Open AccessBalinese Mask Characters Classification using Bag of Visual Words Model
Author(s) -
Komang Budiarta,
Dewa Made Wiharta,
Komang Oka Saputra
Publication year - 2021
Publication title -
journal of electrical, electronics and informatics
Language(s) - English
Resource type - Journals
eISSN - 2622-0393
pISSN - 2549-8304
DOI - 10.24843/jeei.2021.v05.i01.p05
Subject(s) - character (mathematics) , feature (linguistics) , artificial intelligence , computer science , cluster analysis , pattern recognition (psychology) , process (computing) , natural language processing , linguistics , mathematics , philosophy , geometry , operating system
Mask, often known by Balinese as “Tapel”, is made of pule wood. It depicts the representation of characters in the “badbad” or legend. Bali has many types of mask dances that are often performed, which makes tourists interested in visiting Bali. Unfortunately, many tourists do not know the information contained in Balinese masks. The most important information contained in the character of the Balinese masks. The characters of each mask are different even though they have the same type. As mask art is also a cultural heritage from generation to generation, it needs to be preserved. It is necessary to have information in the form of technology that can distinguish the characters from Balinese masks. In this study, bag of visual word method in the classification process of Balinese mask characters is used, where in this method, there are several algorithms used, namely SURF as feature detection, K-Means as a clustering process to get the value of feature quantization, and SVM as a classification of Balinese mask character. The result of the accuracy level obtained from the testing process is 80%.