Bahasa Indonesia

Abstrak

This research aims to develop a spoofing detection system using the Support Vector Machine (SVM) method with texture feature extraction of Local Binary Patterns (LBP) and Gray-Level Co-occurrence Matrix (GLCM). The system is intended to address the challenge of student attendance by integrating face recognition technology into attendance management. Spoofing, which is an attempt to counterfeit faces, poses a challenge in face-based security systems. Therefore, the system focuses on spoofing detection by comparing texture patterns between real and fake faces. The model is able to identify face spoofing attempts with an accuracy rate of 94% after tuning the C and gamma parameters. Furthermore, the anti-spoofing attendance system is tested through black box testing and provides results that meet expectations. The system is able to start classes, record student attendance, and generate valid attendance reports. The entire system functions have been thoroughly tested and achieve a 95% accuracy rate in spoofing detection.