Deep Convolutional Generative Adversarial Network Application in Batik Pattern Generator

Abstract

Batik has been alleged as one of the oldest cultural heritages worldwide. Since the initiation, batik has been identified with various types and patterns. Various kinds of batik making techniques have long been popularized by the wider community. DCGAN serves as a new idea for the modern batik world. Such algorithm is capable of reproducing a novel image to produce batik patterns previously unnotified. Hence, this study aims to propose a DCGAN model to craft a new type of batik pattern. By utilizing a large dataset of batik images and a variety of the proposed DCGAN models, the proposed algorithm has succeeded in crafting a novel and diverse synthetic batik pattern.