Enhancing Ulos Batik Pattern Recognition through Machine Learning: A Study with KNN and SVM

Abstract

This research aims to develop an automated classification system to accurately identify and classify Ulos batik patterns using K-Nearest Neighbors (K-NN) and Support Vector Machine (SVM) techniques. The method is based on computer vision technology and texture analysis using the Gray-Level Co-occurrence Matrix (GLCM). The dataset consists of 1,800 images of Ulos fabric categorized into six main motif classes. The preprocessing process involves converting images to grayscale and extracting features with GLCM. Two classification algorithms, K-NN and SVM, were used for modeling, with evaluation using confusion matrix metrics and Area Under Curve (AUC). Evaluation results show that the K-NN model has an accuracy of 82%, while SVM has an accuracy of 57%. The analysis also highlights the superiority of K-NN in distinguishing Ulos fabric patterns. This research contributes to cultural preservation and the development of the creative industry by introducing an effective automated classification system for Ulos fabric patterns.

Keywords

• image classification
• ulos batik
• computer vision
• k-nearest neighbors
• support vector machine