Stampede Shield

Inspiration

The idea came to us while we were brainstorming unique emergency management ideas, and it came to us how dangerous a rapidly populating crowd can be. Once inside of a crowd, it's hard to find yourself out of the current as more people start flowing in. A solution for this? An application that collects real-time footage of a crowd and highlights areas that are of high risk to become dangerous.

What it does

Stampede Shield utilizes YOLOv8's object detection software, paired with real-time data and predictive analysis, to monitor foot traffic and identify density zones where danger could rise. Stampede Shield can also be used to keep track of the number of people it has registered within a certain amount of time to keep watch for fights or areas where people are fleeing from.

Challenges we ran into

The biggest challenge we encountered was trying to keep track of communication between our frontend and backend systems. It was difficult at some times to find files that were modified in the backend and were intended to be used in the frontend.

Accomplishments that we're proud of

We were more than proud to be able to exercise the true potential of YOLOv8's computer vision model and the accuracy that it can maintain when working with larger crowds, as well as crowds with other objects mixed in such as cars or animals.

What we learned

We learned a lot about using computer vision software and what it looks like to actually implement in a code base, as well as gaining strength in almost every area as we all had multiple roles on the team.

What's next for Stampede Shield

We are proud of what we created in 24 hours, but want to continue working on Stampede Shield still. Some things we want to improve is making the UI more intuitive, especially for the quadrant system, and also fully setting up the domain we have.

Built With

• css
• fastapi
• html
• javascript
• python
• react
• yolov8