Open AccessThe Implementation of Backpropagation Artificial Neural Network for Recognition of Batik Lampung Motive
Author(s) -
Rico Andrian,
Bambang Hermanto,
Raja Kamil
Publication year - 2019
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1338/1/012062
Subject(s) - backpropagation , artificial neural network , artificial intelligence , value (mathematics) , computer science , mathematics , machine learning
United Nations Educational, Scientific, and Culture Organization (UNESCO) has recognized batik cloth is one of the world cultural heritage that originated from Indonesia, exactly on October 2, 2009. Batik in Indonesia has a motive that many, varied and almost every motive of batik various regions have similar motives, but if viewed in more detail batik cloth from different regions are not the same. Certain people who have expertise and knowledge in the field of batik that can distinguish batik motive from various regions. Lampung is one area in Indonesia that has a cloth motive that characterizes the Lampung area used as batik cloth. This study discusses the backpropagation artificial neural network that will be used for the classification of pattern batik motive Lampung. Batik motive lampung used is sembagi, siger ratu agung, jung agung and siger clove cengkih, while for batik is not a motive Lampung used parang kusumo and broken parang. Stages to be done are scaling, grayscale, tresholding and classification. Comparison of training data and data testing used is 70:30 and 80:30 with the need of backpropagation neural network that is epoch = 2000, learning rate = 0.1 and target error = 0.001. The greatest accuracy value is found in the 70:30 data is 92%.