Semantic Madurese Batik Search with Cultural Computing of Symbolic Impression Extraction and Analytical Aggregation of Color,Shape and Area Features

• Main Authors: Khotibul Umam, Ali Ridho Barakbah, Achmad Basuki
• Format: Article
• Language: English
• Published: Politeknik Elektronika Negeri Surabaya 2017-07-01
• Series: Emitter: International Journal of Engineering Technology
• Online Access: https://emitter.pens.ac.id/index.php/emitter/article/view/177
• ISSN: 2355-391X; 2443-1168

Lack of information media about Madurese batik Causes low awareness of younger generation to maintain the production of Madurese batik. Actually, Madurese Batik also has a high philosophy, which the motif and colour reflect the character of the Madurese. Madurese Batik has useful motif as a mean of traditional communication in the form of certain cultural symbols. We collected images of Madurese Batik by identifying the impression of Madurese Batik motif taken from several literature books of Madurese Batik and also the results of observation of experts or craftsmen who understand about Madurese Batik. This research proposed a new approach to create on application which can identify Madurese Batik impression by using 3D-CVQ feature extraction methods to extract color features, and used Hu Moment Invariant for feature feature extraction. Application searching of Madurese Batik image has two ways of searching, those are based on the image input Madurese Batik and based on the input of impression Madurese batik. We use 202 madurese batik motifs and use search techniques based on colors, shapes and aggregations (color and shape combinations).  Performance results using based on image queries used: (1) based on color, the average precision 90%, (2) based on shape, the average precision 85%, (3) based on aggregation, the average precision 80%, the conclusion is the color as the best feature in image query. While the performance results using based on the impression query are:  (1) based on color, the average value of true 6.7, total score 40.3, (2) based on shape, the average value of true 4.1, total score 24.1, and (3) based on the aggregation, the average value of true 2.5, the total score is 13.8, the conclusion is the color as the best feature in impression query.