Implementasi Metode K-Nearest Neighbors pada Klasifikasi Citra Aksara Lontara Menggunakan Ekstraksi Fitur Histogram of Oriented Gradients

Arhinza, Rayhan Saneval (2024) Implementasi Metode K-Nearest Neighbors pada Klasifikasi Citra Aksara Lontara Menggunakan Ekstraksi Fitur Histogram of Oriented Gradients. Undergraduate thesis, UPN Veteran Jawa Timur.

	Text (Cover) 20081010126-cover.pdf Download (2MB)
	Text (Bab 1) 20081010126-bab1.pdf Download (1MB)
	Text (Bab 2) 20081010126-bab2.pdf Restricted to Repository staff only until 4 June 2026. Download (4MB) \| Request a copy
	Text (Bab 3) 20081010126-bab3.pdf Restricted to Repository staff only until 4 June 2026. Download (4MB) \| Request a copy
	Text (Bab 4) 20081010126-bab4.pdf Restricted to Repository staff only until 4 June 2026. Download (13MB) \| Request a copy
	Text (Bab 5) 20081010126-bab5.pdf Download (731kB)
	Text (Daftar pustaka) 20081010126-daftarpustaka.pdf Download (1MB)

Abstract

The culture in Indonesia is incredibly diverse, one aspect of which is the regional languages that have their own native scripts. The Lontara script is one of the many traditional scripts in Indonesia. This script is used by the Bugis and Makassar communities. The name "Lontara" comes from the word "lontar," which is an endemic plant in the South Sulawesi Province. The Lontara script is still used by the Bugis and Makassar people, especially in traditional activities, customary ceremonies, and daily life. The Lontara script is one of the scripts in Indonesia that is endangered. Additionally, research on the Lontara script is relatively scarce. It is necessary to conduct research so that more studies on the Lontara script can be undertaken and used as references and foundations for future research. In this study, the K-NN algorithm is implemented for classification, and Histogram of Oriented Gradients is used for feature extraction to recognize patterns in the Lontara script. Based on testing results from several testing schemes, the highest accuracy achieved was 0.9638, with a total data error of 27, using a 90:10 dataset split, an image size of 64×64 pixels, and a k value of 5.

Item Type:

Thesis (Undergraduate)

Contributors:

Contribution	Contributors	NIDN/NIDK	Email
Thesis advisor	Sari, Anggraini Puspira	0716088605	anggraini.puspita.if@upnjatim.ac.id
Thesis advisor	Akbar, Fawwaz Ali	0017039201	fawwaz_ali.fik@upnjatim.ac.id

Subjects:

T Technology > T Technology (General)

Divisions:

Faculty of Computer Science > Departemen of Informatics

Depositing User:

Rayhan Saneval Arhinza

Date Deposited:

04 Jun 2024 07:46

Last Modified:

04 Jun 2024 07:46

URI:

https://repository.upnjatim.ac.id/id/eprint/24117