Automated Animal Classification using LiDAR
Led by Bing Ouyang, Ph.D.
Affiliated Home Campus: Harbor Branch
Affiliated Department: Harbor Branch Oceanographic Institute
REU Scholar: Leia Rich
REU Scholar Home Institution: Swarthmore College
Marine hydrokinetic (MHK) installations harvest energy through ocean currents and waves. However, it is essential such installations “do no harm” – avoid harming protected marine animals. This issue can be mitigated through installing a system to monitor the surroundings of the installations. The unobtrusive multistatic serial LiDAR imager (UMSLI) has several advantages over most widely used acoustic and camera systems, including not using visible artificial light that would attract marine life during nighttime operations. This project focused on improving a simulation code for the serial laser imaging system which can be used to predict the system’s performance in an underwater environment. The simulation code was expanded to replicate the real-life setup with the receiver array. This model can be used to generate the training dataset to improve the robustness of the UMSLI classifier.
Click here to watch the student presentation.