Abstract—Previously conducted research has exhibited the feasibility of building a content-based image retrieval system for batik images. However, certain types of batik cloths exhibit very distinctive shapes and patterns that occur repetitively throughout the design, and serves as the main identifying feature of the cloth. Such a shape is called the batik motif. The Generalized Hough Transform is a well-known method for recognizing shapes within images, and in this paper we present initial experiments of applying this method for identifying motifs of certain batik cloths, particularly Jogjakarta batik. The results are quite promising, and show that the method can potentially be used to automate identification of well-known batik motifs in various images. I. INTRODUCTION ATIK is a very important part of the cultural heritage of Indonesia. As an acknowledgment of this, UNESCO has designated Indonesian batik as a Masterpiece of Oral and Intangible Heritage of Humanity on October 2, 2009, and has insisted that Indonesia preserve this heritage. Previously conducted research at the Faculty of Computer Science, University of Indonesia, has exhibited the feasibility of building a content-based image retrieval system for batik images using certain features such as the log- Gabor filter. However, certain types of batik cloths exhibit very distinctive shapes and patterns that occur repetitively throughout the design, and serves as the main identifying feature of the cloth. Being able to identify the occurrence of such motifs would enable a powerful and flexible batik analysis tool. In Section 2 a brief introduction to our content- based batik retrieval system is presented, including the overall web-service based architecture. Section 3 presents the Generalized Hough Transform (GHT), a well-known method for recognizing shapes within images, and how it can be used to recognize recurring batik motifs. Our experimental design, results, and analysis are presented in Sections 4 and 5, before concluding in Section 6. II. CONTENT-BASED RETRIEVAL OF BATIK Our batik CBIR is designed using a service- oriented architecture [1]. This kind of architecture consists of clients and server applications which communicate with each other via the Internet. As a client, a mobile application receives a query image from a user, while the server is the back-end application which performs image processing and then returns the results to the client. This work is supported by an RUUI (Riset Unggulan Universitas Indonesia) 2009 grant from the DRPM (Direktorat Riset dan Pengabdian Masyarakat) Universitas Indonesia. Recognition of Batik Motifs using the Generalized Hough Transform Hadaiq Rolis Sanabila and Ruli Manurung Pattern Recognition and Image Processing Lab Faculty of Computer Science, University of Indonesia hadaiq@gmail.com, maruli@cs.ui.ac.id B Fig.1. Batik CBIR Architecture