Citation

Soon, Nyean Cheong and Ng, Chin Kit (2018) Cost-Effective Outdoor Car Park System with Convolutional Neural Network on Raspberry Pi. Journal of Engineering and Applied Sciences, 13 (9 SI). pp. 7062-7067. ISSN 1816-949X

Text 7062-7067.pdf - Published Version Restricted to Repository staff only Download (1MB)

Abstract

Difficulty in finding a vacant parking space has always been a problem encountered by drivers especially in metropolitan areas. This study proposed a cost-effective vision-based outdoor parking space vacancy detection system, ConvPark to assist vehicle drivers by providing information regarding the availability of parking spaces. The system is designed based on Convolutional Neural Network (CNN) technology and is implemented through a Raspberry Pi to identify the occupancy status of parking spaces live via. an IP camera. This system has been deployed at a university car park for real-time detection of vacant parking spaces. The use of CNN classifier in the proposed system provides superiority in term of automatic image features extraction and robustness against environmental variations as compared to other computer vision-based methods. Evaluation outcomes demonstrated that our proposed system can achieve excellence performance in term of detection accuracy by precisely determining the occupancy status of parking spaces under different environmental conditions.

Item Type:	Article
Uncontrolled Keywords:	Neural network
Subjects:	Q Science > QA Mathematics > QA71-90 Instruments and machines > QA75.5-76.95 Electronic computers. Computer science
Divisions:	Faculty of Engineering (FOE)
Depositing User:	Ms Suzilawati Abu Samah
Date Deposited:	18 Apr 2021 14:23
Last Modified:	18 Apr 2021 14:23
URII:	http://shdl.mmu.edu.my/id/eprint/7607

Downloads

Edit (login required)