Case Study: Deep Learning World achieves rapid, automated post-disaster building damage assessment from satellite imagery with REI Systems

Deep Learning World - Customer Case Study

Deep Learning World confronted the need to quickly and accurately assess post‑disaster damage over large areas to support response and recovery efforts. REI Systems showcased a solution using semantic segmentation and custom deep learning models applied to before-and-after satellite imagery to automatically detect buildings and identify damage, reducing dependence on slow, manual aerial or in-person surveys.

REI Systems implemented an automated pipeline that compares pre- and post-event satellite images to map damaged neighborhoods and individual structures, enabling much faster identification and prioritization of response, funding, and insurance adjustments. The solution cuts down on human review time and was presented by REI Systems as a case study at Deep Learning World Las Vegas 2020.

Deep Learning World