Classification of Batik Motifs Using Convolutional Neural Networks

5 Pages Posted: 2 Oct 2018

See all articles by Jonathan Tristanto

Jonathan Tristanto

Tarumanagara University

Janson Hendryli

Tarumanagara University

Dyah Herwindiati

Tarumanagara University

Date Written: August 1, 2018

Abstract

Batik is Indonesia’s traditional cultural heritage which is a cloth-making technique using canting and malam. Each region of Indonesia usually produces their own batik motifs. This paper proposes the convolutional neural networks to recognize the batik motifs from images. The dataset consists of batik motifs from Lasem, Solo, and Yogyakarta region. In total, there are 13 classes of batik motif in the dataset. Experimental results show that the model performance is not sufficient enough to accurately detect most of the batik motifs.

Suggested Citation

Tristanto, Jonathan and Hendryli, Janson and Erny Herwindiati, Dyah, Classification of Batik Motifs Using Convolutional Neural Networks (August 1, 2018). International Conference on Information Technology, Engineering, Science & its Applications. Available at SSRN: https://ssrn.com/abstract=3258935 or http://dx.doi.org/10.2139/ssrn.3258935

Jonathan Tristanto (Contact Author)

Tarumanagara University ( email )

Jl. Letjen S. Parman No.1
Jl. Tanjung Duren Utara No. 1
Jakarta, IN DKI Jakarta 11470
Indonesia

Janson Hendryli

Tarumanagara University

Jl. Letjen S. Parman No.1
Jl. Tanjung Duren Utara No. 1
Jakarta, IN DKI Jakarta 11470
Indonesia

Dyah Erny Herwindiati

Tarumanagara University

Jl. Letjen S. Parman No.1
Jl. Tanjung Duren Utara No. 1
Jakarta, IN DKI Jakarta 11470
Indonesia

Register to save articles to
your library

Register

Paper statistics

Downloads
30
Abstract Views
175
PlumX Metrics