<![CDATA[Technological developments are having considerable effects on a lot of industries, particularly in the security sector. One of the important technologies in security sector is face recognition. Face recognition is a technology that verify and identify individual identity using face. There are many processes that involved in face recognition technology such as face detection methods. Face detection is a process of searching for faces in images. Each face detection method has different way to searching the face in image. It can affect the performance of face recognition technology itself. In this study, an analysis comparison between different face detection methods for face recognition was carried out. Face detection methods that used in this study was haar cascade classifier, dlib, and mediapipe. Technology that used to identify faces was Convolutional Neural Network (CNN). CNN model was trained with different face detection methods. Then it was used to carry out a simulation in identifying faces. The result of the comparison was shown in the form of performance metrics. The performance metrics include confusion matrix, accuracy, precision, recall, and f1-score. Based on the simulation that has been carried out, CNN model with haar cascade classifier face detection method generated the highest accuracy value of 98%, precision value of 98.08%, recall value of 98%, and f1-score of 97.99%.]]>
Copyrights © 2025
