Inspiration
Parking at RIT can be difficult at times and there are times when I try to park in a lot and can't find a single spot. I hope to be able to check where is a good place to park on campus before even arriving at the lot.
What it does
We offer real time data analytics of RIT parking lot capacity and usage, while tracking that usage, we are able to relay which buildings and spaces across the campus receive attention and at what times during the week so RIT can focus its resources in places that benefit students most.
How we built it
We used React for the front end and Flask for the backend. The algorithm was written in python using cv2 and yolov8s for car tracking.
Challenges we ran into
We ran into issues with taking data, recording videos was difficult because it required getting a good shot at a good angle and without zoom interruptions (since we don't have access to live feed cameras at RIT that do exist). The reason for this was because it would crash the software without leaving any error messages. Making the frontend dynamic as well gave issues as the design was slowly perfected through lots of trial and error.
Accomplishments that we're proud of
We are proud that we were able to accomplish all of this in less then 24 hours between the two of us. We were also able to make the campus zoom feature work smoothly and efficiently.
What we learned
We learned more about the world of image recognition and machine learning algorithms for that.
What's next for Park-a-lytics
We aim to increase the size of the parking lots that we are able to use (expanding to every single space), gain access to 24/7 live feeds and expand to the entire RIT campus as a proof of concept before going to other campuses nationwide. We believe that our product is worthwhile as well since it is able to track movement across campus and see engagement across the semester and year with students to better allow campuses to utilize their resources at little extra cost.
Log in or sign up for Devpost to join the conversation.