AI · IoT Full-stack

Doctor-Green

Doctor-Green is a smart-farm platform that unifies an AI module diagnosing crop disease from camera video with IoT devices that measure and control the growing environment, all operated on a single web platform, so diagnosis and environment control live on one screen. I planned and built the whole pipeline solo, from sensor to dashboard. It grew out of the VGG16 plant-disease-diagnosis app that won the Excellence Award at the 2024 ICT·SW Women's Start-up Competition, and was rebuilt in 2026 as a full smart-farm platform on YOLOv8. ESP32 sensor nodes collect environment data, relay it through a Mac-based Flask server into Supabase, and a Next.js web app visualizes it live. From the same app, actuators such as LEDs and fans are controlled in real time through Supabase Realtime subscriptions. For diagnosis, YOLOv8 runs on USB-camera video to detect and diagnose 7 crop-disease classes. Training uses the Roboflow strawberry-disease-detection-dataset v4 with an augmentation strategy for class imbalance; the first run (50 epochs) reached mAP50 ≈ 0.95, and a second YOLOv8l run (150 epochs) is in progress to push accuracy further. The web app is deployed on Vercel as a live demo.