Development Of An Enhanced Self-Organizing Feature Map Iris-Based Access Control System

Access control systems play a critical role in maintaining data integrity and confidentiality. Utilizing biometric recognition improves security by utilizing distinct behavioral and physiological characteristics. This study aims to address issues of accuracy and computing efficiency by creating a sophisticated access control system with iris-based biometrics and a novel fusion of Chicken Swarm Optimization (CSO) and Self-Organizing Feature Maps (SOFM). Iris images were acquired, preprocessed, and subjected to feature extraction via Log-Gabor filters to highlight critical attributes. The CSO algorithm optimized the SOFM's learning parameters, resulting in an improved clustering and recognition process. Comparative analysis against the standard SOFM demonstrated significant enhancements in sensitivity (96.09%), specificity (90.59%), precision (96.85%), and accuracy (94.72%). Additionally, processing time was reduced, ensuring feasibility for real-time applications. These results establish the Modified CSO-SOFM as a robust and efficient method for biometric access control. Future efforts will focus on incorporating multimodal biometrics and advanced optimization strategies to further elevate system performance. Keywords: Enhanced Self-Organizing Feature Map, Iris-Based Access Control System, Biometrics CISDI Journal Reference Format Jeremiah Y.S., Ojo O. J., Adigun E. B. & Ogunkan S. K. (2025): Development Of An Enhanced Self-Organizing Feature Map Iris-Based Access Control System. Computing, Information Systems, Development Informatics & Allied Research Journal. Vol 16 No 1, Pp 35-44. Available online at www.isteams.net/cisdijournal. dx.doi.org/10.22624/AIMS/CISDI/V16N1P5