IMPLEMENTASI SISTEM DETEKSI KETERSEDIAAN SLOT PARKIR MOTOR DI FASILKOM UNSRI KAMPUS BUKIT BESAR BERBASIS ANDROID MENGGUNAKAN METODE YOLOv8

HUSTHA, FARIZ and Oklilas, Ahmad Fali (2024) IMPLEMENTASI SISTEM DETEKSI KETERSEDIAAN SLOT PARKIR MOTOR DI FASILKOM UNSRI KAMPUS BUKIT BESAR BERBASIS ANDROID MENGGUNAKAN METODE YOLOv8. Undergraduate thesis, Sriwijaya University.

	Text RAMA_56201_09011282025072.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (3MB) | Request a copy
	Text RAMA_56201_09011282025072_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_09011282025072_0015107201_01_front_ref.pdf - Accepted Version Available under License Creative Commons Public Domain Dedication. Download (860kB)
	Text RAMA_56201_09011282025072_0015107201_02.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (398kB) | Request a copy
	Text RAMA_56201_09011282025072_0015107201_03.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (551kB) | Request a copy
	Text RAMA_56201_09011282025072_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_09011282025072_0015107201_05.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (125kB) | Request a copy
	Text RAMA_56201_09011282025072_0015107201_06_ref.pdf - Bibliography Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (201kB) | Request a copy
	Text RAMA_56201_09011282025072_0015107201_07_lamp.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (1MB) | Request a copy

Abstract

Motorcycle parking is a problem for students, lecturers, and employees at Fasilkom UNSRI Bukit Besar campus Palembang due to limited parking space and lack of information about empty parking slots. This causes difficulties in finding a safe and convenient parking lot, as well as wasting time and fuel. The purpose of this research is to develop a motorcycle parking slot availability detection system at Fasilkom UNSRI Bukit Besar Campus based on Android with the YOLOv8 method. The YOLOv8 model was trained with a single class "motorcycle" dataset and showed 90,7% precision and 78,8% recall on the training dataset. The mAP50 value of 85,9% shows good performance at low overlap, while mAP50-95 of 59,8% shows challenges at high overlap. On the validation dataset, precision reached 84,2% and recall 70,1%, with an mAP50 value of 77% and mAP50-95 of 45,8%. Testing with the testing dataset showed 83,4% precision, 72,3% recall, 78,6% mAP50, and 48% mAP50-95. Testing with parking lot CCTV videos resulted in 79% accuracy from 7 videos of 1 hour duration per video, influenced by factors such as camera angle, camera height, lighting, camera quality, and weather. The video output was restreamed using Flask in Python and displayed on an Android application developed with Android Studio IDE and Python.

Item Type:	Thesis (Undergraduate)
Uncontrolled Keywords:	Deteksi Slot Parkir, YOLOv8, Android, Object Detection
Subjects:	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:	Fariz Hustha
Date Deposited:	19 Jul 2024 09:02
Last Modified:	19 Jul 2024 09:02
URI:	http://repository.unsri.ac.id/id/eprint/152076

Actions (login required)

View Item