Zawala / jetson-plant_disease_detection

Disease Identification using ZEDD Stereo Camera Dual Vision

Overview

This codebase uses Keras, TensorFlow, and PyTorch to scan a video feed from the ZEDD stereo camera dual vision and identify diseases.

Installation

On Linux PC

1. Create a virtual environment in the code folder: virtualenv env or python3 -m venv env.
2. Activate the virtual environment: source env/bin/activate.
3. Install the required packages: pip3 install -r requirements/requirements.txt.
4. Run the code: python3 sharingan.py.

On Jetson

Disease Identification using ZEDD Stereo Camera Dual Vision

Overview

This codebase uses Keras, TensorFlow, and PyTorch to scan a video feed from the ZEDD stereo camera dual vision and identify diseases.

Installation

On Linux PC

1. Create a virtual environment in the code folder: virtualenv env or python3 -m venv env
2. Activate the virtual environment: source env/bin/activate
3. Install the required packages: pip3 install -r requirements/requirements.txt
4. Run the code: python3 sharingan.py

On Jetson

1. Do not create a virtual environment as it will cause core dumps. Instead, just install Python3.
2. Install the required packages: pip3 install -r requirements/requirements_jetson.txt

Usage

Training a Neural Network

1. Create folders with names like categories.json
2. Insert the correct pictures in the folders
3. Run train.py or tomato-leaf-disease-classification.ipynb (after populating labeled data as in the notebook)

For help with the notebook, visit https://github.com/divyansh1195/Tomato-Leaf-Disease-Detection-.git to train data for use with ZEDD.

Testing the Trained Model

To test the trained model, load the Keras model into the Flask app in line 37, then start the Flask app to use test pictures by uploading and evaluating results.

Logs

[Include logs or output from the project, if applicable.]