Dense Hand-CNN: A Novel CNN Architecture based on Later Fusion of Neural and Wavelet Features for Identity Recognition

Elaraby A Elgallad,Adel M,Wael Ouarda

doi:10.14569/ijacsa.2019.0100647

Elaraby A Elgallad, Adel M + Show 1 more

Open Access

https://doi.org/10.14569/ijacsa.2019.0100647

Copy DOI

Abstract

Biometric recognition or biometrics has emerged as the best solution for criminal identification and access control applications where resources or information need to be protected from unauthorized access. Biometric traits such as fingerprint, face, palmprint, iris, and hand-geometry have been well explored; and matured approaches are available in order to perform personal identification. The work emphasizes the opportunities for obtaining texture information from a palmprint on the basis of such descriptors as Curvelet, Wavelet, Wave Atom, SIFT, Gabor, LBP, and AlexNet. The key contribution is the application of mode voting method for accurate identification of a person at the fusion decision level. The proposed approach was tested in a number of experiments at the CASIA and IITD palmprint databases. The testing yielded positive results supporting the utilization of the described voting technique for human recognition purposes.

Highlights

Biometrics is an authentication method that uses human characteristics to identify a person
We introduced a novel approach in decision level technique, the Mode Voting Technique (MVT)
This work statements two palmprint recognition systems depending on the mode voting technique, and compares the performance of the systems for image processing time

Summary

Introduction

Biometrics is an authentication method that uses human characteristics to identify a person. Biometric can be divided into two broad types as in [1]: physical and behavioral. Physical biometric is a biometric system that evaluates the physical characteristic of a human body to recognize a person, such as fingerprint, face, retina, etc. Behavioral characteristic analyzes the human behavioral traits, such as gait, signature, keystroke, etc. Behavioral biometric is less secured than physical biometric because people can change their behavior anytime they want. People can adjust their signature, keystroke, or walking pattern . Multimodal biometric systems, which incorporate more than one biometric, with appropriate security measures are acknowledged as more robust and more accurate than unimodal biometrics, because even when the score of one biometric recognition is poor due to environmental conditions, the final outcome can be positive because the score from another biometric recognition is considered

Methods

Results

Discussion

Conclusion