GitHub - MehtaAkshita/BC71_FORTIFY

BC71_FORTIFY - SIH_2020 - RKGIT - Ministry of Housing & Urban affairs, Gov. of India

[] [] []

Overview

Our cities are expanding and most of the people have to travel more than 2 hours daily to commute between their place of work and stay. How can we reduce this time?

Project information ℹ️

Team Name: Fortify
Problem Statement Number: BC71
Ministry: Ministry of Housing and Urban Affairs
Short Project Description: A deployable solution for many major road and traffic problems faced in day to day life.

Project Modules	Description
Congestion Prevention	Computer Vision based approch to control traffic lights based upon density calculation of vehicles
Facial Wallet for Public Transport	Faster checkin based for public transport system using facial recognition
Rash Drive Detection	Computer Vision based approch to detect vehicles who are constantly changing lanes
Emergency Vehicle Priority	Alert based system to notify vehicles in front of emergency vehicles to keep a lane empty
Pothole Detection	Using Yolov2 to idenetify potholes on the road and notify concerned authorities for the fix

Team Name: FORTIFY
Team Members: _ Akshita Mehta @MehtaAkshita, Anshuman Aggarwal @anshuman73, Naman Bansal @psifrous, Shreyas Papinwar @shreyaspapi, Swapnil Panwala @panwalas, Zubin Choudhary @Iamzubin
Repository Link: https://github.com/MehtaAkshita/BC71_FORTIFY
Labels: ML, Image & Video Processing, Facial recognition, Android App, Web Development

Platforms & Technologies used 📦

Backend

Yolov2
OpenCV
Keras
TensorFlow
Flask
Dlib

Frontend

Dash by Plotly
JavaScript
HTML
CSS
Android Studio

Database

Firebase

API's

Plate Recognizer API
Way2Sms API
Google Maps API

Project Synposis ⚛️

Connected Cars is an initiative to improve traffic conditions, shorten duration of time taken by emergency vehicles to reach desired destination, avoiding potholes and crashes along with making cities smart in an efficient and effective way by making vehicles communicate i.e. help. Moverover, our concept aims to use the existing infrastructure installed in roads, crossings and cars, thus making it feasible to deploy economically!

Our project comprises of the following modules:

Congestion Prevention: Don't you get fed up when you have to wait on traffic signal even when there is no car in the other lane. Can we control traffic lights based on car density in a particular lane? Yes!

Pothole Detection : Potholes can cause severe damages to cars, passengers and to pedistrians too. It can damage your tire's sidewall or belts. Even a minor impact may knock your vehicle out of alignment. A pothole strike can damage your shocks or struts, or harm your suspension. Can we detect them and inform about that to the concerned authority and maintain a database? Sure yes!

Facial Wallet: Public Transportation system is widely used by many in the cities, to fastrack the ticketing services and remove long lines, we bring to you facial wallet which used face detection algorithm to detect the faces and cut the amount directly from the installed cctv's
Emergency Vehicle Priority: Many people die in India because emergency vehicles don't reach required destination in time. Can we save life by informing people in advance about an incoming emergency vehicle? Definitely yes!
Rash Driving Detection: Detects rash driving on roads especially changing of lanes very frequently.

So drive happy and safe 😊

Other

In this project we have developed 7 Web-based modules and integraded an Android Application

Current Opportunity 🚀 : We can use these modules on existing infrastructure i.e CCTV's installed on public places/traffic signals, and in government vehicles, we don't need any new infrastructure, we will just extract more information from the current infrastructure present.