This project is for a class called "Modeling with Drones 2". It uses drone acquired video, drone flight logs, and other assets to detect, track, and segment individual vehicles in a parking lot, read their license plates, and validate that they have a parking permit that is appropriate for the parking space. It uses YOLOv8, EasyOCR, OpenCV, and GeoPandas for the core functionality.
More details are descried in a storymap entitled Information Instead of Imagery.
This project runs on one particular dataset, and I haven't included the video and log files of that dataset in the repo. Perhaps someday I'll finish the project and make it more general purpose, but probably not.