scholarly journals Novel Biometric Fusion System using GA-PSO and ANN

2020 ◽  
Vol 9 (4) ◽  
pp. 3122-3129
Keyword(s):  
Rejection Rate ◽  
Fusion System ◽  
Scale Invariant ◽  
Fusion Strategy ◽  
Bee Colony ◽  
Face Features ◽  
Authentication System ◽  
Nature Inspired Algorithms ◽  
High Level ◽  
Biometric Fusion

Technology advancements have led to the emergence of biometrics as the most relevant future authentication technology. On practical grounds, unimodal biometric authentication systems have inevitable momentous limitations due to varied data quality and noise levels. The paper aims at investigating fusion of face and fingerprint biometric characteristics to achieve a high level personal authentication system. In the fusion strategy face features are extracted using Scale-Invariant Feature Transform (SIFT) algorithm and fingerprint features are extracted using minutiae feature extraction. These extracted features are optimized using nature inspired Genetic Algorithm (GA). The efficiency of the proposed fusion authentication system is enhanced by training and testing the data by applying Artificial Neural Network (ANN). The quality of the proposed design is evaluated against two nature inspired algorithms, namely, Particle Swarm Optimization (PSO)and Artificial Bee Colony (ABC) in terms of False Acceptance Rate (FAR), False Rejection Rate (FRR) and recognition accuracy. Simulation results over a range of image sample from 10 to 100 images have shown that the proposed biometric fusion strategy resulted in FARof 2.89, FAR 0.71and accuracy 97.72%.Experimental evaluation of the proposed system also outperformed the existing biometric fusion system.

BRAIN SIGNATURES PERSPECTIVE FOR HIGH-SECURITY AUTHENTICATION

2020 ◽  
Vol 32 (04) ◽  
pp. 2050025
Author(s):  
Nikhil Rathi ◽  
Rajesh Singla ◽  
Sheela Tiwari
Keyword(s):  
Classification Accuracy ◽  
Information Transfer ◽  
Rejection Rate ◽  
Internet Banking ◽  
P300 Speller ◽  
Security Authentication ◽  
Information Transfer Rate ◽  
Shoulder Surfing ◽  
Authentication System ◽  
High Level

In the recent past, the web (internet) has emerged as the most interactive authentication system for all of us (i.e. Internet banking passwords, system or building access, and e-payment platforms, etc.) and as a result, traditional authentication systems (like passwords or token-based) are never again more secure i.e. they are vulnerable to attacks. As a result, the security of individual information and safe access to a system winds up prime necessities. Therefore, the EEG-based authentication system has recently become a reasonable key for high-level security. This study centers upon P300 evoked potential-based authentication system designing. In this paper, a new visual stimulus paradigm (i.e. [Formula: see text] P300 speller) using pictures of different objects as stimuli for a person authentication system is designed instead of the conventional character-based paradigm (i.e. [Formula: see text] speller) for increasing the classification accuracy and Information Transfer Rate (ITR). The trial begins by exhibiting a collection of pictures of various objects on four corners of the PC screen comprising of random object pictures (non-target) alongside password pictures (target) that trigger P300 reactions. The P300 reaction’s rightness then checks the identity of the subject concerning the focused pictures (Target). The proposed investigation model achieves higher classification accuracy of 96.78%, along with 0.03075 False Rejection Rate (FRR), 0.03297 False Acceptation Rate (FAR), and ITR of [Formula: see text]. This study has shown that P300-based authentication system has an advantage over conventional methods (Password, Token, etc.) as EEG-based systems cannot be mimicked or forged (like Shoulder surfing in case of password) and can still be used for disabled users with a brain in good running condition. The classification results revealed that the performance of the QDA classifier outperformed other classifiers based on accuracy and ITR.

