RANCANG BANGUN SISTEM PENCATATAN KEHADIRAN DAN PERHITUNGAN JAM KERJA PEGAWAI MENGGUNAKAN HOG DAN FACE RECOGNITION
Abstract
Attendance systems that only record clock-in and clock-out times do not fully represent employee
presence during working hours. This study aims to develop an attendance recording and workinghour
calculation system using Histogram of Oriented Gradients (HOG) and Face Recognition
to detect, identify, and monitor employees in real time. The system was developed using the
System Development Life Cycle (SDLC) method with Python, OpenCV, a webcam,
and the face_recognition library. Testing results show that the system can recognize registered
employees, record working time, apply pause and resume mechanisms, and generate attendance
and working-hour reports. System performance is still influenced by lighting conditions, facial
position, facial-data quality, and camera stability.

