Open AccessAnalysis SURF feature extraction and SVM classification for the facial image recognition from various angles
Author(s) -
Aripin Rambe,
Poltak Sihombing,
. Tulus
Publication year - 2020
Publication title -
iop conference series. materials science and engineering
Language(s) - English
Resource type - Journals
eISSN - 1757-899X
pISSN - 1757-8981
DOI - 10.1088/1757-899x/725/1/012138
Subject(s) - support vector machine , biometrics , artificial intelligence , pattern recognition (psychology) , feature extraction , computer science , face (sociological concept) , facial recognition system , feature (linguistics) , image (mathematics) , identification (biology) , feature vector , computer vision , social science , linguistics , philosophy , botany , sociology , biology
Biometric is a series of procedures used to measure the physical properties of a person based on your physical characteristics of a person’s behavior in identification and verification. One facial biometrics ie, feature extraction Speed Up Robust Feature (SURF) will be suitably used for extracting the characteristics of the face image. Support Vector Machine (SVM) will be used as a method of classification. The face data used in this study were obtained from the National Cheng Kung University (NKCU). SVM classification results with the help of SURF as a model feature extraction with the determination of the number of k = 50 gained 94.60% accuracy rate, k = 500 acquire a 100% accuracy rate and the number of k = 1000 classification results decreased with 93.70% accuracy rate.