Metode Haar-Cascade dan LBPH Untuk Face Detection dan Recognition dalam Pencatatan Kehadiran

Abstract

Face recognition technology has been widely implemented in various fields such as security systems, access control, and automated attendance recording due to its non-invasive and real-time capabilities. This study aims to design and implement a real-time face detection and recognition system using a laptop camera by combining the Haar Cascade Classifier and Local Binary Pattern Histogram (LBPH) methods based on the OpenCV library. The research method applied is Research and Development (R&D), which focuses on developing a functional system and evaluating its effectiveness. The Haar Cascade method is utilized to detect facial regions efficiently using Haar-like features and a cascade classifier, while the LBPH method is employed to recognize faces by extracting local texture features and representing them in histogram form. The system workflow includes dataset collection, dataset training, face detection, face recognition, and attendance recording. Experimental results show that the proposed system is capable of detecting and recognizing registered users in real time with satisfactory accuracy under normal lighting conditions. The system can differentiate registered faces from unregistered ones and successfully record attendance based on recognized identities. However, performance may be affected by variations in lighting intensity, face orientation, and image quality. Overall, this research demonstrates that the combination of Haar Cascade and LBPH methods can be effectively applied to real-time face recognition-based attendance systems and may serve as a reference for further development in biometric identification applications.