Inspiration
The inspiration of this project is to assist those in need and help first responders following natural disasters such as hurricanes. Similar events that have happened recently enlightened the idea of creating a project that helps those affected and informs first responders where help is needed.
What it does
This hurricane preparation and relief tool serves as a centralized platform for citizens and local/state governments to document and visualize hurricane impacts in real time. Users can upload images of hurricane-related issues—such as downed power lines, water outages, blocked roads, structural damage, sinkholes, gas leaks, hazardous debris, or stranded individuals—which are then categorized and mapped. A map interface displays these reports as markers, allowing users to click for images and additional details at specific locations. For governments, this provides a high-resolution view of critical areas, facilitating swift, targeted relief efforts and resource allocation based on reported damage and hazard density.
How we built it
Used React for front end, Leaflet for the map, MongoDB for the database, Nodejs for backend, and wastonx.ai meta-llama/llama-3-2-11b-vision-instruct to analyze user submitted pictures
Challenges we ran into
Analyzing the Images properly using the AI model, representing them properly and displaying the icons properly on the map. Communicating to the backend showed problems as well
Accomplishments that we're proud of
We all personally gained experience especially with frontend, also being able to store data in a database and be able to use this efficiently to help real world problems.
What we learned
Learned about the intricacies that go into building a full stack application.
What's next for Holler
Using AI to see what areas could be affected by storm surge next, clicking on the report and being able to see the photo submitted.
Log in or sign up for Devpost to join the conversation.