scholarly journals THREE –WAY SECURITY USING IMAGE BASED AUTHENTICATION SYSTEM

2016 ◽  
pp. 46-47
Author(s):  
SARIKA KHALADKAR ◽  
SARITA MALUNJKAR ◽  
POOJA SHINGOTE
Keyword(s):  
Instant Messaging ◽  
Image Authentication ◽  
Control Mechanisms ◽  
Authentication System ◽  
One Time Password ◽  
Personal Accounts ◽  
Brute Force Attack ◽  
The One ◽  
High Level ◽  
User Friendly

Secure environments protect their resources against unauthorized access by enforcing access control mechanisms. So when increasing security is an issue text based passwords are not enough to counter such problems. The need for something more secure along with being user friendly is required. This is where Image Based Authentication (IBA) comes into play. This helps to eliminate tempest attack, shoulder attack, Brute-force attack. Using the instant messaging service available in internet, user will obtain the One Time Password (OTP) after image authentication. This OTP then can be used by user to access their personal accounts. The image based authentication method relies on the user’s ability to recognize pre-chosen categories from a grid of pictures. This paper integrates Image based authentication and HMAC based one time password to achieve high level of security in authenticating the user over the internet.

scholarly journals Optimization of Classified Satellite Images using DWT and Fuzzy Logic

2013 ◽  
Vol 2 (02) ◽  
pp. 08-12
Author(s):  
Mr. Kaustubh Patil
Keyword(s):  
Satellite Images ◽  
Clustering Algorithm ◽  
Speckle Noise ◽  
Intermediate Stage ◽  
Sar Images ◽  
Fusion Strategy ◽  
Difference Image ◽  
Fuzzy Clustering Algorithm ◽  
High Level ◽  
Fuzzy C Means Algorithm

The image taken by a satellite can be enhanced in terms of its resolution based on the interpolation can be obtained by DWT. Using DWT, the image at the input is divided into several sub bands and the speckle noise is also removed. Thereafter, the high-level images and low-level image at the input can be combined, to produce a better image applying IDWT. An intermediate stage for approximating high level is proposed here. The variation in detection approaches for SAR images are done by using image fusion strategy and novel fuzzy clustering algorithm. To retrieve an enhanced image, wavelet fusion directives are considered to combine the wavelet coefficients. A fuzzy C-means algorithm is proposed for identifying the altered and unaltered regions in the combined difference image.

Stellar Mass Black Hole for Engineering Optimization

2017 ◽  
pp. 62-90
Author(s):  
Premalatha Kandhasamy ◽  
Balamurugan R ◽  
Kannimuthu S
Keyword(s):  
Black Hole ◽  
Artificial Bee Colony ◽  
Optimization Problems ◽  
Optimal Solution ◽  
Cuckoo Search ◽  
Optimization Method ◽  
Stellar Mass ◽  
Bee Colony ◽  
Nature Inspired Algorithms ◽  
The Universe

In recent years, nature-inspired algorithms have been popular due to the fact that many real-world optimization problems are increasingly large, complex and dynamic. By reasons of the size and complexity of the problems, it is necessary to develop an optimization method whose efficiency is measured by finding the near optimal solution within a reasonable amount of time. A black hole is an object that has enough masses in a small enough volume that its gravitational force is strong enough to prevent light or anything else from escaping. Stellar mass Black hole Optimization (SBO) is a novel optimization algorithm inspired from the property of the gravity's relentless pull of black holes which are presented in the Universe. In this paper SBO algorithm is tested on benchmark optimization test functions and compared with the Cuckoo Search, Particle Swarm Optimization and Artificial Bee Colony systems. The experiment results show that the SBO outperforms the existing methods.

