AI Plant Disease Detection

This repository showcases a Vue.js frontend and a Python(flask) backend for detecting plant diseases using a Convolutional Neural Network (CNN). The frontend allows users to upload images of plant leaves, while the backend handles inference using a trained model.

---

Table of Contents

1. Overview
2. Project Structure
3. Dataset
4. Prerequisites
5. Installation
6. Usage
   • Running the Frontend
   • Running the Backend
7. Model Details

---

Overview

• Goal: Detect plant diseases from leaf images using a trained CNN model.
• Frontend: Built with Vue.js (Vite as the build tool).
• Backend: Python (Flask in app.py) for model inference.
• Machine Learning: A Jupyter Notebook (model.ipynb) for training the CNN model.

---

Project Structure

1.  AI_plant_disease_detection
    ├── node_modules/                 # Node.js dependencies
    ├── src/
    │   ├── components/
    │   │   └── PlantDiseaseDetection.vue  # Vue component for disease detection UI
    │   ├── App.vue                        # Root Vue component
    │   ├── index.html                     # Main HTML template for Vite
    │   └── main.js                        # Entry point for Vue application
    ├── app.py                         # Python backend for model inference
    ├── model.ipynb                    # Jupyter notebook for training the CNN model
    ├── package.json                   # Node.js project configuration
    ├── package-lock.json              # Auto-generated file for exact dependency versions
    ├── Procfile.txt                   # Deployment configuration (e.g., Heroku)
    ├── README.md                      # Project documentation (this file)
    ├── requirements.txt               # Python dependencies
    └── vite.config.js                 # Vite configuration
   
   

---

Dataset

To train or retrain the CNN model, you need a labeled dataset of plant leaf images. One popular dataset is available on Kaggle. Steps:

Download the dataset from Kaggle.

Unzip and place the images in a suitable directory (e.g., data/) at the project root.

Update any file paths in model.ipynb as needed.

Note: The dataset is not included in this repository. You must download it separately from Kaggle.

---

Prerequisites

Node.js (version 14+ recommended)

Python 3.7+

pip (Python package manager)

Virtual environment tool (optional but recommended)

---

Installation

1. Clone the Repository

    git clone https://github.com/vara-prasad-07/ AI_plant_disease_detection.git
    

---

2. Install Node.js Dependencies

    npm install
   

3. Install Python Dependencies

     pip install -r requirements.txt
   

---

Usage

1. running frontend:

     npx vite
   

This starts a local development server. By default, open http://localhost:5173 (or the URL shown in the console) in your browser.

---

Running the Backend

1. Ensure your virtual environment is active (if used).

2. start python server:

   python app.py
   

This will run a Flask (or similar) app on a local port (e.g., http://127.0.0.1:5000).

3. Check API Endpoints (if any):

For instance, POST /predict might be where images are sent from the frontend.

---

Model Details

Architecture: Convolutional Neural Network (CNN) with multiple convolution, pooling, and dense layers.

Frameworks:

1. Python: TensorFlow/Keras (or PyTorch, depending on your setup)

2. Vue: Composition API or Options API (depending on your preference)

3. Training:

Use model.ipynb to train or fine-tune the CNN.

Hyperparameters (learning rate, batch size, epochs) can be modified in the notebook.