EUVIMEDEUVIMED entdecken
Crossref

CLASSIFICATION OF MALAKA WOVEN CLOTH TITLES BASED ON DIGITAL IMAGES USING THE CONVOLUTIONAL NEURAL NETWORK (CNN) METHOD

Redemtus Nahak, Rangga Pahlevi Putra, Fitri Marisa

JOURNAL OF SCIENCE AND APPLIED ENGINEERING · 26.10.2026 · 9 (2) · 7-14

DOI: 10.31328/jsae.v9i2.7603 ↗

Kurz gefasst

Fachartikel in der Zeitschrift JOURNAL OF SCIENCE AND APPLIED ENGINEERING, online erschienen am 26.10.2026. Verfasst von Redemtus Nahak und 2 weiteren Autor:innen.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Malaka woven cloth is one of the cultural heritages of Malaka Regency, East Nusa Tenggara, which has various distinctive motifs such as Marobo, K'nun Rua, and Sorat Los. Unfortunately, public understanding of these motifs is still limited, so a technology-based approach is needed to document and introduce them more widely. This study aims to classify the types of Malaka woven cloth based on digital images using the Convolutional Neural Network (CNN) method. Two CNN architectures used in this study are VGG-19 and ResNet-50. The data used are 900 images for training and 300 images for testing, which have gone through pre-processing stages such as cropping, augmentation, and normalization. The test results show that the VGG-19 architecture provides the highest accuracy reaching 99% with very low classification errors, while ResNet-50 is only able to achieve a maximum accuracy of 86% and shows difficulty in distinguishing certain motifs, especially K'nun Rua. Thus, the CNN method, especially the VGG-19 architecture, has proven effective in classifying Malaka woven fabrics, and has the potential to support efforts to preserve and digitize local culture through image processing technology.

Zitieren

APA

Nahak, R., Putra, R. P., & Marisa, F. (2026). CLASSIFICATION OF MALAKA WOVEN CLOTH TITLES BASED ON DIGITAL IMAGES USING THE CONVOLUTIONAL NEURAL NETWORK (CNN) METHOD. JOURNAL OF SCIENCE AND APPLIED ENGINEERING, 9(2), 7-14. https://doi.org/10.31328/jsae.v9i2.7603

Vancouver

Nahak R, Putra RP, Marisa F. CLASSIFICATION OF MALAKA WOVEN CLOTH TITLES BASED ON DIGITAL IMAGES USING THE CONVOLUTIONAL NEURAL NETWORK (CNN) METHOD. JOURNAL OF SCIENCE AND APPLIED ENGINEERING. 2026;9(2):7-14. doi:10.31328/jsae.v9i2.7603

BibTeX
@article{Nahak2026,
  author = {Nahak, Redemtus and Putra, Rangga Pahlevi and Marisa, Fitri},
  title = {CLASSIFICATION OF MALAKA WOVEN CLOTH TITLES BASED ON DIGITAL IMAGES USING THE CONVOLUTIONAL NEURAL NETWORK (CNN) METHOD},
  journal = {JOURNAL OF SCIENCE AND APPLIED ENGINEERING},
  year = {2026},
  volume = {9},
  number = {2},
  pages = {7--14},
  doi = {10.31328/jsae.v9i2.7603}
}
Kostenlos starten →