Tomato Leaf Disease Classification Web App

Overview

The dataset, which consists of 8 categories of tomato leaves from PlantVillage were used as dataset to train the Keras model. It was then converted to a TensorflowJS model and deployed using a NodeJS Express server on Digital Ocean. Solutions to remedy the diseases are provided as well.

Setup

1. Clone this repository into your local machine
2. Install NodeJS
3. In the local directory (/tomato), open Terminal and run the following:

npm install
node server.js

Built With

• Keras - High level neural network API, running on top of Tensorflow
• TensorflowJS - Library for machine learning in Javascript
• ExpressJS - Minimalist web framework for NodeJS
• Digital Ocean - Cloud hosting provider

Acknowledgements

Sincere acknowledgements to Mr Gary Chan (project facilitator), Dr Jimmy Goh (AI expert), Ms Grace Yap (Temasek Foundation Scale Organizer), all other lecturers of Republic Polytechnic who helped in any part and my teammates.

• SP Mohanty - PlantVillage Dataset
• DeepLizard - TensorflowJS tutorial
• Traversy Media - ExpressJS tutorial
• Plant Natural - Solutions for tomato leaf diseases