Security of Multimodal Biometric Fusion System

2011 ◽  
Vol 5 (4) ◽  
pp. 264-270
Author(s):  
Zhifang Wang ◽  
Shuangshuang Wang ◽  
Qun Ding
Keyword(s):  
Fusion System ◽  
Biometric Fusion

A Generalized and Robust Anti-Predatory Nature-Inspired Algorithm for Complex Problems

2019 ◽  
Vol 10 (1) ◽  
pp. 75-91 ◽  
Author(s):  
Rohit Kumar Sachan ◽  
Dharmender Singh Kushwaha
Keyword(s):  
Standard Deviation ◽  
Defense Mechanism ◽  
Mathematical Formulation ◽  
Predatory Behavior ◽  
Rank Test ◽  
Bee Colony ◽  
Self Defense ◽  
Nature Inspired Algorithm ◽  
Nature Inspired Algorithms ◽  
Result Analysis

This article describes how nature-inspired algorithms (NIAs) have evolved as efficient approaches for addressing the complexities inherent in the optimization of real-world applications. These algorithms are designed to imitate processes in nature that provide some ways of problem solving. Although various nature-inspired algorithms have been proposed by various researchers in the past, a robust and computationally simple NIA is still missing. A novel nature-inspired algorithm that adapts to the anti-predatory behavior of the frog is proposed. The algorithm mimics the self defense mechanism of a frog. Frogs use their reflexes as a means of protecting themselves from the predators. A mathematical formulation of these reflexes forms the core of the proposed approach. The robustness of the proposed algorithm is verified through performance evaluation on sixteen different unconstrained mathematical benchmark functions based on best and worst values as well as mean and standard deviation of the computed results. These functions are representative of different properties and characteristics of the problem domain. The strength and robustness of the proposed algorithm is established through a comparative result analysis with six well-known optimization algorithms, namely: genetic, particle swarm, differential evolution, artificial bee colony, teacher learning and Jaya. The Friedman rank test and the Holm-Sidak test have been used for statistical analysis of obtained results. The proposed algorithm ranks first in the case of mean result and scores second rank in the case of “standard deviation”. This proves the significance of the proposed algorithm.

A user-specific and selective multimodal biometric fusion strategy by ranking subjects

2013 ◽  
Vol 46 (12) ◽  
pp. 3341-3357 ◽  
Author(s):  
Norman Poh ◽  
Arun Ross ◽  
Weifeng Lee ◽  
Josef Kittler
Keyword(s):  
Fusion Strategy ◽  
Biometric Fusion

scholarly journals New Authentication Model for Multimodal Biometrics Based on Shape Features Vectors

2019 ◽  
Vol 54 (6) ◽  
Author(s):  
Sundos Abdulameer Alazawi ◽  
Huda Abdulaaliabdulbaqi ◽  
Yasmin Makki Mohialden
Keyword(s):  
High Performance ◽  
Feature Fusion ◽  
Personal Data ◽  
Rejection Rate ◽  
Shape Features ◽  
Multimodal Biometrics ◽  
Invariant Moments ◽  
Measurement And Analysis ◽  
Authentication System ◽  
Form Features

Biometrics is the science and technology dealing with the measurement and analysis of the biological features of the human body. The analysis is based on comparing the value of certain measured features with the form features in the database. Unimodal Biometric Systems have many limitations regarding precision in the identification/authentication of personal data. To accurately identify a person, a multimodal biometrics system such as combining face and fingerprint characteristic is used. Many such multi-biometrics fusion possibilities exist that can be utilized as an authentication system. In this paper, we present a new authentication system of the multimodal biometrics method for both face and fingerprint characteristics based on general shape feature fusion vectors. There are two main phases in our method: first, the fused shape features for both face and fingerprint images are extracted in accordance with central moments, and second, these features were recognized for retrieval of an authorized person using direct Euclidian distance. Experimentally, we tested about 100 shape features vectors, and observed that our method allows to improve the multimodal biometrics model when we are using the same features for two biometric images. A new method has a high-performance precision when invariant moments are used to extract shape features vectors and when similarity measurements computed based on direct Euclidean distance in the experiments are performed. We recorded False Acceptance Rate, False Rejection Rate, and Accuracy, FAR and FRR where the accuracy of the model is 91 %.

Sign in / Sign up

Export Citation Format

Share Document