Tomato Disease Detector

The repository is a Detector project that allows you to easily detect tomato's disease using Simple Web services. Currently, a total of 7 classes of disease can be detected with bounding box.

• Image file available
• Video file available
• New yolo models can be added
• Other format models can be added

Table of Contents

• Structure
• Usage
• Examples
• Team
• License

Structure

• Dataset: https://www.aihub.or.kr/aidata/129
• Model
  • Detection
  • Yolov5 Github : https://github.com/ultralytics/yolov5
• Tool
  • Frontend : Streamlit - https://docs.streamlit.io/en/stable/#
  • Backend : FastAPI - https://fastapi.tiangolo.com/
  • Annotation Tool : CVAT - https://github.com/openvinotoolkit/cvat

Usage

1. Clone This Repository

   $ git clone https://github.com/IVADL/PDD-prototype.git

2. docker-compose commands

   $ docker-compose build
   $ docker-compose up

3. Visit Streamlit UI

• visit http://localhost:8501

4. Run model

• Select model : yolov5
• Test Image or video upload
• Click 'Detect Plant Disease Button'

Example

1. Image Detect
2. Video Detect

Team

The project was conducted at the Korea Lab of Artificial Intelligence and formed a team called IVADL.

• Harim Kang Git-hub
• Seonmin Kim Git-hub
• Eunbi Park Git-hub

License

MIT