data (20-meter resolution with a five-day revisit cycle) and Planet data (3-meter resolution), specifically targeting the evaluation of vegetation cover impact, employing the NDVI index for accurate and timely updates.
Supervisor: Zhiwei Li
Urban flooding increasingly threatens megacities, especially in low-income countries. Despite the gravity of the issue, most flood risk studies have predominantly focused on high-income regions like the European Union and the U.S., leaving developing nations relatively under-examined.
Many of these studies hinge on census data combined with land cover information, relying on results from flood inundation models that have inherent uncertainties. Moreover, the majority of research has a localized focus, often overlooking broader patterns that could emerge from comparing cities at different developmental stages.