Face Authentication in E-learning using Local Binary Pattern and Haar cascade

Face Authentication is one way to secure a system against attacks, especially useful for systems such as E-Learning which is used by many people as a medium of online learning for students. We review several methods that have been tried in the E-Learning system, discuss the advantages and disadvantages of each method. The results of the review it was found that face detection alone would be better if using Deep Convolutional Neural Network when compared to Local Binary Pattern. As for the effective feature recognition method for E-Learning is Deep-Learning, because the accuracy level generated by deep learning is very good at 95%, and the speed of the method for someone’s technician requires only 1.7 ms (millisecond), Haar is good enough for used, but the resulting accuracy rate is only 68%, and the time required for someone’s technician is 4s (second). © 2018 IEEE.