An effective biometric identification system using a new image Fuzzy modeling for reliable feature extraction
Zakaria Tidjani, Salim Chitroub, Abdallah Meraoumia, Mokhtar Smahi · 2017
Feature extraction for an optimal data representation is crucial for any biometric identification system. A new fuzzy-based feature extraction algorithm for an effective biometric identification system is proposed in this paper. The Sugeno-Takagi Fuzzy System is used for the image modeling. The system parameter are systematically adjusted by using a recursive algorithm. The proposed algorithm is integrated in a new biometric identification system that we have developed. The CASIA multispectral palmprint database is used for the evaluation process. The obtained results have shown that the proposed systems outperform many other systems developed in the literature.