da/sec - Biometrics and Internet Security Research Group
Refine
Document Type
- Article (8)
- Doctoral Thesis (2)
- Preprint (1)
Language
- English (11)
Has Fulltext
- yes (11)
Is part of the Bibliography
- yes (11)
Keywords
- Biometrics (access control) (3)
- Feature extraction (2)
- Fingerprint recognition (2)
- - (1)
- Biometric Identification (1)
- Biometric Systems (1)
- Computational Workload Reduction (1)
- Databases (1)
- Face Recognition (1)
- Face recognition (1)
Mobile Contactless Fingerprint Presentation Attack Detection: Generalizability and Explainability
(2024)
Contactless fingerprint recognition is an emerging biometric technology that has several advantages over contact-based schemes, such as improved user acceptance and fewer hygienic concerns. Like for most other biometrics, Presentation Attack Detection (PAD) is crucial to preserving the trustworthiness of contactless fingerprint recognition methods. For many contactless biometric characteristics, Convolutional Neural Networks (CNNs) represent the state-of-the-art of PAD algorithms. For CNNs, the ability to accurately classify samples that are not included in the training is of particular interest, since these generalization capabilities indicate robustness in real-world scenarios. In this work, we focus on the generalizability and explainability aspects of CNN-based contactless fingerprint PAD methods. Based on previously obtained findings, we selected four CNN-based methods for contactless fingerprint PAD: two PAD methods designed for other biometric characteristics, an algorithm for contact-based fingerprint PAD and a general-purpose ResNet18. For our evaluation, we use four databases and partition them using Leave-One-Out (LOO) protocols. Furthermore, the generalization capability to a newly captured database is tested. Moreover, we explore t-SNE plots as a means of explainability to interpret our results in more detail. The low D-EERs obtained from the LOO experiments (below 0.1% D-EER for every LOO group) indicate that the selected algorithms are well-suited for the particular application. However, with an D-EER of 4.14%, the generalization experiment still has room for improvement.
We propose MCLFIQ: Mobile Contactless Fingerprint Image Quality, the first quality assessment algorithm for mobile contactless fingerprint samples. To this end, we re-trained the NIST Fingerprint Image Quality (NFIQ) 2 method, which was originally designed for contact-based fingerprints, with a synthetic contactless fingerprint database. We evaluate the predictive performance of the resulting MCLFIQ model in terms of Error-vs.-Discard Characteristic (EDC) curves on three real-world contactless fingerprint databases using three recognition algorithms. In experiments, the MCLFIQ method is compared against the original NFIQ 2 method, a sharpness-based quality assessment algorithm developed for contactless fingerprint images and the general purpose image quality assessment method BRISQUE. Furthermore, benchmarks on four contact-based fingerprint datasets are also conducted. Obtained results show that the fine-tuning of NFIQ 2 on synthetic contactless fingerprints is a viable alternative to training on real databases. Moreover, the evaluation shows that our MCLFIQ method works more accurately and is more robust compared to all baseline methods on contactless fingerprints. We suggest considering the proposed MCLFIQ method as a starting point for the development of a new standard algorithm for contactless fingerprint quality assessment.
Gender classification on normalized iris images has been previously attempted with varying degrees of success. In these previous studies, it has been shown that occlusion masks may introduce gender information; occlusion masks are used in iris recognition to remove non-iris elements. When, the goal is to classify the gender using exclusively the iris texture, the presence of gender information in the masks may result in apparently higher accuracy, thereby not reflecting the actual gender information present in the iris. However, no measures have been taken to eliminate this information while preserving as much iris information as possible.
We propose a novel method to assess the gender information present in the iris more accurately by eliminating gender information in the masks. This consists of pairing iris with similar masks and different gender, generating a paired mask using the OR operator, and applying this mask to the iris. Additionally, we manually fix iris segmentation errors to study their impact on the gender classification.
Our results show that occlusion masks can account for 6.92% of the gender classification accuracy on average. Therefore, works aiming to perform gender classification using the iris texture from normalized iris images should eliminate this correlation.
