Analisis Tekstur Fraktal untuk Pengenalan Motif Batik dengan Metode SVM-RBF

This research discusses the recognition and classification of batik motifs using the Fractal Texture Analysis-based Segmentation (SFTA) method integrated with Support Vector Machine (SVM). Batik, as an Indonesian cultural heritage, is the art of painting silk cloth with various motifs and patterns that reflect cultural values. To address the challenge of recognizing diverse batik motifs, this study proposes a fractal-based approach for extracting features from batik images. This method measures the fractal dimension of the image using the Box Counting Method, allowing it to depict unstructured organic textures with high precision. The extracted fractal features are then processed using various feature selection methods such as Chi-Square, Mutual Information, Variance Threshold, and others. Experimental results show that the "Dispersion Ratio" feature selection method achieves the highest accuracy of approximately 69.93% with SVM-RBF parameters (C=80), demonstrating its ability to identify relevant features for batik motif recognition. These findings make a significant.