Design of a road marking violation detection system at railway level crossings

Susilawati, Helfy and Nurpadillah, Sifa and Sediono, Wahju and Nurdin, Agung Ihwan (2025) Design of a road marking violation detection system at railway level crossings. International Journal of Electrical and Computer Engineering (IJECE), 15 (1). 883 - 893. ISSN 2088-8708

	PDF - Published Version Restricted to Registered users only Download (749kB) | Request a copy
	PDF (Scopus) - Supplemental Material Restricted to Registered users only Download (193kB) | Request a copy

Abstract

When a train passed through a railway-level crossing, a common phenomenon was that many vehicles attempted to overtake others by crossing into lanes designated for oncoming traffic, resulting in both roads becoming congested with motorized vehicles. At that time, no system was in place to enforce penalties for violating road markings at level crossings. Therefore, a system capable of detecting such violations when trains pass through was needed. The designed system utilized a Raspberry Pi 4, a webcam, and an ultrasonic sensor. The single shot detector (SSD) method was employed for vehicle classification. The optical character recognition (OCR) method was used for character recognition on license plates. The research involved object detection at level crossings using varied objects (cars and motorcycles) with license plates categorized into two types: white background plates with black numbers and black background plates with white numbers. Based on the research results, turning on the webcam when the bar opened and closed using an ultrasonic sensor got an average error of 0.573% and 0.582%. The system could distinguish objects with an average recognition delay of 0.554 seconds and 0.702 seconds for car and motorbike objects. Regarding number plate detection, the success rate of character recognition stood at 64.45%.

Item Type:	Article (Journal)
Additional Information:	6584/11826
Uncontrolled Keywords:	Optical character recognition Railway level crossing Raspberry Pi Road markings Single shot detector
Subjects:	T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK7800 Electronics. Computer engineering. Computer hardware. Photoelectronic devices > TK7885 Computer engineering
Kulliyyahs/Centres/Divisions/Institutes (Can select more than one option. Press CONTROL button):	Kulliyyah of Engineering > Department of Mechatronics Engineering
Depositing User:	Dr.-Ing. Wahju Sediono
Date Deposited:	08 Jan 2025 15:46
Last Modified:	08 Jan 2025 15:46
URI:	http://irep.iium.edu.my/id/eprint/118261

Actions (login required)

View Item