View on GitHub

volcanic-disaster-recovery

MIDS W210 Capstone

VolcanoResponse

Mission

To Improve the efficiency of disaster response resources for volcano eruptions in developing countries, using computer vision estimates based on recent satellite image data.

Problem

Developing nations lack the same volcanic detection and monitoring infrastructure as developed nations that are important in preparing for eruptions. We aim to bridge this resource and knowledge gap by using computer vision and data science on satellite imagery to pinpoint which areas are likely to be damaged alongside damage severity estimates. International Disaster Response Organizations can utilize this knowledge to be better informed on what decisions to make and how to effectively utilize resources to protect communities and save lives.

Check Out the VolcanoResponse Dashboard Here

Watch this video to learn how to use the dashboard.

Dashboard Image

VolcanoResponse is focused on improving disaster response to volcano eruptions affecting 8% of the population that are within 100 kilometers of potentially active volcanoes in developing countries. Using satellite imagery across multiple channels, we are utilizing computer vision processing and prediction techniques within our MVP intended for disaster response organizations. We seek to use our MVP to contribute to peer learning about combining satellite data and machine learning for the task of volcanic damage classification.

Team

Ren Tu Jeffrey Adams Mackenzie Lee Anne Marshall
Ren Tu Jeffrey Adams Mackenzie Lee Anne Marshall

MIDS Capstone Summer 2022