Current Awardee

Air Pollution Prediction using Traffic Surveillance Camera Footage and Deep Learning

Mei Tessum
Mei Tessum

College: Agricultural, Consumer and Environmental Sciences
Award year: 2021-2022
NCSA collaborators: Volodymyr Kindratenko, Dawei Mu

Traffic-related air pollution is a major health burden in the United States. "Hyperlocal" pollution quantification is increasingly recognized as important for environmental sustainability and social equity. However, measuring pollution with traditional direct-measurement techniques at the required levels of granularity is not feasible as these methods do not scale well and have high demand in cost and labor. The team proposes to create a system that uses traffic camera footage and deep learning to predict traffic-related pollution concentrations. The initial results produced by the proposed research will help to establish a larger project in cooperation with traffic surveillance networks in multiple metropolitan areas. The proposed project represents a step toward a scalable system for hyperlocal pollution estimates.