Open AccessExpression Invariant Features for Face Recognition
Author(s) -
Mayank Mohit,
Neralla Harichandana,
Pendem Bhagyasri,
Praveen Kumar
Publication year - 2019
Publication title -
international journal of engineering and advanced technology
Language(s) - English
Resource type - Journals
ISSN - 2249-8958
DOI - 10.35940/ijeat.b3099.129219
Subject(s) - sadness , surprise , facial expression , anger , computer science , facial recognition system , artificial intelligence , face (sociological concept) , invariant (physics) , expression (computer science) , identifier , biometrics , emotion classification , three dimensional face recognition , speech recognition , pattern recognition (psychology) , computer vision , face detection , psychology , mathematics , communication , social psychology , linguistics , philosophy , mathematical physics , programming language
Personal Computer sourced Face Recognition has been a sophisticated and well-found technique which is being rationally utilized for most of the authenticated cases. In reality, there is a number of situations where the expressions of the face will be different. We are here able to instinctively detect the five universal expressions: smile, sadness, anger, surprise, neutral by studying face geometry by determining which type of facial expression has been carried out. Using some facial data with variant expressions. We hereby made some experimentations to calculate the accuracies of some machine learning methods by making some changes in the face images such as a change in expressions, which at last needed for training and recognition identifiers. Our objective is to take the features of neutral facial expressions and add them with the other expressive face images like smiling, angry, sadness to improve the accuracy.