Research Article Pattern Mathematical Model for Fingerprint Security Using Bifurcation Minutiae Extraction and Neural Network Feature Selection Nesreen Alsharman , 1 Adeeb Saaidah , 2 Omar Almomani , 2 Ibrahim Jawarneh , 3 and Laila Al-Qaisi 2 1 Computer Science Department, e World Islamic Sciences Education University, Amman, Jordan 2 Computer Network and Information Systems Department, e World Islamic Sciences Education University, Amman, Jordan 3 Mathematics Department, Al-Hussein Bin Talal University, Ma’an, Jordan Correspondence should be addressed to Nesreen Alsharman; nesreen.alsharman@wise.edu.jo Received 11 November 2021; Revised 3 February 2022; Accepted 25 February 2022; Published 16 April 2022 Academic Editor: Luigi Catuogno Copyright © 2022 Nesreen Alsharman et al. is is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Biometric based access control is becoming increasingly popular in the current era because of its simplicity and user-friendliness. is eliminates identity recognition manual work and enables automated processing. e ﬁngerprint is one of the most important biometrics that can be easily captured in an uncontrolled environment without human cooperation. It is important to reduce the time consumption during the comparison process in automated ﬁngerprint identiﬁcation systems when dealing with a large database. Fingerprint classiﬁcation enables this objective to be accomplished by splitting ﬁngerprints into several categories, but it still poses some diﬃculties because of the wide intraclass variations and the limited interclass variations since most ﬁngerprint datasets are not categories. In this paper, we propose a classiﬁcation and matching ﬁngerprint model, and the classiﬁcation classiﬁes ﬁngerprints into three main categories (arch, loop, and whorl) based on a pattern mathematical model using GoogleNet, AlexNet, and ResNet Convolutional Neural Network (CNN) architecture and matching techniques based on bifurcation minutiae extraction. e proposed model was implemented and tested using MATLAB based on the FVC2004 dataset. e obtained result shows that the accuracy for classiﬁcation is 100%, 75%, and 43.75% for GoogleNet, ResNet, and AlexNet, respectively. e time required to build a model is 262, 55, and 28 seconds for GoogleNet, ResNet, and AlexNet, respectively. 1. Introduction Biometrics science is used to identify people using their physical characteristics. ese characteristics are ﬁngerprint, iris, palm, face, DNA, and voice [1]. Among these charac- teristics, the ﬁngerprint is one the most accurate and reliable for identifying a person [2] since ﬁngerprints are the unique biometric characteristics of any person; therefore, it is used in forensic divisions worldwide for criminal investigations where even the twins have nonidentical ﬁngerprints. erefore, ﬁngerprints have been conﬁrmed to be good and secure biometrics. e process of ﬁngerprint identiﬁ- cation is to conﬁrm or refuse if a scanned ﬁngerprint belongs to a speciﬁc person or not. e increasing commercial applications and number of civilians that depend on ﬁn- gerprint-based identiﬁcation lead to a huge ﬁngerprint database. Matching speciﬁc ﬁngerprints stored in the da- tabase is computationally time-consuming. e subject of automatic ﬁngerprint identiﬁcation has received intensive attention among researchers. To solve automatic ﬁngerprint identiﬁcation, ﬁnger- prints can be stored in databases based on the characteristics of their ridge and furrow patterns. In general, ﬁngerprints can be divided into three major classes known as whorl (W), loop (L), and arch (A) according to Galton [3]. e Galton classiﬁcation scheme is shown in Figure 1. e database of Hindawi Security and Communication Networks Volume 2022, Article ID 4375232, 16 pages https://doi.org/10.1155/2022/4375232