Securing Face Recognition System Using Blockchain Technology

Saumya Shankar,Jitendra Madarkar,Poonam Sharma
DOI: https://doi.org/10.1007/978-981-15-6318-8_37
2020-01-01
Abstract:AbstractFacial recognition is a wide area of a computer vision which is mostly used for security purpose. The main motive of the face recognition system is to authentic a person from a given training database. The face-based biometric system and surveillance cameras are deployed everywhere, for that need a strong face recognition system. The recognition system needs a large number of training samples that need to store in any storage center but when the face data is collected and stored, that time hackers can access and manipulate that data. So, a platform is presented that is secure and tamper-proof from these data breaches as well as hacks and is not compromising with data availability, by using blockchain to store face images. The absolute infeasibility of editing a historical record in a blockchain makes it tamper-proof (immutable) ensuring security. The storage of data on multiple computers provides accessibility. For face recognition, VGGFace deep neural network is used for automatic extraction of features and logistic regression for classification. documentclass[12pt]{minimal}usepackage{amsmath}usepackage{wasysym}usepackage{amsfonts}usepackage{amssymb}usepackage{amsbsy}usepackage{mathrsfs}usepackage{upgreek}setlength{oddsidemargin}{-69pt}egin{document}$$ldots $$end{document}
What problem does this paper attempt to address?