Classification of Ceramic Roof Tiles Using the CNN Method
SAGA: Journal of Technology and Information System · 11.02.2026 · 3 (2) · 415-423
DOI: 10.58905/saga.v3i2.511 ↗Kurz gefasst
Fachartikel in der Zeitschrift SAGA: Journal of Technology and Information System, online erschienen am 11.02.2026. Verfasst von Achmad Danu Zakaria und 3 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY-SA 4.0).
Abstract (Originalsprache)
The research on tile classification using Convolutional Neural Network (CNN) aims to improve and address issues in the sorting process within the tile manufacturing industry. The accuracy level in manual sorting processes is very low due to errors caused by visual limitations and physical fatigue. By leveraging the capabilities of Convolutional Neural Network (CNN), a model was developed to classify tiles. This research involved several processes, including literature review, dataset collection, dataset splitting, preprocessing, Convolutional Neural Network (CNN) design, training, testing, and result evaluation. The study used 69 tile images divided into three classes: KW 1, KW 2, and KW 3. The results of testing the Convolutional Neural Network (CNN) on tile classification using 100 epochs with a data split of 90% training and 10% validation yielded an accuracy rate of 100%.
Zitieren
APA
Zakaria, A. D., Eviyanti, A., Mauliana, M. I., & Azinar, A. W. (2026). Classification of Ceramic Roof Tiles Using the CNN Method. SAGA: Journal of Technology and Information System, 3(2), 415-423. https://doi.org/10.58905/saga.v3i2.511
Vancouver
Zakaria AD, Eviyanti A, Mauliana MI, Azinar AW. Classification of Ceramic Roof Tiles Using the CNN Method. SAGA: Journal of Technology and Information System. 2026;3(2):415-423. doi:10.58905/saga.v3i2.511
BibTeX
@article{Zakaria2026,
author = {Zakaria, Achmad Danu and Eviyanti, Ade and Mauliana, Metatia Intan and Azinar, Azmuri Wahyu},
title = {Classification of Ceramic Roof Tiles Using the CNN Method},
journal = {SAGA: Journal of Technology and Information System},
year = {2026},
volume = {3},
number = {2},
pages = {415--423},
doi = {10.58905/saga.v3i2.511}
}
EUVIMED