THE IMPLEMENTATION OF GRAY LEVEL CO-OCCURRENCE MATRIX AND CONVOLUTIONAL NEURAL NETWORKS FOR CLASSIFYING YOGYAKARTA BATIK MOTIFS

DANI, ANANDA RIZKI (2025) THE IMPLEMENTATION OF GRAY LEVEL CO-OCCURRENCE MATRIX AND CONVOLUTIONAL NEURAL NETWORKS FOR CLASSIFYING YOGYAKARTA BATIK MOTIFS. Tugas Akhir thesis, Informatics.

Text 5210411362_AnandaRizkiDani_Abstrak.pdf Download (11kB)

Abstract

Yogyakarta batik motifs exhibit a rich diversity that reflects the cultural heritage of Indonesia. However, the identification process for these motifs is still predominantly manual, requiring expert knowledge and limiting scalability in digital applications. This study aims to develop an automated batik motif classification system using a hybrid approach that combines Gray Level Co-occurrence Matrix (GLCM) for texture feature extraction and a Convolutional Neural Network (CNN) based on MobileNetV2 for image classification. GLCM is chosen for its ability to extract detailed texture information, while MobileNetV2 is employed for its lightweight yet effective pattern recognition capabilities. The dataset comprises 3,223 images categorized into five batik motifs: Batik Ceplok, Batik Kawung, Batik Truntum, Batik Parang, and Batik Ciptoning, sourced from institutions such as the Yogyakarta Palace Batik Museum and Kaggle. Experimental results demonstrate a high classification accuracy of 99%, validating the effectiveness of the proposed approach in distinguishing intricate batik patterns. The final output of this study is a classification system that can be integrated into a mobile client-server application, enabling real-time automatic detection of Yogyakarta batik motifs. Keywords: Batik, CNN, GLCM, Classification, MobileNetV2

Item Type:	Thesis (Skripsi, Tugas Akhir or Kerja Praktek) (Tugas Akhir)
Subjects:	T Technology > T Technology (General)
Divisions:	Fakultas Sains Dan Teknologi > S1 Informatika
Depositing User:	Kaprodi S1 Informatika UTY
Date Deposited:	16 Jul 2025 02:13
Last Modified:	16 Jul 2025 02:13
URI:	http://eprints.uty.ac.id/id/eprint/18156

Actions (login required)

View Item