Inspiration
We wanted to build a cheap way to reliably track the capacity of a given parking lot using a combination of ultrasonic/ laser sensors, as well as computer vision.
What it does
Based on the reading of the two physical sensors and a webcam, it determines whenever it detects a vehicle or person going past, and continues its internal count that way.
How We built it
We used a combination of many technologies, including Python, Pyserial, OpenCV, and the Google Cloud Vision API.
Challenges I ran into
The physical sensors themselves were at times extremely inaccurate.
Accomplishments that I'm proud of
At the end, we managed to fix the bugs surrounding the faulty hardware.
What I learned
How to build software to interface with hardware.
What's next for ParkTrackr
Implementing a mobile app that can interface with the data to empower users with the data to make their daily lives more efficient.
Log in or sign up for Devpost to join the conversation.