Face Detection and Recognition with improved accuracy using Principal Component Analysis and comparing with Symlet algorithm

Abstract

The purpose of this research is to analyse the performance of the principal component analysis (PCA) method and the SYMLET algorithm when it comes to detecting and recognising faces with a high degree of accuracy. G-power is a tool that determines the total number of samples necessary for good face detection utilising the PCA and SYMLET algorithm. When certain parameters, such as minimum power, acceptable error rate, and allocation ratio, are held constant at 0.8, 0.05, and 1 respectively, a total of 256 samples can be produced from the experiment. It has been determined that PCA has an accuracy of 96.97%, and SYMLET has an accuracy of 52.63% correspondingly. There is a significant difference of 0.03 between the two groups (p less than 0.05). It has been determined from the findings of this research that the PCA algorithm detects faces in a substantially more accurate manner than the SYMLET algorithm does.