Utilization of Double Random Phase Encoding and Local binary pattern for Cancelable Face Recognition System
Paper ID : 1105-ICEEM2023 (R1)
Authors
Samy Elbeheiry *
Faculty of Electronic Engineering, Menoufia University Menouf, Egypt
Abstract
The widespread adoption of biometric technologies
for human authentication has led to concerns about privacy
and anonymity. To address the risk of biometric template theft,
this paper proposes a novel cancelable face pattern generation
technique based on Component Local Binary Pattern (LBP) and
the Optical Double Random Phase Encoding (DRPE) algorithm.
LBP extracts essential facial features. The Optical DRPE algorithm then applies encryption, making the transformed biometric
templates resistant to reverse engineering.To quantitatively assess
the performance of the proposed cancelable face recognition
method, various metrics have been employed. The evaluation
includes calculating the Equal Error Rate (EER)based on the
false rejected rate (FRR), the false acceptance rate, and the
Area Under the Receiver Operating Characteristic(ROC) curve
(AROC). Evaluation on diverse face databases demonstrates
the method’s effectiveness in enhancing security and achieving
improved Receiver Operating Characteristic (ROC) performance
compared to traditional methods
Keywords
Biometric, Cancelable Biometric, Scale Invariant Feature Transform, Local Binary Pattern
Status: Accepted