Convolutional neural network-based feature extraction using multimodal for high security application
- PDF / 1,152,939 Bytes
- 11 Pages / 595.276 x 790.866 pts Page_size
- 97 Downloads / 259 Views
SPECIAL ISSUE
Convolutional neural network‑based feature extraction using multimodal for high security application Priti Shende1 · Yogesh Dandawate2 Received: 16 July 2020 / Revised: 30 September 2020 / Accepted: 30 October 2020 © Springer-Verlag GmbH Germany, part of Springer Nature 2020
Abstract An efficient biometrics-based security system is the prime need in modern security industry. Biometric modalities are unique features of any human being based on which a computer system can recognise, authenticate or verify a person. In this paper we propose a convolutional neural network-based face, fingerprint, palm vein identification system. Main purpose of this paper is to propose a convolutional neural network with minimum layers for face, fingerprint and palm vein, achieving high accuracy and reducing the complexity. The network is of two convolutional layers, two ReLU layers and two Maxpooling layesr with ten hidden layers in Fully connected layer. The dataset of 4500 images is generated for all the modalities. Dataset images are used for 60% training, 10% validation and testing 30%. Proposed CNN architecture’s accuracy is 95% for face, 94% for fingerprint and 99% palm-vein. The CNN used with minimum layers has performed consistently for all the biometric modalities maintaining good accuracy. Keywords Biometrics · Convolutional neural networks · Feature level extraction · High security
1 Introduction Biometric technology is used as a person’s identification and authentication tool. There is a growing demand for biometric-based authentication system in comparison with traditional e-cards, password and key-and-lock systems in homes, industry, defence, banking security systems. Biometric system is a pattern recognition tool which uses physical characteristics of human beings like a fingerprint, iris, palm-vein, face or behavioural characteristics like a signature, voice, gait for identification [1]. Multimodal biometric systems use multiple modalities of the same individual for processing the system. Multibiometric system improves the recognition rates. Using multiple traits, many security systems are developed, such as authentication and verification, forensic investigations, e-commerce. Banking frauds have * Priti Shende [email protected] Yogesh Dandawate [email protected] 1
Dr. D. Y. Patil Institute of Technology, Pimpri, Pune, Maharashtra, India
Vishwakarma Institute of Information Technology, Kondhwa, Pune, Maharashtra, India
2
been the prime motivation for the emergence of the biometric security system, therefore developing a robust biometric system for Automated Teller Machines is necessary where a person does not have to carry debit cards. The person’s body itself will act as a debit card for withdrawing the money; hence biometrics like palm-vein, fingerprint and face of a person is selected for authentication and identification. Face recognition is the most commonly used biometric; due to its accessibility of the biometric feature. The face is an identity of a person due to its uniqu
Data Loading...
