Authors:

I Gusti Agung Gede Arya Kadyanan

Abstract:

“Endek fabric is one of the woven fabric crafts from Province of Bali, Indonesia. In its development, nowadays endek fabric is widely used as a traditional clothing or it can be used as a school or office uniforms. Most of people still don’t know if endek fabric has a variety of motifs or designs, so in this research will be explained the classification of Endek Bali fabric types based on its motifs using one of the deep learning methods. In this research, the classification of Endek Bali fabric types consist of preprocessing, training and testing which using the Convolutional Neural Network (CNN) algorithm. For the CNN architecture that will be used is the LeNet-5 architecture. In the CNN algorithm, the feature extraction process is carried out at the convolution layer, then the classification process is carried out in fully connected layer. Based on the research which conducted, it using 4 pieces of Endek Bali fabric class, where 75% (75 data) of each endek fabric class will be used as training data, whereas 25% (25 data) of each endek fabric class will be used as testing data, with the overall data for each class amounting to 100 data. In the best training scenario that using 0.00001 as the learning rate and 0.0001 as the minimum error change value, the highest accuracy value is obtained with an average accuracy by 80%.”

Keywords

Keyword Not Available

Downloads:

Download data is not yet available.

References

References Not Available

PDF:

https://jurnal.harianregional.com/jik/full-77654

Published

2022-04-29

How To Cite

KADYANAN, I Gusti Agung Gede Arya. Deep Learning Pengembangan Aplikasi Deep Learning untuk Identifikasi Kain Endek Bali.Jurnal Ilmu Komputer, [S.l.], v. 15, n. 1, p. 32-39, apr. 2022. ISSN 2622-321X. Available at: https://jurnal.harianregional.com/jik/id-77654. Date accessed: 02 Jun. 2025.

Citation Format

ABNT, APA, BibTeX, CBE, EndNote - EndNote format (Macintosh & Windows), MLA, ProCite - RIS format (Macintosh & Windows), RefWorks, Reference Manager - RIS format (Windows only), Turabian

Issue

Vol 15 No 1 (2022): April 2022

Section

Articles

Creative Commons License This work is licensed under a Creative Commons Attribution 4.0 International License