Batik Image Classification Using SIFT Feature Extraction, Bag of Features and Support Vector Machine
Ryfial Azhar, Desmin Tuwohingide, Dasrit Debora Kamudi, Sarimuddin, Nanik Suciati · Procedia Computer Science · 2015
Batik is a traditional fabric of Indonesian cultural heritage. Automatic batik image classification is required to preserve the wealth of traditional art of Indonesia. In such classification, a method to extract unique characteristics of batik image is important. Combination of Bag of Features (BOF) extracted using Scale-Invariant Feature Transform (SIFT) and Support Vector Machine (SVM) classifier which had been successfully implemented in various classification tasks such as hand gesture, natural images, vehicle images, is applied to batik image classification in this study. The experimental results show that average accuracy of this method reaches 97.67%, 95.47% and 79% in normal image, rotated image and scaled image, respectively.