Amina Bassit

Modern biometric recognition systems extract distinctive feature vectors of biometric samples using deep neural networks to measure the amount of (dis-)similarity between two biometric samples. Studies have shown that personal information (e.g., health condition, ethnicity, etc.) can be inferred, and biometric samples can be reconstructed from those feature vectors, making their protection an urgent necessity. State-of-the-art biometrics protection solutions are based on homomorphic encryption (HE) to perform recognition over encrypted feature vectors, hiding the features and their processing while releasing the outcome only. However, this comes at the cost of those solutions’ efficiency due to the inefficiency of HE-based solutions with a large number of multiplications; for (dis-)similarity measures, this number is proportional to the vector’s dimension. In this paper, we tackle the HE performance bottleneck by freeing the two common (dis-)similarity measures, the cosine similarity and the squared Euclidean distance, from multiplications. Assuming normalized feature vectors, our approach pre-computes and organizes those (dis-)similarity measures into lookup tables. This transforms their computation into simple table lookups and summations only. We integrate the table lookup with HE and introduce pseudo-random permutations to enable cheap plaintext slot selection, which significantly saves the recognition runtime and brings a positive impact on the recognition performance. We then assess their runtime efficiency under encryption and record runtimes between 16.74ms and 49.84ms for both the cleartext and encrypted decision modes over the three security levels, demonstrating their enhanced speed for a compact encrypted reference template reduced to one ciphertext.

Most deep learning‐based image classification models are vulnerable to adversarial attacks that introduce imperceptible changes to the input images for the purpose of model misclassification. It has been demonstrated that these attacks, targeting a specific model, are transferable among models performing the same task. However, models performing different tasks but sharing the same input space and model architecture were never considered in the transferability scenarios presented in the literature. In this paper, this phenomenon was analysed in the context of VGG16‐based and ResNet50‐based biometric classifiers. The authors investigate the impact of two white‐box attacks on a gender classifier and contrast a defence method as a countermeasure. Then, using adversarial images generated by the attacks, a pre‐trained face recognition classifier is attacked in a black‐box fashion. Two verification comparison settings are employed, in which images perturbed with the same and different magnitude of the perturbation are compared. The authors’ results indicate transferability in the fixed perturbation setting for a Fast Gradient Sign Method attack and non‐transferability in a pixel‐guided denoiser attack setting. The interpretation of this non‐transferability can support the use of fast and train‐free adversarial attacks targeting soft biometric classifiers as means to achieve soft biometric privacy protection while maintaining facial identity as utility.

Bloom filters (BFs) and homomorphic encryption (HE) are prominent techniques used to design biometric template protection (BTP) schemes that aim to protect sensitive biometric information during storage and biometric comparison. However, the pros and cons of BF‐ and HE‐based BTPs are not well studied in literature. We investigate the strengths and weaknesses of these two approaches since both seem promising from a theoretical viewpoint. Our key insight is to extend our theoretical investigation to cover the practical case of iris recognition on the ground that iris (1) benefits from the alignment‐free property of BFs and (2) induces huge computational burdens when implemented in the HE‐encrypted domain. BF‐based BTPs can be implemented to be either fast with high recognition accuracy while missing the important privacy property of ‘unlinkability’, or to be fast with unlinkability‐property while missing the high accuracy. HE‐based BTPs, on the other hand, are highly secure, achieve good accuracy, and meet the unlinkability‐property, but they are much slower than BF‐based approaches. As a synthesis, we propose a hybrid BTP scheme that combines the good properties of BFs and HE, ensuring unlinkability and high recognition accuracy, while being about seven times faster than the traditional HE‐based approach.

Biometric verification has been widely deployed in current authentication solutions as it proves the physical presence of individuals. Several solutions have been developed to protect the sensitive biometric data in such systems that provide security against honest-but-curious (a.k.a. semi-honest) attackers. However, in practice, attackers typically do not act honestly and multiple studies have shown severe biometric information leakage in such honest-but-curious solutions when considering dishonest, malicious attackers. In this paper, we propose a provably secure biometric verification protocol to withstand malicious attackers and prevent biometric data from any leakage. The proposed protocol is based on a homomorphically encrypted log likelihood-ratio (HELR) classifier that supports any biometric modality (e.g., face, fingerprint, dynamic signature, etc.) encoded as a fixed-length real-valued feature vector. The HELR classifier performs an accurate and fast biometric recognition. Furthermore, our protocol, which is secure against malicious adversaries, is designed from a protocol secure against semi-honest adversaries enhanced by zero-knowledge proofs. We evaluate both protocols for various security levels and record a sub-second speed (between 0.37s and 0.88s) for the protocol secure against semi-honest adversaries and between 0.95s and 2.50s for the protocol secure against malicious adversaries.