Alignment-Free Cancelable Iris Key Binding Scheme

• Main Author: Ch’ng, Kai Liang
• Format: Final Year Project / Dissertation / Thesis
• Published: 2020
• Subjects: TJ Mechanical engineering and machinery
• Online Access: http://eprints.utar.edu.my/4223/1/1503759_FYP_report_%2D_Kai_Liang_Chng.pdf; http://eprints.utar.edu.my/4223/

The application of iris in the usage of biometric recognition systems has gain popularity over the years. It has raised the public concern on the reliability and consistency that could be achieved by using iris as part of the biometric recognition. This study aims to determine how the proposed framework performed when it is tested with a different database and the strength of the security. The proposed framework consists of 3 different processes, which include the Bloom Filter, Indexed First One Hashing, and the Key Binding Process. Four different databases have been used to test out the framework and the security’s strength. Initial parameters have been optimized through extensive trials with the aids of using the Taguchi Method. The three main parameters that are crucial to determine the performance and security strength have been tested and the results have been tabulated in this report. The results indicate that the database with better quality will show much more promising results in terms of performance. The strength of the security is partially dependent on the performance. The trade-off between the performance and strength of the security can be observed across four different databases. Further research with a more recent public database that shall have better quality is needed to fully identify the performance and strength of the security for the proposed methods.