Computer Vision Based Parking Lot System

Mark Stiller, Dr. Tae Song Lee, and Dr. Sairam Tangirala, Applied Physics Lab, School of Science and Technology, Georgia Gwinnett College, 1000 University Center Lane Lawrenceville, GA 30043

In the institutions and venues having multiple parking decks, it can sometimes be difficult to locate empty parking spots. Examples include academic institutions, commercial parking decks, etc. A research on the prevalence of smart parking lots indicates a recent focus on the monitoring of electric charging stations within the traditional parking lots, rather than a real time monitoring of all the parking spots inventory.

Our work focuses on designing, developing and prototyping an implementation of a hybrid computer-vision and client-facing app system to eliminate user’s guesswork and to improve efficiency of the parking lot system. In our approach, the real-time video-feed of a parking lot's entrance/exit is processed using a portable, independent, computing system. The processed information is transmitted to a remote server via telemetry using LoRa (Long Range) protocol. LoRa is a low-power wide-area network protocol developed by Semtech Corporation that works independently without any other network (like Wifi, WAN, LAN) dependencies. The server then processes telemetry-data and updates the parking spots inventory in real-time. The client facing app then uses the real-time inventory data to provide current parking lot status to the user. Additional features like notifications about full, closed, busy lots also are also planned to be implemented.

Additional Abstract Information

Presenter: Mark Stiller

Institution: Georgia Gwinnett College

Type: Oral

Subject: Computer Science

Status: Approved

Time and Location

Session: Oral 4
Date/Time: Tue 11:00am-12:00pm
Session Number: 408
List other presenters in this same room and session