Open AccessOPTIMIZED PCA BASED FACE RECOGNITION
Author(s) -
Rayaan Grewal
Publication year - 2021
Publication title -
international journal of research in informative science application and techniques
Language(s) - English
Resource type - Journals
ISSN - 2581-5814
DOI - 10.46828/ocpu/ijrisat/21577
Subject(s) - eigenface , facial recognition system , particle swarm optimization , artificial intelligence , computer science , face (sociological concept) , pattern recognition (psychology) , principal component analysis , novelty , fitness function , similarity (geometry) , computer vision , machine learning , genetic algorithm , image (mathematics) , social science , philosophy , theology , sociology
In this paper an algorithm to solve the problem of automatic face recognition ispresented. The novelty of the algorithm is the ability to combine the principalcomponent analysis (PCA) with Modified Particle Swarm Optimization (MPSO) toimprove the execution time and to obtain better face recognition results. Theefficiency of face recognition system is impmeasure the similarity of an input face compared with a database of faces. The use ofthe fitness function helps to obtain more accurate results in a faster way. The resultsobtained are excellent even when the system wA comparison of the results obtained with the algorithm without MPSO versus thealgorithm using MPSO is also presented. The algorithm is also implemented on thecolored images of the human faces.Keywords: Particle Swarm Optimization, Face Recognition, Eigenfaces, EvolutionaryComputer Vision.