Analyze the Risk of COVID-19 in Crowded Locations

This project uses computer vision to analyze visual data (image or video) captured in public locations in order to estimate the density of crowds and responses to governments' policies of social distancing.




The COVID-19 pandemic has more than four million infections and 300,000 deaths as of mid May 2020 worldwide. In response to the pandemic, many governments have imposed “social distancing” policies and restricted “non-essential” activities. As a result, public locations witness significant decreases of crowds. Moreover, many governments recommend wearing masks in public locations. This project uses publicly available visual data to understand the changes in density, usage, and whether people are following recommended guidelines over time.

This project is supported by the National Science Foundation OCA-2027524. The team analyzes only aggregated information and does not recognize any individual for privacy protection. The project follows Purdue Institutional Review Board's protocol 2020-460.





The members are expected to have finished one semester of calculus and one programming course.


Meeting Times:

  • Summer 2020: Monday and Wednesday 10:00 - 10:50 am EST, online
  • Fall 2020: TBD (online)