Septoria Leaf Spot Detector (SLSD)

Introduction

Welcome to SLSD! This project is designed to detect the necrotic area on the black cottonwood leaf.

Key Features

• User-Friendly Interface: Provides an intuitive GUI, easy-to-use.
• Batch Processing Capability: Detects multiple images at once. Just select multiple images or folder.
• Time-Effective: It can detect +1,000 images within an hour.
• Comprehensive Reporting & Customization: Automatically annotates images, generates detailed CSV reports.

Installation

If using Anaconda 3, create a clean environment and activate it.
In Anaconda Prompt, type the following these steps to install the project:

# Create Conda Environment
conda create -n YOUR_ENVIRONMENT_NAME python==3.10

# Activate Conda Environment
conda activate YOUR_ENVIRONMENT_NAME

# Install the PyPI & git (if there's an error about package dependencies)
conda install pip
conda install git

# Clone the repository
git clone https://github.com/th00tames1/leafspot.git

# Navigate to the project directory
cd leafspot

# Install dependencies
pip install -r requirements.txt

How To Use

It's easy to run!

python run.py

GUI software will appear soon. It may take up to about one minute on the first run.

1. Open Images by click 'Open File' or 'Open Folder'.
   The example images are located in the 'test_images folder'. Or, you can download by use this google drive link: https://drive.google.com/drive/folders/10MSsa04RvnxVbzmnlWHY5Ggod6un33EV?usp=sharing

2. Click 'Run'.

3. The prediction result images and .csv reports will be saved.

License

Copyright Heechan Jeong (Advanced Forestry Systems Lab, OSU), 2025.

Distributed under the terms of the MIT license, Leaf Spot Detector is free and open source software.