OPTIMASI HYPERPARAMETER YOLOV8 MENGGUNAKAN ONE FACTOR AT A TIME (OFAT) DAN RANDOM SEARCH UNTUK DETEKSI OBJEK SAMPAH

Maulana, Muhammad Aldi (2025) OPTIMASI HYPERPARAMETER YOLOV8 MENGGUNAKAN ONE FACTOR AT A TIME (OFAT) DAN RANDOM SEARCH UNTUK DETEKSI OBJEK SAMPAH. Undergraduate thesis, UPN Veteran Jawa Timur.

	Text (Cover) Cover.pdf Download (657kB)
	Text (Bab 1) Bab 1.pdf Download (248kB)
	Text (Bab 2) Bab 2.pdf Restricted to Repository staff only until 4 December 2027. Download (925kB)
	Text (Bab 3) Bab 3.pdf Restricted to Repository staff only until 4 December 2027. Download (730kB)
	Text (Bab 4) Bab 4.pdf Restricted to Repository staff only until 4 December 2027. Download (2MB)
	Text (Bab 5) Bab 5.pdf Download (200kB)
	Text (Daftar Pustaka) Daftar Pustaka.pdf Download (183kB)
	Text (Lampiran) Lampiran.pdf Restricted to Repository staff only Download (262kB)

Abstract

Waste management has become a global issue that poses serious threats to both the environment and human health. Ineffective waste handling can lead to environmental pollution, ecosystem degradation, and numerous negative impacts on society. This study aims to develop a waste object detection system using the YOLOv8 method by applying hyperparameter optimization techniques, namely OFAT and Random Search. The dataset used in this research was obtained from the Jambangan Waste Disposal Site (TPS Jambangan) in Surabaya and through direct image collection using a smartphone. The dataset underwent several pre-processing stages, including data splitting, annotation, data division, augmentation, and model training using YOLOv8. The evaluation was conducted by comparing two model scenarios: the optimized model and the non-optimized (default) model, using a 70:20:10 data split ratio. The results show that the YOLOv8 model optimized with OFAT and Random Search achieved the best detection performance with a mean Average Precision (mAP) of 87.5%, precision of 87.4%, recall of 79.7%, and an F1-score of 87.5%. In comparison, the default model achieved an mAP of 84.4%, precision of 91.9%, recall of 75.2%, and an F1-score of 82.6%. Although a decrease in recall was observed, the optimized model outperformed the default model in other metrics, achieving a 3.1% improvement in mAP. The findings of this study indicate that applying hyperparameter optimization using OFAT and Random Search is effective in improving the performance of the YOLOv8 model, thereby enhancing the accuracy of waste object detection. The developed model demonstrates strong applicative potential to support various automated detection systems that can help improve efficiency in waste management.

Item Type:

Thesis (Undergraduate)

Contributors:

Contribution	Contributors	NIDN/NIDK	Email
Thesis advisor	Puspaningrum, Eva Yulia	NIDN0005078908	evapuspaningrum.if@upnjatim.ac.id
Thesis advisor	Rahajoe, Ani Dijah	NIDN0012057301	anidijah.if@upnjatim.ac.id

Subjects:

T Technology > T Technology (General)

Divisions:

Faculty of Computer Science > Departemen of Informatics

Depositing User:

Muhammad Aldi Maulana

Date Deposited:

04 Dec 2025 04:58

Last Modified:

04 Dec 2025 04:58

URI:

https://repository.upnjatim.ac.id/id/eprint/47767