Feature extraction methods for batik pattern recognition: A review

Abstract

Batik is one of the world’s cultural heritages with various decorative patterns. Each batik has its unique pattern characteristic such as color intensity, ornament visualisation and ornament size. Batik with a large-sized ornament and a high-intensity color on its background will result in a clear edge and easy-to-recognise ornaments. While batik with a small-sized ornament and a low-intensity color of the background will produce slight edge ornaments, thus it becomes difficult to recognise. In order to recognise any patterns of batik, some features that can represent each pattern are becomes necessary. Batik pattern features can be obtained from various methods of feature extraction. This paper will address various methods of feature extraction of batik for recognising batik patterns. These methods of feature extraction are categorised into several groups based on visual features produced by batik including feature extraction of color, texture formed in batik, and shape of batik ornament. The grouping of extraction methods in batik aims to describe the methods for extracting features from batik pattern to improve the recognition level of batik pattern.