Automatic Fruit Counter

Automatic fruit counter is web projects that classify and automatically calculate fruits.

1. Image file

2. Webcam

🍌Index

🍓Architecture
🍓Frontend
🍓Backend
🍓Contributors
🍓Framework
🍓Docker Prerequisites

🍓Architecture

🍏 Frontend : React
🍏 Web Server : Nginx(Reverse Proxy)
🍏 WSGI : gunicorn
🍏 Application Server : Django
🍏 DataBase : MongoDB

🍓Frontend

🍍Design

🖍Design Document

🍍Structure

🍏 font_end/src/pages : Directory for pages
🍏 front_end/src/pages/components : Directory for components used on multiple pages
🍏 front_end/src/pages/logo : Directory for logos

⬆️ Back to Top

🍓Backend

🍍Object Detection Model

🍏 YOLOv5
🍏 YOLOv5 github
🍏 DataSet : 6,000 images / 63 classes / 16,830 annotations

🍏 Training result

metrics/precision	metrics/mAP_0.5
metrics/mAP_0.5:0.95	metrics/recall

	obj_loss	box_loss	cls_loss
Train
Validation

base model : YOLOv5s base model : YOLOv5m

⬆️ Back to Top

🍓Contributors

Name	Country	Where to find you	Role
MinJae Kim	Republic Of Korea	Github	Machine Learning
Chunghwan Kim	Republic Of Korea	Github	Back End, Devops
Chaeyeon Song	Republic Of Korea	Github	Front End
SiEun Lee	Republic Of Korea	Github	Front End
Dahyeon Jeong	Republic Of Korea	Github	Front End

⬆️ Back to Top

🍓Framework

• ReactJS
• Django

⬆️ Back to Top

🍓Docker Prerequisites

• Windows https://docs.docker.com/docker-for-windows/install/
• Mac https://docs.docker.com/docker-for-mac/install/

1. git clone

git clone https://github.com/2021SiliconValleyOnlineInternshipTeam-J/FruitBilling.git

2. docker compose build

docker-compose build

3. docker compose up

docker-compose up

⬆️ Back to Top