This Author published in this journals
All Journal ICONESTH
Claim Missing Document
Check
Articles

Found 1 Documents
Search

PERFORMANCE OF COMPUTER VISION METHOD FOR OVER-HEIGHT VEHICLE DETECTION Twk Mohd Iqbal; Ilahi, Ridha; Munandar, Munandar; Irhamna, Reza
International Conference on Education, Science, Technology and Health (ICONESTH) 2024: The 2nd ICONESTH
Publisher : International Conference on Education, Science, Technology and Health (ICONESTH)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.46244/iconesth.vi.480

Abstract

Prevention of accidents for over-height vehicles that will pass through tunnels or underpasses is very necessary. This prevention is done by providing an early warning to over-height vehicles. Early warning is carried out using a system that is able to detect oversized vehicles from the tunnel or underpass height allowed to be passed. This system must have high accuracy and be resistant to various environmental conditions. Interference in detection can occur due to rain, wind, lack of light, night time, various backgrounds such as walls, trees, and two-way roads. This research examines the durability of computer vision-based over-height vehicle detection methods against environmental condition influences. The methods used are Gaussian Mixture Model (GMM) and blob detection. Samples were taken in 3 different location conditions and detected in 5 frame variations. The results of this research show that GMM and blob detection is excellent performance and resilience against environmental conditions. This method can detect over-height vehicles with an accuracy reaching 100% with the number of frames at 5 and 7. It is expected that this detection method can provide more accurate early warnings for over-height vehicles, thereby increasing road user safety.