Advanced Research on High-security-level Error-correction-based Iris Recognition System
Kuo-Chun Lin, Yen‐Ming Chen · 2023
Over recent years, biometric systems have been popularly applied in various scenarios. Unfortunately, potential threats associated with these systems may lead to significant social concerns and pose risks to security. In this paper, an error-correction-based iris recognition (EC-IR) scheme which uses multi-scale dominating feature points (msDFPs) is produced to improve the security level against a stricter concern of error-correction-based attack. The EC-based attack can be initiated if an attacker is thoroughly knowledgeable about the structure of the adopted error correcting code. Hence, the system can be compromised with much fewer attempts compared to the brute force attack. To improve the number of security bits under this attack, the extraction method of the msDFPs is proposed to modify the essence of original iris data. As a result, it has been demonstrated that the proposed msDFPs-based EC-IR scheme provides a well-balanced design that ensures both a large value of security bits and an optimized recognition performance.