Morehead State University
Enabling Smart Agriculture with Computer Vision
Grade Level at Time of Presentation
Senior
Major
Computer Science
Institution 23-24
Morehead State University
KY House District #
100
KY Senate District #
18
Faculty Advisor/ Mentor
Tyler Ward; Dr. Kouroush Jenab; Dr. Jorge Ortega-Moody
Department
Department of Engineering & Technology Management
Abstract
This research project focuses on the development of a computer vision application to detect diseases in five of Kentucky's most commonly exported crops: soybeans, corn, alfalfa (hay), wheat, and tobacco. Using the state-of-the-art YOLOv8 object detection algorithm, we aim to create a robust and versatile tool for disease detection and classification. To facilitate this, we curated and annotated a comprehensive image dataset composed of various disease states and healthy samples for each crop. Our long-term objective is to adapt this application for deployment on drones, which can fly over farms, capturing images and enabling rapid and widespread disease detection, offering farmers the opportunity to respond promptly to potential threats. This research represents a significant contribution to precision agriculture and crop management, addressing the need for efficient disease detection methods in a rapidly evolving agricultural landscape. The integration of computer vision and drone technology has the potential to revolutionize crop monitoring, improve yield predictions, and enhance overall farm productivity, thereby ensuring food security and economic sustainability in Kentucky and beyond.
Enabling Smart Agriculture with Computer Vision
This research project focuses on the development of a computer vision application to detect diseases in five of Kentucky's most commonly exported crops: soybeans, corn, alfalfa (hay), wheat, and tobacco. Using the state-of-the-art YOLOv8 object detection algorithm, we aim to create a robust and versatile tool for disease detection and classification. To facilitate this, we curated and annotated a comprehensive image dataset composed of various disease states and healthy samples for each crop. Our long-term objective is to adapt this application for deployment on drones, which can fly over farms, capturing images and enabling rapid and widespread disease detection, offering farmers the opportunity to respond promptly to potential threats. This research represents a significant contribution to precision agriculture and crop management, addressing the need for efficient disease detection methods in a rapidly evolving agricultural landscape. The integration of computer vision and drone technology has the potential to revolutionize crop monitoring, improve yield predictions, and enhance overall farm productivity, thereby ensuring food security and economic sustainability in Kentucky and beyond.