https://github.com/santiagoasp98/tomato-disease-classification

Tomato disease classification using deep learning with TensorFlow, Keras and FastAPI.
https://github.com/santiagoasp98/tomato-disease-classification

deep-learning disease-prediction fastapi keras neural-network python tensorflow tomato

Last synced: 4 months ago
JSON representation

Tomato disease classification using deep learning with TensorFlow, Keras and FastAPI.

• Host: GitHub
• URL: https://github.com/santiagoasp98/tomato-disease-classification
• Owner: santiagoasp98
• Created: 2025-01-13T19:55:52.000Z (5 months ago)
• Default Branch: main
• Last Pushed: 2025-01-16T21:06:53.000Z (5 months ago)
• Last Synced: 2025-01-16T21:21:55.846Z (5 months ago)
• Topics: deep-learning, disease-prediction, fastapi, keras, neural-network, python, tensorflow, tomato
• Language: Jupyter Notebook
• Homepage:
• Size: 29.7 MB
• Stars: 0
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
# Tomato Disease Classification Project

This project implements a **Tomato Disease Classifier** using a **Convolutional Neural Network (CNN)** built with **TensorFlow** and **Keras**. The classifier is capable of identifying common tomato diseases from leaf images with **quite high accuracy**.

## Snapshot

![web snapshot](snapshots/snapshot1.png "Web Application Snapshot")

## Features

- **Deep Learning Model**: A CNN trained on a labeled dataset of tomato leaf images to detect diseases with **almost 100% accuracy on the test set**.

- **Interactive Web Interface**: A clean, user-friendly frontend built with **HTML**, **CSS**, and **JavaScript**, allowing users to upload images and view predictions.

- **FastAPI Backend**: A lightweight, high-performance API backend for serving predictions.

- **Real-Time Predictions**: Users can upload an image of a tomato leaf and get instant classification results, including a confidence score.

## Technologies Used

- **Frontend**:

  - HTML, CSS, JavaScript

  - Responsive design with a clean layout

- **Backend**:

  - **FastAPI** for handling API requests

  - TensorFlow/Keras for the machine learning model

- **Deployment**:

  - Easily deployable on any platform that supports Python and FastAPI

  - Model inference optimized for fast responses

## Highlights

- **High Accuracy**: Achieved **nearly 100% accuracy** on the test dataset, making the model highly reliable for real-world applications.

- **Image Preview**: Users can preview uploaded images before making predictions.

- **Confidence Visualization**: The interface displays a confidence score for the predicted disease class.

## Future Work

- Improve web application.

- Make a mobile app that allows users to take photos with their phone.