Perbandingan Pengaruh Reduksi Dimensi Metode Principal Component Analysis (PCA) dengan Metode Independent Component Analysis (ICA) dalam Pengenalan Motif Batik

Daffa', Muhammad Falikhuddin (2024) Perbandingan Pengaruh Reduksi Dimensi Metode Principal Component Analysis (PCA) dengan Metode Independent Component Analysis (ICA) dalam Pengenalan Motif Batik. Undergraduate thesis, UPN Veteran Jawa Timur.

	Text (Cover) 20081010023.-cover.pdf Download (1MB)
	Text (Bab 1) 20081010023.-bab1.pdf Download (203kB)
	Text (Bab 2) 20081010023.-bab2.pdf Restricted to Repository staff only until 30 October 2026. Download (437kB)
	Text (Bab 3) 20081010023.-bab3.pdf Restricted to Repository staff only until 30 October 2026. Download (1MB)
	Text (Bab 4) 20081010023.-bab4.pdf Restricted to Repository staff only until 30 October 2026. Download (3MB)
	Text (Bab 5) 20081010023.-bab5.pdf Download (187kB)
	Text (Daftar Pustaka) 20081010023.-daftarpustaka.pdf Download (198kB)
	Text (Lampiran) 20081010023.-lampiran.pdf Restricted to Repository staff only Download (260kB)

Abstract

Batik is one of Indonesia's widely recognized cultural heritages and has a unique philosophical value in each motif. To understand the philosophical value contained in batik motifs, batik motif recognition is important. One of the approaches used for batik motif recognition is through image analysis and pattern recognition. By involving large batik motif image data, dimension reduction techniques are needed to make it more efficient. One of the dimension reduction methods is feature extraction. Classic feature extraction that is still popular is Principal Component Analysis (PCA) and Independent Component Analysis (ICA). Research on the comparative use of the two feature extraction methods has been conducted using facial image data that has relatively uniform and structured characteristics. Therefore, in this study, more complex data is used in the form of batik motif images that have many variations in patterns. K-Nearest Neighbors (KNN) testing using reduced matrix data by retaining 20 PCA and ICA components was able to achieve the highest accuracy and shorter computation time. In Convolutional Neural Network (CNN) testing with inverse reshape image data, the use of PCA and ICA is still not optimal in accuracy because the image data contains a lot of noise that interferes with the CNN model. However, PCA and ICA are able to speed up computation time.

Item Type:

Thesis (Undergraduate)

Contributors:

Contribution	Contributors	NIDN/NIDK	Email
Thesis advisor	Puspaningrum, Eva Yulia	NIDN0005078908	evapuspaningrum.if@upnjatim.ac.id
Thesis advisor	Junaidi, Achmad	NIDN0710117803	achmad.junaidi.if@upnjatim.ac.id

Subjects:

T Technology > T Technology (General) > T385 Computer Graphics

Divisions:

Faculty of Computer Science > Departemen of Informatics

Depositing User:

Muhammad Falikhuddin Daffa'

Date Deposited:

31 Oct 2024 04:12

Last Modified:

31 Oct 2024 04:12

URI:

https://repository.upnjatim.ac.id/id/eprint/31731