Deep-Analysis of Palmprint Representation Based on Correlation Concept for Human Biometrics Identification

Deep-Analysis of Palmprint Representation Based on Correlation Concept for Human Biometrics Identification

Raouia Mokni (College of Computer Engineering and Science, Sattam Prince Bin Abdulaziz University, Kharj, Saudi Arabia & Faculty of Economics and Management of Sfax, University of Sfax, Sfax, Tunisia), Hassen Drira (IMT Lille Douai, Univ. Lille, CNRS, UMR 9189 -CRIStAL, Lille, France) and Monji Kherallah (Faculty of Sciences of Sfax, University of Sfax, Sfax, Tunisia)
Copyright: © 2020 |Pages: 19
DOI: 10.4018/IJDCF.2020040103

Abstract

The security of people requires a beefy guarantee in our society, particularly, with the spread of terrorism throughout the world. In this context, palmprint identification based on texture analysis is amongst the pattern recognition applications to recognize people. In this article, the researchers investigated a deep texture analysis for the palmprint texture pattern representation based on a fusion between several texture information extractions through multiple descriptors, such as HOG and Gabor Filters, Fractal dimensions and GLCM corresponding respectively to the frequency, model, and statistical methodologies-based texture features. They assessed the proposed deep texture analysis method as well as the applicability of the dimensionality reduction techniques and the correlation concept between the features-based fusion on the challenging PolyU, CASIA and IIT-Delhi Palmprint databases. The experimental results show that the fusion of different texture types using the correlation concept for palmprint modality identification leads to promising results.
Article Preview
Top

Introduction

The need for automated secure access to physical or virtual environments, especially for personalized services, is growing. These needs require reliable means to verify the identity of a person who is having access to the system. However, conventional means, such as passwords or magnetic cards associated with a personal code, have a number of shortcomings. For instance, a password may be forgotten or stolen by another individual, or even given to someone else and access cards may also be lost or stolen.

This is how the science of automatic identification of people through the exploitation of their unique physical and behavioral traits, such as face, signature, voice, fingerprints, shape of the hand, palmprint, etc., appeared the most reliable solution. Each of these different traits is called biometric modality.

Among these physical modalities, we are interested in the palmprint, which is the most mature biometric technology that has been used for more than a century. Hitherto, more attention has been paid to palmprint identification systems through research teams because it presents diverse advantages compared to other modalities in the biometric system such as its usability, stability over time, low cost, affordable materials, high recognition accuracy, etc.

The palmprint is particularly solicited in biometrics: it is the inside of the hand between the wrist and the roots of the fingers. It illustrates the most discriminating physical characteristic for the recognition of its skin patterns including both the principal lines and the texture representations, which are appropriated for the use of the real application within palmprint recognition.

In fact, an overview of the related work methods about palmprint identification applications is presented. Till now, these already existing applications can be divided into two popular categories, Global approach and Structural approach, based mainly on the extraction and the analysis of the different palmprint representations.

  • Structural approaches: They analyze the structure of the palmprint and include the palmprint lines, such as the principal lines (cf., (Han, Cheng, Lin, & Fan, 2003), (Mokni, Drira, & Kherallah, 2017a), etc.), the wrinkles (cf., (Chen, Zhang, & Rong, 2001), etc.), the ridges and minutiae (cf., (Duta, Jain, & Mardia, 2002), etc.) for represented the palmprint pattern.

Although this type of approach relies on taking into account the structural particularity of the palmprint representations since it is stable over time, unluckily, those representations alone cannot provide satisfactory information to identify the person efficiently.

  • Global approaches: They use the whole area of the palmprint as input to their recognition algorithm. Over recent years, researchers have been interested in this type of approach that presents different descriptors or methods to analyze the texture pattern of an image. In fact, different proposals have been presented. We distinguish three methodologies based on the texture feature analysis such as Statistical, Frequency and Model. Each one of them involves various descriptors.

In fact, (1) Statistical methodology includes Eigenpalms (cf.,(Lu, Zhang, & Wang, 2003), etc.), Fisherpalms (cf., (Wu, Zhang, & Wang, 2003), etc.), Local Binary Pattern (LBP) (cf., (Hammami, Ben-Jemaa, & Ben-Abdallah, 2014), etc.), Gray Level Co-occurrence Matrix (GLCM) (cf., (Latha & Prasad, 2015)), (Mokni, Elleuch, & Kherallah, 2016), etc.), (2) Frequency methodology involves different descriptors such as Gabor filters (GF) (cf., (Kumar & Shekhar, 2010), (Ben-Khalifa, Rzouga, & BenAmara, 2013), etc.), Scale-Invariant Feature Transform (SIFT) (cf., (Hammami et al., 2014)), Wavelets (cf., (Kekre, Sarode, & Tirodkar, 2012), etc.), and (3) Model methodology groups Blanket or Fractal, Multi-Fractal dimensions (cf., (Guo, Zhou, & Wang, 2014), etc.).

Complete Article List

Search this Journal:
Reset
Open Access Articles: Forthcoming
Volume 12: 4 Issues (2020): 3 Released, 1 Forthcoming
Volume 11: 4 Issues (2019)
Volume 10: 4 Issues (2018)
Volume 9: 4 Issues (2017)
Volume 8: 4 Issues (2016)
Volume 7: 4 Issues (2015)
Volume 6: 4 Issues (2014)
Volume 5: 4 Issues (2013)
Volume 4: 4 Issues (2012)
Volume 3: 4 Issues (2011)
Volume 2: 4 Issues (2010)
Volume 1: 4 Issues (2009)
View Complete Journal Contents Listing