Tomato leaf disease detector

The brief

Tomato farmers lose crops to diseases they spot too late. Build an image classifier that recognises common tomato leaf diseases from a photo and gives the recommended treatment. Good looks like: high accuracy on held-out photos, a test on photos taken in real fields, and a small app.

What you will build

1. Prepare and augment the leaf images
2. Train a transfer-learning CNN
3. Test on photos with real backgrounds
4. Make the model small enough for a phone
5. Build an app that shows the disease and treatment

Free, open data to build it with

• Tomato leaf disease detection ↗ Kaggle
• PlantVillage Dataset ↗ Kaggle

Check each dataset's licence before you publish your work. Your mentor gives your team its own version of the brief (data, scope or region), so every team's project is its own.