SISTEM PENGENALAN WAJAH SECARA REAL TIME MENGGUNAKAN METODE CONVOLUTIONAL NEURAL NETWORK DAN SUPPORT VECTOR MACHINE
Total View This Week0

Institusion

Universitas Sriwijaya

Author

ALFIANSYAH ALFIANSYAH (STUDENT ID : 09011281520131)
Sri Desy Siswanti (LECTURER ID : 0007127410)
Huda Ubaya (LECTURER ID : 0216068101)

Subject

T1-995 Technology (General) 

Datestamp

2021-01-12 02:39:05 

Abstract :
Face recognition system is a branch of biometric science that is useful for identifying someone based on their physical characteristics, in this case their face. In general, face recognition system consists of three stages, face detection, feature extraction and face recognition. In this study Multi Task Cascade Convolutional Neural Network was used for the face detection process, then VGGFace for the feature extraction process and Support Vector Machine for the classification process. The system is trained and tested using 1000 data from 10 different people with a comparison of 8: 2 training and test data. From the results of the system testing of the data, the system accuracy is 98% with 38s processing time. Then from the results of realtime system testing, the accuracy of the system is 96.67% with an average processing time of 0.155s per data. 

File :
RAMA_56201_09011281520131.pdf
RAMA_56201_09011281520131_TURNITIN.pdf
RAMA_56201_09011281520131_0007127410_0216068101_01_front_ref.pdf
RAMA_56201_09011281520131_0007127410_0216068101_02.pdf
RAMA_56201_09011281520131_0007127410_0216068101_03.pdf
RAMA_56201_09011281520131_0007127410_0216068101_04.pdf
RAMA_56201_09011281520131_0007127410_0216068101_05.pdf
RAMA_56201_09011281520131_0007127410_0216068101_06_ref.pdf
RAMA_56201_09011281520131_0007127410_0216068101_07_lamp.pdf