top of page

ai tool

proof of concept

STATUS: Just started

This page updates you on our progress

Training City Tegucigalpa

"caminante, no hay camino: se hace camino al andar"

Antonio Machado

Artificial intelligence tool at the service of 
vulnerable communities for climate resiliency
in secondary fast urbanizing urban regions of the Global South

Product is a public flooding risk heat map for Resiliency Planning 
location prediction of vulnerable communities
provides economical & automatic updates
communities complement mapping on the ground (COVID open)

 
what WE are doing 

Where

Honduras

 

  1. Location of our Proof of Concept  is Honduras: our training city is Tegucigalpa and our Testing City is TBD.  Although Tegucigalpa is not fast urbanizing,  it presents myriad climate issues  and poverty
     

  2. Additionally, given that Hurricane Mitch struck in 1998, Tegucigalpa was  center o f several studies that will be used to train for and  to validate results
     

  3. We started developing this AI tool concept pre COVID.  Now, COVID highlights  the importance of  using remote sensing tools

questions

More with less
 

  1. Can we, with 20% of info, achieve 80% of certainty using  Machine Learning?
     

  2. Can we provide a userfriendly system that delivers fast and economical risk-heat- maps and its updates for  secondary/tertiary cities?
     

  3. What does it take to adapt algorithms to other regions?

TECHnology

Deep neural networks 

 

 

  1. Training City: Tegucigalpa, Honduras
     

  2. Testing City: TBD
     

  3. Data Sources: geospatial imagery (satellite, DEM) and open georeferenced datasets. We derived analytics-ready layers from these datasets such as topography, and distance to point of interest.
     

  4. Methods:  Convolutional neural networks (U-net architecture) for classification and prediction for vulnerable communities and hybrid models for flooding susceptibility.

stage one 

BId for wix 2.png
Dymax for wix 2.png

First stage: identification of informal settlements and vulnerable communities using machine learning, computer vision technique.

 

Output comparison with  existing AIDB map.

Screen%20Shot%202020-09-23%20at%204.15_e
Screen%20Shot%202020-09-23%20at%204.13_e

Vulnerable Communities on Hillshade.
 

Tegucigalpa, Municipality of the Central District, and Comayaguela.

Elevation DEM graphic:  
red  shows highest;  yellow shows lowest.  

 

Tegucigalpa, Municipality of the Central District, and Comayaguela. 

Screen%20Shot%202020-09-23%20at%204.16_e
Screen Shot 2020-09-23 at 4.17.04 PM.png

First stage: flooding susceptibility in Tegucigalpa, Municipality of the Central District, and Comayaguela.

Comparison: TPI map with  existing map of flooding susceptibility and research articles.

bottom of page