Home // INTELLI 2022, The Eleventh International Conference on Intelligent Systems and Applications // View article

A Deep Learning based Unoccupied Parking Space Detection Method for City Lots

Authors:
Hamid Reza Tohidypour
Yixiao Wang
Panos Nasiopoulos
Mahsa T. Pourazad

Keywords: deep learning; city lots; object recognition; YOLOv4.

Abstract:
Nowadays, finding a vacant parking space in populated metropolitan cities is a challenging task, leading to serious traffic congestion with environmental and productivity ramifications. Although many different systems have been proposed and tested for unsupervised parking lot space detection over the years, they have been proven to be either impractical or costly to maintain. In this paper, we propose a deep learning based approach that uses the video captured by a vehicle mounted camera to accurately detect and count the number of available parking spaces in real-time. Our system achieves an impressive average detection accuracy of 90.59% for unoccupied spaces and 95.66% for the occupied spaces.

Pages: 11 to 15

Copyright: Copyright (c) IARIA, 2022

Publication date: May 22, 2022

Published in: conference

ISSN: 2308-4065

ISBN: 978-1-61208-977-5

Location: Venice, Italy

Dates: from May 22, 2022 to May 26, 2022