Open AccessAn Approach for Pose Invariant Face Recognition System Using Log-Gabor Feature
Author(s) -
Madhura S. Shettar,
Poornima Byahatti
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/925/1/012030
Subject(s) - artificial intelligence , biometrics , pattern recognition (psychology) , computer science , facial recognition system , feature extraction , three dimensional face recognition , computer vision , classifier (uml) , face (sociological concept) , gabor wavelet , chin , euclidean distance , invariant (physics) , face detection , mathematics , medicine , social science , discrete wavelet transform , wavelet transform , sociology , wavelet , mathematical physics , anatomy
Biometrics is the process based on humans’ characters like behavioural and physical. Among the biometrics traits, face recognition has attained much importance for the authentication of individuals. Face recognition with pose variation is one of the major challenges among the illumination and expression. The system is widely used as surveillance which can withstand in places like ATMs, Airports, and Streets to track the criminals. The face recognizing mainly depends on the feature extraction from the face. Project aim is to extract those features which are invariant even though the pose changes. In the proposed work face recognition by using Log Gabor features is addressed. Firstly, colour images with variation in poses are captured followed by the extraction of facial features components like eyes, nose, mouth, and chin through Log Gabor features. The identification of the person is done using KNN classifier where the matching is carried out by calculating the Euclidean distance between the training and test pairs. Finally, experimental results show the performance of the system for different number of the users.