Kaji Sistem Pengolahan Citra Untuk Pendeteksian dan Pengenalan Wajah Secara Real-Time

Abstract

At present, identification systems are developing rapidly. This development makes an advance in biometric-based security systems. Face recognition is one of many forms of identification in biometrics that is made based on differences in the shape of a person's face that has high accuracy. The use of detected faces can be used in a variety of useful ways. As in the field of security, a person's identity recognition system through the face to secure the cellphone, automatic door opening, absence of facial presence, even in the search for someone who is missing and the search for criminals. The Raspberry Pi device with the ability to detect and recognize a person's face in real time is a design made in this paper. Raspberry Pi can recognize faces in real time using the Python programming language, the LBPH algorithm and the OpenCV library as well as a database that has been created for the face recognition process. In manual testing using a Raspberry Pi and a webcam camera, the results showed that the test succeeded in recognizing faces with the largest percentage accuracy of faces, namely 92% based on the closest distance between the camera and the target as far as 50 cm. The further away the target face is from the camera, the smaller the percentage of face recognition ability, it even becomes unrecognizable.