Implementasi Face Recognition Pada Absensi Karyawan Menggunakan Local Binary Pattern Histogram dan SHA 256 bit

Abstract

Attendance is an activity that states the presence or absence of an employee. One of the technology models used to express the presence of an employee is a fingerprint machine. PT Global Andalas Cargotama has implemented an employee attendance system using a fingerprint. However, in the process of implementation, the fingerprint machine has weaknesses that are vulnerable to manipulation, fingerprint identification which is often hampered due to the need for routine machine maintenance and attendance recapitulation which is still done manually and not in real time. The face recognition system is an application to recognize faces in real time and accurately and can be applied to overcome the weaknesses of the fingerprint machine. This study uses the LBPH (Local Binary Pattern Histogram) method in recognizing employees' faces when doing attendance, this is based on several studies that have been done previously, this method has a high level of accuracy in recognizing faces. Meanwhile, SHA256 bit cryptography is used to solve the problem of attendance manipulation. The results of facial recognition both when attendance comes in and attendance goes home will be stored on the server computer locally to state employee attendance. The test results from this study showed that the attendance application was able to recognize the employee's face accurately with an average accuracy rate of 0.9511 by paying attention to the room lighting parameters, accessories on the face and the distance from the camera to the face. Through this research, it is expected to be able to overcome the problem of the weakness of the fingerprint machine and the attendance process at PT Global Andalas Cargotama to be more effective.