Classification technique for face-spoof detection in artificial neural networks using concepts of machine learning

Abstract

In biometric technology, face recognition techniques are considered the most signif icant research area. This technology is abundantly used in security services, smart cards, surveillance, social media, and ID verification. The number of countermea sures is gradually increasing, and many systems have been initiated to distinguish genuine access and fake attacks. In our paper, we propose a Convolutional Neu ral Network (CNN), which can obtain fine distinctions and abilities in a supervised manner. Deep convolutional neural networks have prompted a progression of break throughs for image classification. This paper introduces various architectures of CNN for detecting face spoofing using many convolutional layers. We have used VGG-16 under Convolutional Neural Networks (CNN) architecture in the proposed system for learning about the feature classification. Our proposed system has show cased an accuracy of 98% for Convolutional Neural Network (CNN), 63% for VGG16,and 50% for Support Vector Machine (SVM) respectively.