For my undergraduate senior thesis project at the University of Hawaii at Hilo, I trained a machine learning model based on the YOLO v5 object detection algorithm.
The goal for my project was to create a machine learning model that could successfully identify grazing behaviors of the yellow tang (Zebrasoma flavescens). Working with footage from the MEGA Lab Camera Livestream, I was able to sucessfully create a model which could detect grazing behaviors in these reef surgeonfish. Most of the code was written in Python with additional data analysis taking place in R. Image data was annotated using the online Roboflow machine learning toolbox. The project took place over my senior year of study in the marine science department at UH Hilo. The corresponding write-up can be found here and here is an example video analyzed by the model: