CAMERA-BASED DETECTORS AS AN ALTERNATIVE TO DETECTING TRAINS IN A LEVEL CROSSING IMPLEMENTATION
Based on data from Indonesia Directorate General of Railways in 2017, it is mentioned that the problems at the level crossing of railroad tracks are mostly caused by human error factors themselves. The current train headway and the crossing system that is still operated manually can increase the potential for accidents. Therefore, the development of alternative camera-based detectors to support the railroad crossing automation system is needed at this time. The development of this camera-based train detector uses the basic program You Only Look Once (YOLO), where YOLOv3 has proven to be accurate enough to detect moving objects. The development results show promising results for several types of alternative trains.
Key Words : Detectors; Train; YOLOv3
S. P. Rajendran, L. Shine, R. Pradeep, S. Vijayaraghavan, â€œReal-Time Traffic Sign Recognition using YOLOv3 based Detectorâ€, Proc. 10th ICCCNT 2019, India, July 2019.
Q-C. Mao, H-M. Sun, Y-B. Lu, R-S. Jia, â€œMini-YOLOv3: Realtime Object Detector for Embedded Applicationsâ€, IEEE Access Vol. 7, September 2019
H. Chen, Z. He, B. Shi, T. Zhong, â€œResearch on Recognition Method of Electrical Components Based on YOLO V3â€, Proc. Special Section on Artificial Intelligence Technologies for Elecric Power System, IEEE, November 2019
A. Avramovic, â€œReal-Time Large Scale Traffic Sign Detectionâ€, 14th Symposium on Neural Network and Aplication, Serbia, November 2018.
V. Kharchenko, I. Chyrka, â€œDetection of Airplanes on the Ground using YOLO Neural Networkâ€, IEEE Int. Conf. Mathematical Methods in Electromagnetic Theory, Ukraine, 2018.
J. Lin, M. Sun, â€œA Yolo-based Traffic Counting Systemâ€, Conf. on Technologies and Applications of Artificial Intelligence, Taiwan, 2018.
M. Shahud, P. Praneetpolgrang, J. Bajracharya, S. Petcharee, â€œThai Traffic Sign Detection and Recognition Using Convolutional Neural Networkâ€, IEEE, 2018.
S. Bismantoko, H. Luthfiyah, G. M. Putra, S. Hidayat, â€œTinjauan Teknis Deteksi Objek Berbasis Image Processing dalam Kasus Uji Coba Parkir Sepeda motor di Jalanâ€, Jurnal online STMIK, Indonesia. 2018.
Open Access Policy
MIPI provides immediate open access to its content on the principle that making research freely available to the public supports a greater global exchange of knowledge.
MIPI by TIRBR-BPPT is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License. Permissions beyond the scope of this license may be available at http://ejurnal.bppt.go.id/index.php/MIPI