Slums are generally characterized by extremely high population densities, irregular arrangement of small buildings, and poor living conditions. These areas often harbor homeless populations. Our approach involves:
-
Identification of Slum Areas: We identify areas with extensive slum presence, such as Dharavi, one of the largest slums. This is done through surveys and user inputs.
-
Location Mapping: Collected data is marked on maps to track slum areas.
-
Machine Learning Integration: Using satellite imagery (Sentinel) and machine learning, we train models to predict probable slum locations based on collected data.
-
Validation and Tracking: Probable locations are validated against user inputs and surveys. This helps in tracking homeless populations effectively.
-
Additional Features: We have incorporated features such as:
- User-input location storage in a database.
- Contacting nearby NGOs.
- Donation options via QR codes.
- Reporting suspicious activities like kidnapping, smuggling, or rape cases.
- Organizing and attending events like free health checkups, treatments, food distribution, and educational sessions.
- Mumbai_details.html: Details of slums or homeless populations displayed using various maps (Google map, Satellite map, and Machine Learning classified map).
- Support-us.html: Allows users to donate for homeless people using QR codes.
- contactauthorities.html: Allows users to contact managing authorities.
- contactNGO.html: Allows users to contact nearby NGOs.
- index.html: Home page.
- contribute.html: Allows users to contribute the location of homeless people stored in a database.
- file_report.html: Allows users to report any deformities or suspicious activities related to homeless people shown on the map.
- index-logout.html: Allows users to logout from the website.
- Login.css: Stylesheet for the login page.
- login.html: Login page.
- mission.html: Describes the mission of the project.
- signup2.css: Stylesheet for the signup page.
- signup2.html: Signup page.
- package.json: Stores important metadata about the project.
- feature_collection.js: Dataset for ML classified map to be imported in Google Earth Engine.
- ml_map_gee.js: Machine learning code used in Google Earth Engine editor to predict probable slum areas.
- github_repo_turing_crypt.txt: GitHub repository link.
For development, you will only need Node.js and a node global package, Yarn, installed in your environement.
-
Just go on official Node.js website and download the installer. Also, be sure to have
git
available in your PATH,npm
might need it (You can find git here). -
You can install node.js and npm easily with apt install, just run the following commands.
$ sudo apt install nodejs $ sudo apt install npm
-
You can find more information about the installation on the official Node.js website and the official NPM website.
If the installation was successful, you should be able to run the following command.
$ node --version
v8.11.3
$ npm --version
6.1.0
If you need to update npm
, you can make it using npm
! Cool right? After running the following command, just open again the command line and be happy.
$ npm install npm -g
Contributions are always welcome!
Client: HTML, CSS, JAVASCRIPT
Server: NodeJS, MongoDB
MACHINE LEARNING: Pre-processing, Multiclass Classification using Random Forest, NLTK, PyTorch, ReLU
Dashboard: PowerBI