Face Morphing Attacks pose a threat to the security of identity documents, especially with respect to a subsequent access control process, because they allow both involved individuals to use the same document. Several algorithms are currently being developed to detect Morphing Attacks, often requiring large data sets of morphed face images for training. In the present study, face embeddings are used for two different purposes: first, to pre-select images for the subsequent large-scale generation of Morphing Attacks, and second, to detect potential Morphing Attacks. Previous studies have demonstrated the power of embeddings in both use cases. However, we aim to build on these studies by adding the more powerful MagFace model to both use cases, and by performing comprehensive analyses of the role of embeddings in pre-selection and attack detection in terms of the vulnerability of face recognition systems and attack detection algorithms. In particular, we use recent developments to assess the attack potential, but also investigate the influence of morphing algorithms. For the first objective, an algorithm is developed that pairs individuals based on the similarity of their face embeddings. Different state-of-the-art face recognition systems are used to extract embeddings in order to pre-select the face images and different morphing algorithms are used to fuse the face images. The attack potential of the differently generated morphed face images will be quantified to compare the usability of the embeddings for automatically generating a large number of successful Morphing Attacks. For the second objective, we compare the performance of the embeddings of two state-of-the-art face recognition systems with respect to their ability to detect morphed face images. Our results demonstrate that ArcFace and MagFace provide valuable face embeddings for image pre-selection. Various open-source and commercial-off-the-shelf face recognition systems are vulnerable to the generated Morphing Attacks, and their vulnerability increases when image pre-selection is based on embeddings compared to random pairing. In particular, landmark-based closed-source morphing algorithms generate attacks that pose a high risk to any tested face recognition system. Remarkably, more accurate face recognition systems show a higher vulnerability to Morphing Attacks. Among the systems tested, commercial-off-the-shelf systems were the most vulnerable to Morphing Attacks. In addition, MagFace embeddings stand out as a robust alternative for detecting morphed face images compared to the previously used ArcFace embeddings. The results endorse the benefits of face embeddings for more effective image pre-selection for face morphing and for more accurate detection of morphed face images, as demonstrated by extensive analysis of various designed attacks. The MagFace model is a powerful alternative to the often-used ArcFace model in detecting attacks and can increase performance depending on the use case. It also highlights the usability of embeddings to generate large-scale morphed face databases for various purposes, such as training Morphing Attack Detection algorithms as a countermeasure against attacks.
In many forensic scenarios, criminals often attempt to conceal their identity by covering their face and other distinctive body parts. In such situations, physical evidence may, however, reveal other unique characteristics, e.g. hands, which can be used to identify offenders. In this context, several state-of-the-art biometric recognition systems have been proposed recently. These recognition systems offer high identification performance in restricted environments. However, in forensic scenarios, the environment is often unconstrained, making biometric identification considerably more difficult, with a consequent decrease in accuracy. In this article, we explore methods (e.g. hand alignment and information fusion) to improve the identification of subjects within forensic investigations. Experimental results show that explored techniques play an important role in the improvement of the identification performance of existing schemes: the combination of hand alignment and information fusion results in the highest Rank-1 identification performance improvement of up to 13.10% (i.e., 26.30% vs. 13.20%) and 16.30% (i.e., 77.00% vs. 60.70%) with respect to the baseline for the unconstrained databases NTU-PI_v1 and HaGRID, respectively ( https://github.com/ljsoler/IF-HA-HandRecognition ).
The development of large-scale identification systems that ensure the privacy protection of enrolled subjects represents a major challenge. Biometric deployments that provide interoperability and usability by including efficient multi-biometric solutions are a recent requirement. In the context of privacy protection, several template protection schemes have been proposed in the past. However, these schemes seem inadequate for indexing (workload reduction) in biometric identification systems. More specifically, they have been used in identification systems that perform exhaustive searches, leading to a degradation of computational efficiency. To overcome these limitations, we present an efficient privacy-preserving multi-biometric identification system that retrieves protected deep cancelable templates and is agnostic with respect to biometric characteristics and biometric template protection schemes. To this end, a multi-biometric binning scheme is designed to exploit the low intra-class variation properties contained in the frequent binary patterns extracted from different types of biometric characteristics. Experimental results reported on publicly available databases using state-of-the-art Deep Neural Network (DNN)-based embedding extractors show that the protected multi-biometric identification system can reduce the computational workload to approximately 57% (indexing up to three types of biometric characteristics) and 53% (indexing up to two types of biometric characteristics), while simultaneously improving the biometric performance of the baseline biometric system at the high-security thresholds. Code is available at https://github.com/dosorior/FBP-Multi-biometric-Indexing.
