Hello! For my second interview in my CS Students Interview Series, I talked to Audrey Ha, a winner of the 2020 Congressional App Challenge (CAC). The CAC is a prestigious prize that students can win by designing, creating, and coding an app that fits the district-specific challenge. The CAC was created by members of the US House of Representatives with the goal of teaching middle and high school age students how to code. For more information, check out the CAC website: https://www.congressionalappchallenge.us/about/.
Audrey Ha heard about this challenge through her local high school’s AP Java teacher. For the challenge, she focused her app on hurricane relief for two main reasons, the first being that the 2020 hurricane season was a record breaking one. Second, she thought that the speed and accuracy of machine learning could help the US organize relief efforts for natural disasters. Her award-winning app, SurveyHurricane, uses artificial intelligence to accurately and quickly detect damaged houses on aerial imagery of storm-impacted regions. Her app uses object detection and image classification neural networks to locate these damaged houses.
It took Audrey 2 months to create Survey Hurricane as there was a lot of work that went into it. In order to build her app, she used open-source storm aerial imagery from the National Oceanic and Atmospheric Administration as her database. She then trained a Faster R-CNN model from the Google TensorFlow API to locate houses on aerial images. She also trained an Inception ResNet V2 model to classify whether each house is intact or damaged. Last, Audrey integrated both of these models into a web app that utilizes the Flask Python microframework. To watch a demo of SurveyHurricane, feel free to check out this video: https://www.youtube.com/watch?v=pWTADXffilE.
When I asked Audrey about why this award was important to her, she told me that she felt very honored as she is deeply passionate about applying machine learning to large scale problems in the community. In the future, she hopes to continue to use technology in a way that has a positive impact on society. Audrey is currently working on databases and computer vision for the Veterans Administration Palo Alto. In college, she plans to study computer science, specifically artificial intelligence, databases, and app development. I look forward to catching up with Audrey in the future!