Face Recognition Menggunakan Algoritma Haar Cascade Classifier Dan Convolutional Neural Network

Abstract

Face  recognition  is  a  biometric  technology  that  is  widely  used  in  the  era  of  the  industrial revolution 4.0 such as smart home, security and presence. In the application of face recognition, a method is needed that can perform facial recognition quickly andwith a high level of accuracy. This study aims to determine  the  level  of  accuracy  and  computation  time  of  the  Haar  Cascade  Classifier  Algorithm  and Convolutional Neural Network in Face Recognition with the mechine learning method. Determining the level ofaccuracy is done by calculating the amount of facial data that can be recognized from the overall face  data.  Calculating  the  computational  time  is  done  by  calculating  the  time  required  during  the  facial recognition  process  by  the  computational  process.  The  process  of  determining  the  level  and  time  of computing  is  carried  out  by  the  python  computing  program  using  the  numpy  and  tensorflow  libraries. Based  on  the  analysis  carried  out  by  the  face  detection  process  using  the  Haar  Cascade  Algorithm  and Convolutional  Neural  Network,  the  program  accuracy  is 98.84% and  the  average  time  needed  to recognize faces is 0.05s