Biometric fingerprint identification hinges on the reliability of its sensors; however, calibrating and standardizing these sensors poses significant challenges, particularly in regards to repeatability and data diversity. To tackle these issues, we propose methodologies for fabricating synthetic 3D fingerprint targets, or phantoms, that closely emulate real human fingerprints. These phantoms enable the precise evaluation and validation of fingerprint sensors under controlled and repeatable conditions. Our research employs laser engraving, 3D printing, and CNC machining techniques, utilizing different materials. We assess the phantoms’ fidelity to synthetic fingerprint patterns, intra-class variability, and interoperability across different manufacturing methods. The findings demonstrate that a combination of laser engraving or CNC machining with silicone casting produces finger-like phantoms with high accuracy and consistency for rolled fingerprint recordings. For slap recordings, direct laser engraving of flat silicone targets excels, and in the contactless fingerprint sensor setting, 3D printing and silicone filling provide the most favorable attributes. Our work enables a comprehensive, method-independent comparison of various fabrication methodologies, offering a unique perspective on the strengths and weaknesses of each approach. This facilitates a broader understanding of fingerprint recognition system validation and performance assessment.
Biometric systems have experienced a large development in recent
years since they are accurate, secure, and in many cases, more user
convenient than traditional credential-based access control systems. Inspite of their benefits, biometric systems are still vulnerable to attack presentations (APs), which can be easily launched by a fraudulentsubject without having a wide expert knowledge. This way, he/she can gain access to several applications, such as bank accounts and smartphone unlocking, where biometric systems are frequently deployed. In order to mitigate such threats and increase the security of biometric systems, the development of reliable Presentation Attack
Detection (PAD) algorithms is of utmost importance to the research
community.In the context of PAD, we explore in this Thesis different strategies and methods in order to improve the generalisation capability of PAD schemes. To that end, we propose the definition of a semantic common feature space which successfully discriminates bona fide presentations (BPs)1 from APs. In essence, this process is seeking for those significant features extracted from known PAI species samples that are observed in unknown PAI species. In addition, we explore several handcrafted techniques in order to build a reliable description of features per biometric characteristic studied. The experimental evaluation shows that a common feature space can be computed through the fusion between generative models and discriminative approaches. Remarkable detection performances for high-security thresholds lead to the construction of a convenient (i.e., low BP rejection rates or Bona fide Presentation Classification Error Rate (BPCER)) and secure (i.e., low AP acceptance rates or Attack Presentation Classification Error Rate (APCER)) PAD subsystem.
Mobile Contactless Fingerprint Recognition: Implementation, Performance and Usability Aspects
(2022)
This work presents an automated contactless fingerprint recognition system for smartphones.We provide a comprehensive description of the entire recognition pipeline and discuss important requirements for a fully automated capturing system. In addition, our implementation
is made publicly available for research purposes. During a database acquisition, a total number of 1360 contactless and contact-based samples of 29 subjects are captured in two different environmental
situations. Experiments on the acquired database show a comparable performance of our contactless scheme and the contact-based baseline scheme under constrained environmental influences. A comparative usability study on both capturing device types indicates that the majority of subjects prefer the contactless capturing method. Based on our experimental results, we analyze the impact of the current COVID-19 pandemic on fingerprint recognition systems. Finally, implementation aspects of contactless fingerprint recognition are summarized.
The growing scope, scale, and number of biometric deployments around the world emphasise the need for research into technologies
facilitating efficient and reliable biometric identification queries. This work presents a method of indexing biometric databases,
which relies on signal-level fusion of facial images (morphing) to create a multi-stage data-structure and retrieval protocol. By
successively pre-filtering the list of potential candidate identities, the proposed method makes it possible to reduce the necessary
number of biometric template comparisons to complete a biometric identification transaction. The proposed method is extensively
evaluated on publicly available databases using open-source and commercial off-the-shelf recognition systems. The results show
that using the proposed method, the computational workload can be reduced down to around 30%, while the biometric performance
of a baseline exhaustive search-based retrieval is fully maintained, both in closed-set and open-set identification scenarios.