A closer look at crop health
Leaf detection, disease classification, and a robot to collect images in the field. My award-winning final year project.
YOLOv5 / CNN / Robotics
The idea
Crop disease detection starts with useful images of the plants. My final year project brought image collection and disease classification into one system.
How I built it
A robotic system autonomously captures field images. YOLOv5 locates leaves, and a CNN classifies disease from the detected regions.
Where it landed
The classifier reached 96% accuracy in the project evaluation. The project received first place in the final year project competition at ITU. This result describes the project evaluation, rather than a guarantee of performance in every field environment.