Classification of Multimodal Biometric Recognition System Using Fuzzy Logic

Resumen

Nowadays, biometric technology plays a major role in secure identity verification yet, they still face challenges like as spoofing attacks, intra-class variations, and noisy input data. In order to solve these issues, this paper proposes a multimodal biometric classification system that improves precision and resilience using fuzzy logic. The system extracts fingerprint patterns using Local Binary Patterns and extracts textural information from iris images using Gabor filters to merge fingerprint and iris identification. Fuzzy logic is essential for this model because it manages biometric data's variability and uncertainty well, something that conventional approaches frequently find difficult. The system is better able to manage noisy data and possible spoofing attempts by implementing a rule-based decision-making framework that is specific to the characteristics of biometric traits.

Our experimental assessment shows that this method works well, with a classification accuracy of 99.7%, outperforming both traditional unimodal and multimodal methods. The ability to combine a variety of biometric features with intelligent decision-making is a higher performance accuracy. Overall, the proposed method is a reliable and adaptable option for identity verification, especially in situations where security and accuracy are crucial. This work provides a practical and creative method for enhancing multimodal biometric systems by fusing the flexibility of fuzzy logic with the benefits of fingerprint and iris biometrics.