Satnet

• 

  Heat map of low income and Industrial areas

• 

  City of Rio De Janeiro

• 

  Picture of the Model we Created

Inspiration

We were inspired by the recent pandemics that rocked the world and we wanted to find a way to prevent them.

What it does

It takes satellite images from cities across the world and classifies them as slums, industrial, how high-income residential to determine the most prominent area for a disease outbreak within a city.

How we built it

We used Tensorflow and AWS EC2 instance to train models on a dataset of 7000 satellite images.

Challenges we ran into

The virtual machine, EC2, suffered from immense throttling and thus its capacitors had to be altered.

Accomplishments that we're proud of

We are incredibly proud of building this model on a virtual machine as it was a challenge considering how large our data files were.

What we learned

What's next for Satnet

We would like to use more powerful satellites and try to detect more features (war damage) across the world to find more ways to better help people.

Built With

• amazon-ec2
• amazon-web-services
• jupyter-notebook
• keras
• matplotlib
• numpy
• pandas
• pillow
• python
• tensorflow