CarbonTrace

CodeUtsava 8.0
Millenium Falcons

Overview

As the Millennium Falcons team, we’re more than just problem-solvers. Together, we're taking on the challenge of coal theft in the mining industry, combining our unique strengths and perspectives to tackle this head-on. The mining industry endures major financial setbacks due to coal theft, especially during transport. Thieves often mask the stolen coal by adding water to maintain the overall weight, making it difficult to detect theft using traditional methods. These conventional monitoring systems, typically manual or limited in scope, fall short in addressing the complexities of theft prevention. The lack of real-time tracking, load integrity monitoring, and anomaly detection results in unchecked inefficiencies and poor accountability.

Note

Key Features:

• Real-time Moisture Fluctuation Tracking: It constantly monitors the moisture of the coal. If it exceeds the base level of the moisture, it will send an alert notification both in the train and the server

• Real-Time Anomaly Detection: It uses opencv to detect faces of anomalies. If detected, then it again sends an alert notifications to both train and the server

• Cetralized Server for Surveillance: The centrlized Surveillance server is used to recieve the alert notifications, along with daily basis transportation data analysis

Getting Started

Cloning into local mahine

To clone this repository into the local machine, you need to follow the following steps use git to clone the repo

git clone "https://github.com/Millenium-Falcons/CarbonTrace.git"

navigate into the clone directory CarbonTrace

cd CarbonTrace

And you can use and inspect the individual piece of codes inside that directory

Frontend

Avirup Ghosal is responsible for the creation and maintanace of the whole frontend. The website is written using a javascript framework React.js. Deployed link: https://coal-trace.vercel.app

For getting started with the frontend, please read the following instructions: Deployed address:

cd CarbonTrace
cd frontend
npm run dev

Backend

Hritesh Saha is responsible for the creation and maintanace of the whole backend for this project. It is written in a node.js framework called Express.js, and it serves as an medium for the frontend, the hardware and the machine learning to communicate.

For setting up the backend, please read the following instructions

cd CarbonTrace
cd Server
node index

Machine Learning

Debjeet Banerjee is responsible for the creation and maintanace of the Machine Learning models. It is written in python is using OpenCV and Scikit-Learn Logistic Regression. It is responsible for processing real-time video for anomaly detection and alerts the user whenever an anomaly is detected. For setting up the GenAI, please read the following instructions:

Navigate in the cloned repository and go inside the machine_learning directory

cd CarbonTrace
cd machine_learning

Install the required modules using the following command for Windows

pip install -r requirements.txt

for Linux

pip3 install -r requirements.txt --break-system-packages

execute the the python code for anomaly detection using OpenCV using the following commands

python opencv_face_detection.py

Contributors

Builders	Tech Stack
Member Role Branch Debjeet Banerjee ML & Hardware ML Hritesh Saha Backend Backend Sayantana Halder Hardware ESP Avirup Ghosal Frontend Frontend	Role Tech used ML/DL Backend Frontend Hardware Deployment

Copyright © 2024 Millenium-Falcons