#TITLE_ALTERNATIVE#

• Main Author: RUKMANA (NIM : 10206068); Pembimbing : Dr. Rizal Kurniadi, DANI
• Format: Final Project
• Language: Indonesia
• Online Access: https://digilib.itb.ac.id/gdl/view/15270
• Institution: Institut Teknologi Bandung

Information regarding the identity of a person's extremely important to be managed and known to be a database. Development of database technology today is not only limited to the data form of writing, but also with biometric data. Any technique to obtain the biometric data information about a person is to identify the image of the person's face. Many reasearch are conducted to identify a person's face is only limited to the grayscale images and the facial image that is used is one or two state variables. The purpose of this study is to design a face recognition system based on backpropagation neural network <br /> <br /> <br /> and edge detection using RGB images regardless of facial expression, face toward the camera position, rotation of the face, accessories on the face images, and environmental conditions. Face recognition system which is formed subsequently tested using the images training and images testing. The test results are then analyzed by using the calculation of the mean squared error and coefficient of determination. The test results by using the training images show the face recognition accuracy of 100% and the accuracy of face recognition with images testing of 43:75%.