wee Abstract— Batik as a decorative cloth has a very important role in many traditional ceremonies in Java and throughout Indonesia. Previously conducted research has exhibited the feasibility of building a content-based image retrieval system for batik images using certain features of the image, namely the log-Gabor filter. In this paper we employ the log-Gabor filter and colour histogram feature. We devise several clustering experiments applied to a collection of previously compiled batik images to measure how well these features can separate apart traditional cloths from Java with those from outside Java, how well they can they separate batiks from different regions, and so forth. The results shows that increasing the granularity of detail in the clustering task leads to decreased results, and that the colour histogram feature consistently performs better. I. INTRODUCTION ATIK as a decorative cloth has a very important role in many traditional ceremonies in Java and other islands throughout Indonesia. In this paper, we explore how two well-known image processing features, namely the log-Gabor filter and the colour histogram, would characterise a collection of batik images. Using these features, we devised four clustering experiments applied to a collection of previously compiled batik images [3]. For example, how well can these features separate apart traditional cloths from Java with those from outside Java, how well can they separate batiks from different regions, and so forth. In Section 2 a brief introduction to various aspects of batik cloths is given. Sections 3 and 4 discusses content-based image retrieval, specifically concerning batik, and the two main features that are used throughout this paper: the log-Gabor filter and colour histograms. Our experimental design, results, and analysis are presented in Sections 5 and 6, before concluding in Section 7. This work is supported by an RUUI (Riset Unggulan Universitas Indonesia) 2009 grant from the DRPM (Direktorat Riset dan Pengabdian Masyarakat) Universitas Indonesia. II. BATIK CLOTHS Batik cloths are used extensively in many aspects of life in an Indonesian lifestyle, such as pregnancy, childbirth, circumsicion, weddings, and so forth. Certain batik patterns were previously reserved for the Javanese traditional monarchy, which lifted batik status as not only a decorative cloth but also as a symbol of status. As of today, such restrictions no longer apply to the Javanese community, hence batik is even more widely used at all social levels and produced in many more different patterns. Even so, until this day, batik still plays an important role in the Javanese monarchy wardrobe. This specific type of batik is more commonly known as the traditional Javanese batik. As an acknowledgment of batik as being part of Indonesian cultural heritage, UNESCO designated Indonesian batik as a Masterpiece of Oral and Intangible Heritage of Humanity on October 2, 2009, and as part of the acknowledgment, has insisted that Indonesia preserve this heritage. Nevertheless, batik cloth is not an exclusive product of solely the Javanese community. Broadening the definition of decorative cloth, other areas in Indonesia also have their own patterned cloth. During the development of the batik cloth culture, merchants bringing batik cloths as their commodity around Indonesia are to be thanked for the distribution of batik cloths. In the meantime, batik makers are then influenced by other cultures and art, creating more patterns, textures and colours to the library of decorative cloth in Indonesia. In course of the distribution of batik culture in various places across Indonesia, other areas have established their own decorative batik cloths, which adds the variety to Indonesian batik in general. The term batik was initially a verb in the Indonesian language, which refers to the activity of dyeing decorative cloth with colour, after applying a resistive technique [1]. The textile that is used to make batik cloths is usually a natural type of fabric such as cotton or silk, and it must be densely woven to go through the Clustering Batik Images based on Log-Gabor and Colour Histogram Features Laksmita Rahadianti, Ruli Manurung, and Aniati Murni Pattern Recognition and Image Processing Lab Faculty of Computer Science, University of Indonesia laksmita_r@yahoo.com, maruli@cs.ui.ac.id, aniati@cs.ui.ac.id B