Research on Rapid Iris Localization Algorithm Based on Projection Analysis
<p>Nan Liu</p> · Academic Journal of Computing & Information Science · 2025
With the accelerated advancement of global digitalization, information interconnectivity has become a fundamental characteristic of modern society. During this transformative process, the security vulnerabilities of traditional identity authentication systems have become increasingly prominent: current systems primarily rely on physical credentials and multi-factor password mechanisms. These static verification methods not only suffer from inherent weaknesses such as susceptibility to loss and forgery but may also lead to severe personal privacy breaches and societal security risks due to information leakage. In the face of increasingly sophisticated cybersecurity threats, the development of novel identity authentication technologies has become an urgent necessity for ensuring digital security. As a critical direction in identity authentication, biometric technology verifies identity by analyzing individuals' unique physiological or behavioral characteristics. Among these, iris recognition technology demonstrates significant technical superiority and broad application prospects due to its liveness detection capability, contactless acquisition, high uniqueness, long-term stability, and inherent anti-counterfeiting properties. However, existing iris recognition technologies still face major challenges in practical applications: not all scenarios can provide sufficient computational resources or deployment space. To address this critical issue, this study innovatively proposes a longitudinal gray-scale integration-based iris segmentation method. This approach can operate efficiently on miniaturized PCs or embedded processors while achieving precise center localization, thereby providing a reliable foundation for subsequent iris texture recognition.