Skip to content

RickeyZh/YRHacks2023

Repository files navigation

YRHacks2023 - IWitness

Inspiration

Even in a futuristic world, crime exists. First responders currently rely on reports from human witnesses. However, this communication is much too slow in an emergency. The lag in response times can end up being fatal. If first responders immediately get notified about a potential threat, it can prevent many deaths and reduce crime rates. IWitness aims to be the means to execute this vision.

What it does

IWitness is a web application aimed at first responders. It is designed to hook up to cameras around the city to detect crime or medical emergencies. Once the AI detects one of those, it pings a suitable person to check the situation accordingly.

How we built it

We used HTML and CSS for the front end. Javascript was used in the backend.

Challenges we ran into

It was our first time using TensorFlow, so it took us a while to understand how to utilize the library within our program. Most of our team started with little experience in web development, which meant we came across many hurdles that we had to utilize Google to get through.

Accomplishments that we're proud of

We are proud that we were actually able to implement everything that we planned out and that our finished product turned out quite well.

What we learned

We learned the basics of TensorFlow and machine learning and became more comfortable working with languages such as HTML, CSS, and JavaScript.

What's next for IWitness

We plan on tuning the detector to make it more accurate and precise at assessing the situation. We plan on moving beyond just the image to get other inputs as well such as sound to increase reliability. We could also try to expand the reach of this project by creating a server to hold all the footage and to connect it with many cameras in the city.

Login Details

Username: admin@gmail.com
Password: password

About

YR Hacks Second Place Winner + Winner of Smart Cities

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Contributors 4

  •  
  •  
  •  
  •