Integration of Social Media and Unmanned Aerial Vehicles (UAVs) for Rapid Damage Assessment in Hurricane Matthew

Recent studies stress the critical role of rapid damage assessment in crisis management to reduce losses in natural disasters. Timely damage-related data collection by traditional empirical inquiry methods (e.g., interview and survey questionnaire) is particularly difficult after natural disasters such as hurricanes and floods. Remote sensing technologies including satellite and unmanned aerial vehicles (UAVs) were popularly employed in previous research in support of damage data collection. Compared with orbit limitation and high operating costs of satellites, UAVs demonstrated its advantages in low operating costs, high operational flexibility, and high spatial resolution of imagery. However, using UAVs alone in data collection for major disasters such as Hurricane Matthew is particularly difficult because the affected area was so extensive. On the other side, previous studies indicate that social media (e.g., Twitter) was widely used as a crowdsourcing platform for citizens’ communication and information sharing during natural disasters. Furthermore, citizens can also communicate information updates related to damage (e.g., burned area of wildfire) in natural disasters. Hence, social media presented its potential in providing timely information to support damage assessment. To conduct a rapid damage assessment, this paper proposed the framework of integration of Twitter and UAVs for damage data collection for Hurricane Matthew in Florida. Firstly, this study used real-time Twitter data to prioritize the affected counties requiring the deployment of UAVs. Secondly, we proposed the employment of UAVs in imagery and video data collection in these affected areas at the city and region level. Lastly, processing techniques for image and video were applied for rapid damage assessment, which can be further geographically represented to support crisis management.