We've wanted to create something for any campus to help filter out those who don't have parking permits from those who do so that they rightfully get a parking spot. Updates would also be sent out so that people are notified if there is space in a parking structure. This would facilitate the parking process for employees or students and parking management.
What it does
Our app is a simple idea that utilizes hardware and software to identify whether or not a car that enters a parking structure has a parking permit or not. A camera scans for a license plate and our software determines whether or not that driver has a parking permit. The application would display registered parkers and unregistered parkers. Also, the app would notify all users if a parking lot is full.
How we built it
For our app, we used a Python program to hook up a simple Logitech camera to use Google Cloud Platform's cloud vision to identify the text on a driver's license plate. Depending on the quality of the picture taken the efficiency of the correct license plate number will decrease. We then used Cloud Firestore to store those license plate IDs. We also used Node.js as a server and NPM as a package manager. In order to send out text messages to notify drivers, we used the Twilio API. For the front-end, we used Bootstrap and Angular to help create a fast and responsive user interface.
Challenges we ran into
Front-end challenges that we've faced was to correctly display the license plate data. Making the UI look pleasing and responsive was also a challenge. For the back-end some of the challenges that we faced were setting up Twilio SMS on Firestore via the Google Cloud Functions, as well as, collecting and putting the correct data into Firestore. Another problem that we had is that when using the Google Cloud Vision API, if the quality of the picture of the license is low then the accuracy of the license plate number will decrease.
Accomplishments that we're proud of
We're proud that we ended up integrating everything well despite numerous issues.
What we learned
We learned how to make a web application that leverages the Google Cloud Platform and also uses hardware.
What's next for ParkifySolutions
Currently, a majority of the essentials of ParkifySolutions are running properly. However, there are still minute details that still needs to either be implemented or perfected, such as setting up a motion sensor camera to take the picture of the license plates, perfect the different uses of the database to allow parking services to use ParkifySolutions properly, and implemented a real time update of the number of parking spots open in a parking structure.