IEEE Geoscience and Remote Sensing Speaker Series Presents Dr. Gordon Christi, JHU/APL

November 7^th, B-11 Auditorium at the new CCRF Lockheed Facility; 11am lunch provided in the social hub outside the auditorium, 12-12:30pm talk (no food allowed in the B-11 auditorium).  Please RSVP by Nov 1 if you would like lunch to be provided.

Abstract: On September 2018, Hurricane Florence hit the Carolinas. More than a million people were forced to evacuate as rainfall hit and the wind speeds amplified. In response, the National Guard immediately started to collect imagery from airborne platforms and used it to manually identify flooded areas and damaged infrastructure. JHU/APL was tasked to accelerate the processing of airborne and spaceborne imagery using computer vision and machine learning to identify flooded areas. Since then, JHU/APL has been extending these capabilities to include building damage assessment and road extraction. The deep learning models trained for these tasks were recently used during JHU/APL’s assistance in the response to Hurricane Dorian. In this talk, Gordon Christie, the algorithms lead for the project, will provide technical details of the approaches and the lessons learned from working with first responders.

Bio: Gordon Christie received his B.S. in Computer Engineering from West Virginia University in 2011, and his M.S. and Ph.D. degrees in Computer Engineering from Virginia Tech in 2013 and 2016, respectively. Since graduating, he has spent over 2.5 years at The Johns Hopkins University Applied Physics Laboratory (JHU/APL) as a senior staff member. Gordon’s research has focused on computer vision (CV), machine learning (ML), remote sensing, and autonomous robotic systems, with a more recent focus on public health and disaster relief problems. Gordon has also been involved in several public challenges led by JHU/APL focused on CV/ML for overhead imagery, including the 2019 IEEE GRSS Data Fusion Contest, IARPA’s Functional Map of the World, and the Urban 3D Challenge.

Relevant articles:

https://www.jhuapl.edu/PressRelease/190926

https://www.jhuapl.edu/PressRelease/190927