Skip to content
This repository has been archived by the owner on Apr 4, 2024. It is now read-only.

Viibrant/automated-checkout

Folders and files

NameName
Last commit message
Last commit date

Latest commit

ย 

History

60 Commits
ย 
ย 
ย 
ย 
ย 
ย 
ย 
ย 
ย 
ย 
ย 
ย 
ย 
ย 
ย 
ย 

Repository files navigation

โš ๏ธ Archive notice ๐Ÿ“š 2024/04

This repo contains my A Level Computer Science coursework.

๐Ÿ”ง I used YOLOv3, then SOTA pretrained object detector, for an automated checkout. It served as an initial foray into applying neural networks within a practical, albeit simplified, retail setting. This project remains as both a record of my early exploration into machine learning, yet also as simple nostalgia.

๐Ÿง  I learnt a lot from this project. The bit I was particularly proud of was getting the frontend websocket code to work, pulling data streamed from the backend and dynamically creating a table with the detected objects. It was my first time working with generator functions, which I hadnโ€™t even heard of before diving into this. Also, this was my first shot at putting together a system that used different programming languages, which I found quite exciting.


Automated-Checkout

Main repository for this project, built using Flask, ImageAI.

Execution

To run (pretty obvious but still):

./start.sh

This will start a Flask web server run locally on your computer that you can connect to, the URL should be outputted to the terminal.

Project consists of web.py, a script that runs the web server itself. Here we find the code instantiating the camera declared in camera.py and the Neural Network declared in, you guessed it, neuralnetwork.py.

Dependencies

pip install tensorflow opencv keras flask flask_socketio imageai pillow numpy

Basically you may as well just install Anaconda that'll probably sort you all out considering