All of our team members live near to separate major intersections. There are a few issues with these intersections such as light timings, and the ever-increasing amount of pedestrians being hit by cars.

What it does

There are cameras mounted to the stoplights which monitor traffic going each way at an intersection. These cameras detect the number of cars going each way throughout the day, and the numbers are compared and the light timings are changed accordingly. We can also use this technology to estimate the speed of the cars. This can be used to aid disabled people when crossing the street as they are more susceptible to being in an accident as they may not see/hear them coming if they are blind/deaf, or simply distracted on their phone through warning the people when an incoming car is approaching at a high enough speed that there wouldn't be enough space for it to stop in time. It alerts the people of this danger through a flashing light and a loud beeping noise. Think of it almost like a collision avoidance system in a car, but for pedestrians on the sidewalk.

How we built it

If smart intersections were to be implemented, we would have to keep costs down in order for these intersections to be more widespread. We are planning on using a Raspberry Pi 3 on each camera that would look through a Pi Noir camera with an extra infrared light for further night vision capabilities. All of these components would be housed inside of a weatherproof 3D printed shell and would receive power through the stoplight's power source.

Challenges we ran into

All of us were new to opencv and had to learn the basics of it before we tried anything too complex. Once we felt we had enough experience to get moving on the actual project, getting the motion tracking to work accurately became an issue because of the many bugs that were in our rough code. We ran out of time trying to get it all to work, but with more time (and energy) it is certainly possible.

Accomplishments that we're proud of

We are proud of creating a product that is feasible to build, produce, and use to improve our communities. Even if we didn't finish it, we are very pleased with the concept that we created in our first ever hackathon as at the beginning of the day, we had absolutely no idea what we were going to make let alone how we could relate it to the topics given.

What we learned

This being our first hackathon, none of us really knew what we were up against. One main takeaway from this experience is that we needed to prepare more than just making an ironic team name and coming up with some basic ideas.

What's next for Smart Intersections?

Getting the proper motion detection working along with speed detection. After that, building and testing a proper prototype.

Built With

Share this project: