PENGGUNAAN METODE YOLO UNTUK DETEKSI KENDARAAN DAN PENENTUAN TINGKAT PELANGGARAN MELAWAN ARUS LALU LINTAS MENGGUNAKAN ALGORITMA ONE DIMENSIONAL CONVOLUTIONAL NEURAL NETWORK PADA JALAN RAYA KOTA PALEMBANG

IRAWAN, GREGORIUS JOSE MAHESA and Oklilas, Ahmad Fali (2023) PENGGUNAAN METODE YOLO UNTUK DETEKSI KENDARAAN DAN PENENTUAN TINGKAT PELANGGARAN MELAWAN ARUS LALU LINTAS MENGGUNAKAN ALGORITMA ONE DIMENSIONAL CONVOLUTIONAL NEURAL NETWORK PADA JALAN RAYA KOTA PALEMBANG. Undergraduate thesis, Sriwijaya University.

	Text RAMA_56201_09011281924068.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (4MB) | Request a copy
	Text RAMA_56201_09011281924068_TURNITIN.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (3MB) | Request a copy
	Text RAMA_56201_09011281924068_0015107201_01_front_ref.pdf - Accepted Version Available under License Creative Commons Public Domain Dedication. Download (1MB)
	Text RAMA_56201_09011281924068_0015107201_02.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (1MB) | Request a copy
	Text RAMA_56201_09011281924068_0015107201_03.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (2MB) | Request a copy
	Text RAMA_56201_09011281924068_0015107201_04.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (1MB) | Request a copy
	Text RAMA_56201_09011281924068_0015107201_05.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (685kB) | Request a copy
	Text RAMA_56201_09011281924068_0015107201_06_ref.pdf - Bibliography Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (555kB) | Request a copy
	Text RAMA_56201_09011281924068_0015107201_07_lamp.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (2MB) | Request a copy

Abstract

Wrong-way traffic violation is often committed by the public and frequently leads to traffic accidents and congestion. This research aims to develop a detection system using the You Only Look Once (YOLO) algorithm to identify and count the number of vehicles based on video recordings. Additionally, this study employs the One-Dimensional Convolutional Neural Network (1DCNN) method to determine the violation level. The dataset consists of 3592 images and 40 videos of motorcycles and cars, along with a reference table of violation levels with 3 columns and 16 rows. The YOLO model achieves a model accuracy of 74.66%, while the accuracy for testing data image readings is 70.13%. The average accuracy for video data readings is 99.34%. The 1DCNN model produces a model accuracy of 50% and reading accuracy of 90%. In this study, it is found that the YOLO model can process video data to detect vehicles, the 1DCNN model can be applied to determine the wrong-way violation, and the output of the violation level in this research has three conditions: "few," "moderate," and "many," applied to the analysis of 40 videos. Based on the results obtained, it can be predicted that Srijaya Negara Street in front of the Sriwijaya University Palembang and H.M. Noerdin Street will experience many violations against traffic direction based on the analyzed violation levels.

Item Type:	Thesis (Undergraduate)
Uncontrolled Keywords:	Tingkat Pelanggaran, Melawan Arus Kendaraan, You Only Look Once (YOLO), One Dimensional Convolutional Neural Network (1DCNN)
Subjects:	Q Science > Q Science (General) > Q300-390 Cybernetics > Q325.5 Machine learning Q Science > Q Science (General) > Q334-342 Computer science. Artificial intelligence. Algorithms. Robotics. Automation.
Divisions:	09-Faculty of Computer Science > 56201-Computer Systems (S1)
Depositing User:	Gregorius Jose Mahesa Irawan
Date Deposited:	09 Jan 2024 07:06
Last Modified:	09 Jan 2024 07:06
URI:	http://repository.unsri.ac.id/id/eprint/137763

Actions (login required)

View Item