Randall Pietersen

Randall (Randy) Pietersen is a PhD candidate whose research is focused on the development of new deep learning (DL) tools and methods for airfield damage assessment. Specifically, Randy’s work is centered on pavement structures and the development of novel, drone-based, remote sensing technologies as a safer, faster alternative to manual, human assessment. As a MathWorks Fellow, Randy will expand his work in the use of drones and hyperspectral image interpretation in the detection and classification of pavement damage and unexploded ordnance. This project will involve creating and characterizing the performance of new convolutional neural networks (CNN) that will reliably classify hyperspectral image data in real time. Randy relies on MATLAB tools to train, test, and compare CNN models and to address the most complex problems in autonomous pavement inspection and UAV-based hyperspectral remote sensing. His work has the potential to strengthen the field of DL research and to protect lives by improving the safety and efficiency of airfield operations for the United States Air Force.