Deep Multi-Biometric Fuzzy Commitment Scheme: Fusion Methods and Performance

Abstract Biometric cryptosystems offer privacy-preserving authentication using biometric data, such as fingerprints or iris scans. However, single modalities suffer from limited entropy, impacting both recognition performance and security. This work investigates the fusion of multiple biometric characteristics to address these limitations. This study provides a detailed description of the Deep Multi-biometric Fuzzy Commitment Scheme and its components. In the experimental setup, we demonstrate how Convolutional Neural Networks (CNNs) are used to tackle the challenge of non-uniform representation vectors by generating uniform embeddings. This process generates the monomodal databases of iris and fingerprint vectors, as well as the multibiometric database. We then evaluate our three proposed fusion methods: concatenation, interleaving, and random shuffling within the fuzzy commitment scheme using error correction methods such as Hadamard and Reed-Solomon codes. The evaluation of performance and security reveals that random shuffling outperforms other methods like interleaving and concatenation in terms of recognition performance. Concatenation displayed the lowest performance. Finally, the findings are summarized and potential improvements are discussed.