SISTEM PENGENALAN KENDARAAN BERBASIS VIDEO DENGAN ALGORITMA YOLOv5 UNTUK SIMULASI PENGENDALIAN LAMPU LALU LINTAS

Hanif, Faisal (2023) SISTEM PENGENALAN KENDARAAN BERBASIS VIDEO DENGAN ALGORITMA YOLOv5 UNTUK SIMULASI PENGENDALIAN LAMPU LALU LINTAS. Undergraduate thesis, UPN Veteran Jawa Timur.

[img]
Preview
Text (Cover)
19081010080.-cover.pdf

Download (1MB) | Preview
[img]
Preview
Text (Bab 1)
19081010080.-bab1.pdf

Download (214kB) | Preview
[img] Text (Bab 2)
19081010080.-bab2.pdf
Restricted to Registered users only until 6 June 2025.

Download (282kB)
[img] Text (Bab 3)
19081010080.-bab3.pdf
Restricted to Registered users only until 6 June 2025.

Download (762kB)
[img] Text (Bab 4)
19081010080.-bab4.pdf
Restricted to Registered users only until 6 June 2025.

Download (971kB)
[img]
Preview
Text (Bab 5)
19081010080.-bab5.pdf

Download (149kB) | Preview
[img]
Preview
Text (Daftar pustaka)
19081010080.-daftarpustaka.pdf

Download (175kB) | Preview

Abstract

Traffic lights in Indonesia is typically set by a timer. The lights will change according to the preset time. This is inefficient due to it disregarding other variables on the road such as vehicle debit and vehicle queue length. Therefore, this research aims to develop a traffic lights control system that could use the cameras at traffic light pole with YOLOv5 object detection method. This method was chosen because it could save money on camera installation because it is already instaled on many urban traffic lights. After developing and testing the system, the result is that the performance is affected by the ambient light condition. The variable tested in this research is ambience lights level. On low lights condition, this system could reach an accuracy of 79,0%, while on a good lights condition, the accuracy is 90,8%. Moreover, the system has successfully completed every test scenario. On test case with incorrect video address input, system has completed it successfully, proven by a message that says that the input address is invalid. It could also show listbox that shows traffic lights color duration calculation result on each video.

Item Type: Thesis (Undergraduate)
Contributors:
ContributionContributorsNIDN/NIDKEmail
Thesis advisorWahanani, Henni EndahNIDN0022097811UNSPECIFIED
Thesis advisorNurlaili, Afina LinaNIDN0013129303UNSPECIFIED
Subjects: Q Science > QA Mathematics > QA76 Computer software
Divisions: Faculty of Computer Science > Departemen of Informatics
Depositing User: Faisal Hanif
Date Deposited: 06 Jun 2023 05:22
Last Modified: 06 Jun 2023 05:22
URI: http://repository.upnjatim.ac.id/id/eprint/14645

Actions (login required)

View Item View Item