Macro-Micro

Wikimedia Commons

Macro-Micro: Education about Scale and Self-Similarity

Michelle Andrews, GIS Specialist

Overview

In collaboration with our colleagues Dr. Sam Porter and Johnnie Jaeger at AISOS (Advanced Imaging Service for Objects and Spaces), we present the Micro/Macro exhibit. Each image represents something either scanned at AISOS or remotely sensed from space. We challenge our visitors to guess the scale and possibly what or where the image represents. This exhibit showcases the exploratory tools that enhance research across disciplines in the College of Liberal Arts.

Using a Physical Drone to Control a Virtual Drone

At its core, TrashBot uses Python 3.8+ for programming language and PyTorch for its machine learning framework. TrashBot utilizes the Ultralytics YOLOv8 architecture, which is implemented in PyTorch and optimized for real-time object detection. Computer Vision Annotation Tool (CVAT) was used to annotate the training datasets sourced from TACO and Open Images Dataset V7 to fine-tuning the model. TrashBot was deployed through the CLI (MacOS) for initial prototyping and deployment. Conda and pip coupled with virtual environments were used to manage dependencies, allowing for reproducible builds across different development and deployment stages.

YOLOv5 vs YOLOv8 compared on Kaggle Trash Dataset

Project Advising

Dr. Eric Shook, Director, GeoCommons

Michelle Andrews, GIS Professional