PERANCANGAN SISTEM BIOMETRIK TELINGA BERBASIS TEKSTUR GLCM MENGGUNAKAN METODE SUPPORT VECTOR MACHINE

In today's era of ever-evolving technology, building a reliable security system for humans requires innovation. to protect personal data, privacy data, and access permissions that can be dangerous if others can access them. Biometrics on the human body continue to be developed to protect the system from theft or access forgery. The ear is one of the biometric characteristics that attracts the attention of researchers in the field of computer vision recognition, so this study discusses the design of an ear biometric system based on Gray Level Co-occurrence Matrix (GLCM) texture using Support Vector Machine (SVM). This study aims to develop an ear biometric identification system that can recognize individuals based on the unique characteristics of the ear. The method used in this study is image processing, which includes several important stages, such as image pre-processing to improve image quality, segmentation to separate the ear area from the background, and feature extraction using Gray Level Co-occurrence Matrix (GLCM) to obtain texture information on the ear image. The final stage of classification is carried out using the Support Vector Machine (SVM) algorithm which is known to be effective in pattern recognition. The results of the study showed that the designed system was able to achieve an average accuracy of 76% to 80%, with an average sensitivity of 80% and a high specificity of 94% in various test scenarios. The average recognition for each individual showed varying results, where the first and fifth individuals achieved 100% accuracy, the second individual 95%, the third individual 85%, and the fourth individual 95%. The conclusion of this study is that the developed texture-based ear biometric system can be an effective alternative in individual identification. This research is expected to be the basis for further development in biometric technology, as well as opening up opportunities for further research on security systems.
Keywords: Biometrics, Ear, Image Processing, GLCM, Support Vector Machine (SVM)