<![CDATA[This study discusses the development of a fingerprint type classification system based on digital image processing using the Support Vector Machine (SVM) method. The system is designed to recognize three main fingerprint patterns: arch, loop, and whorl. The data processing stages include binarization of the fingerprint image and feature extraction using the Histogram of Oriented Gradients (HOG) method. Once the features are extracted, classification is performed using the SVM algorithm with a Radial Basis Function (RBF) kernel to improve separation performance between classes. The dataset used in this study was obtained from the Kaggle platform, and the system was implemented using MATLAB software, complete with a graphical user interface (GUI) to facilitate user interaction. The system’s performance was evaluated by dividing the dataset into 80% training data and 20% testing data. The results show that the model is capable of classifying fingerprint patterns with an accuracy of 89.25%. These findings indicate that the SVM method is effective and can serve as an initial solution for automatic fingerprint-based identification systems. ]]>
Copyrights © 2025
