Coal Composition Control System

This is an implementation of the Coal Composition Control System project of the CompTechSchool 2022.

The AI-based system for the coal fraction analysis on the conveyor in real time.

Structure of this Repo

• docs : documentation
• src : main code
• scripts : scripts for data downloading
• data : downloaded data saved here
• weights : learnt weights
• train : auxiliary code

Requirements

• python3
• pip

Installation & Run

Docker (recommended):

Clone the repo and change to the project root directory:

git clone https://github.com/comptech-winter-school/coal-composition-control
cd coal-composition-control

docker build . -t coal
docker run --name coal-container -it coal

After you've done with the demo stop and remove all running containers and images:

docker system prune -a

From source

Clone the repo and change to the project root directory:

git clone https://github.com/comptech-winter-school/coal-composition-control
cd coal-composition-control

Install requirements:

pip install -r requirements.txt

And run:

streamlit run startup.py

Research notebooks

1. Semantic segmentation
2. Object detection
3. Instance segmentation with Mask R-CNN
4. Instance segmentation with Yolact
5. Classic edge detectors

Team

Pashtetickus	asromahin	0x0000dead	JI411	Dr_Under	LRDPRDX
Павел Данилов	Александр Ромахин	Сергей Ванданов	Александр Лекомцев	Ильдар Мурзагалеев	Богдан Сикач
Team Leader, Developer	Project Manager	Data Analyst, Developer	Data Analyst, Developer	Data Analyst, Developer	Technical Writer