Smart face detection and recognition in Pose variant images using Alexnet CNN compare accuracy with SVM

Abstract

Many face recognition tasks have been carried out in controlled environments, compared to controlled environments face recognition in uncontrolled environments like pose variant images is the most challenging task. The main aim of this work is to recognize the faces in pose variant images with different arbitrary rotations using two different machine learning algorithms Alexnet convolutional neural network (ACNN) and Support Vector Machine (SVM) are trained with 20 samples of pose variant images dataset each, totally 40 samples to recognize the person.: From Matlab simulation results ACNN has achieved better accuracy(96 <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">%</sup> ) compared to SVM(89%). Attained Significance value (P!O.000) using SPSS analysis. From this study it shows that ACNN has higher accuracy than SVM.