This project aims to classify plant diseases using deep learning models. It utilizes the PlantVillage dataset from Kaggle to train models for identifying diseases in plant leaves.
- Download the dataset from Kaggle - PlantVillage Dataset.
- Extract the downloaded data to a folder named "PlantVillage".
- Activate the base environment or create a virtual environment.
conda create -p venv python==3.8 -y
- Install the required dependencies using
requirements.txt
.pip install -r requirements.txt
- Refer to the presentation in the [PPT and Video](PPT and Video) folder for an overview of the project.
- Follow the notebooks in the following sequence:
Split_PlantVillage_Data.ipynb
: Split the dataset into train, test, and validation sets.EDA.ipynb
: Explore the data distribution.- Model Training Notebooks:
- Save the finalized models in the
saved_models
directory with naming conventions:PepperLeaf_Model.h5
,PotatoLeaf_Model.h5
,TomatoLeaf_Model.h5
in their respective plant folder.
- The
ui
directory contains static and template folders for a user-friendly web interface of the app. - Resources folder includes the
model_operations.py
module with necessary functions used inapp.py
. - Test your code using
test.py
in Resources before running it in the app.
- Activate conda environment
conda activate
with requirements installed already. - Run the
python app.py
code in the terminal to start the web page. Access the app by copying the provided localhost link.
- Explore alternative models beyond CNNs.
- Expand the project to include more plant species and fine-tune models for specific plants.
- Enhance production-level deployment and incorporate CI/CD pipelines for automated testing and deployment.
- Collaborate with experts for domain-specific insights and continuous data collection.
This README file provides a step-by-step guide to setting up the project, running the code, and outlines future directions for the project's expansion and improvement.