A Classification Method of Traditional Decor Pattern Based on Support Vector Machines Approach and Gray Level Co-occurrence Matrix with Mean and F-Score
Dhendra Marutho, Muljono Muljono · 2019 International Seminar on Application for Technology of Information and Communication (iSemantic) · 2019
Traditional decor pattern of Europe are based on European cultural assets like structure, clothes, pottery, etc. Each country in Europe has a unique pattern of painting. In this paper, we classify those patterns by country that is Russia, Polland, Turkey, and then Belarus. We are used GLCM for image feature extraction, and F-Score to selection features. KNN and SVM are classification method that we used in this research. The experiment result shows that classification traditional decor pattern of Europe with SVM and F-Score feature selection produce the highest accuracy that is 100 %.