CyberScape| A Flood of Ideas

A Flood of Ideas

Your challenge is to develop a new methodology or algorithm that leverages Earth observation and critical infrastructure datasets to estimate damages to infrastructure caused by flooding. Make a measurable impact on the resilience of nations by helping the Earth observations community contribute to the United Nations’ primary effort to reduce disaster risk!

Estimation of infrastructural damage due to floods

Summary

Using a customized dataset from HIFLD Tsunami Events datasets, with inputs like Cause of tsunami, Intensity, Location, etc, and output of predicted Infrastructural damage on a scale from 1-4 (based on the cost of damage in millions of dollars, deaths, injuries)

How We Addressed This Challenge

We have taken this as a multiple classification problem (supervised learning) and used a Deep Neural networks sequential model, and provided a training dataset of around 27,000+ entries for the system.

How We Developed This Project

Using Tensorflow and keras libraries available in python

How We Used Space Agency Data in This Project

Since the NASA data is all disparate, we combined and customized a couple of datasets available to make one huge training set from HIFLD Tsunami events data. We have input parameters like Cause Code( Represents Cause of Tsunami), Tsunami Intensity and Location (Latitude. longitude), to map to a damage amount output on a scale from 1-4

(1= Damage amounts to less than 1 million $,

2= Damage amounts to 1-5 million $, and so on)

Project Demo

Here is the link to our presentation:

https://docs.google.com/presentation/d/1n56V89jSRMjCBQJiH3NSZ6SdT_94h2jJptWCeEyHW9Q/edit?usp=sharing

Project Code
Data & Resources
Tags
Machine learning
Judging
This project was submitted for consideration during the Space Apps Judging process.