Biometric Presentation Attack Detection: Beyond the Visible Spectrum

The deployment of biometric recognition systems has seen a considerable increase over the last decade, in particular for fingerprint based systems. To tackle the security issues derived from presentation attacks launched on the biometric capture device, automatic presentation attack detection (PAD) methods have been proposed. In spite of their high detection rates on the LivDet databases, the vast majority of the methods rely on the samples provided by traditional capture devices, which may fail to detect more sophisticated presentation attack instrument (PAI) species. In this paper, we propose a multi-modal fingerprint PAD which relies on an analysis of: i) the surface of the finger within the short wave infrared (SWIR) spectrum, and ii) the inside of the finger thanks to the laser speckle contrast imaging (LSCI) technology. On the experimental evaluation over a database comprising more than 4700 samples and 35 PAI species, and including unknown attacks to model a realistic scenario, a Detection Equal Error Rate (D-EER) of 0.5% has been achieved. Moreover, for a BPCER ≤ 0.1% (i.e., highly convenient system), the APCER remains around 3%.

The wide deployment of biometric verification systems has also unveiled security and privacy issues. Among other vulnerabilities, presentation attacks directed to the capture device pose a severe security threat. That is, a gummy finger (known as presentation attack instrument, PAI) can be used to impersonate another subject and thereby gain illegal access to the system. To prevent such attacks, presentation attack detection (PAD) techniques have been proposed in the last decade. For the particular case of fingerprints, most approaches are based on conventional optical or capacitive sensors, acquiring a single image, and which can thus detect only a limited number of PAIs. In this work, we present a new multi-modal PAD approach based on a recently developed capture device, which is able to acquire four different types of samples: i) finger photos, ii) finger vein samples, iii) multi-spectral short wave infrared images, and iv) laser speckle contrast image sequence samples. Different ad-hoc algorithms have been developed for each set of images. The experimental evaluation over 35 PAIs, part of them used only at the test stage, shows the efficiency of the proposed approach in a realistic scenario, with an APCER of 6.6% for a BPCER of 0.2%.

The rapid advancement of premium smartphones with sophisticated high resolution cameras has enabled reliable and convenient contactless acquisition of fingerprint biometrics, making them suitable for verification at scale. However, as with other biometric systems, contactless fingerprint verification remains vulnerable to presentation attacks. Many Presentation Attack Detection (PAD) methods rely on supervised learning that requires labelled attack data, which limits generalisation. We propose an unsupervised PAD approach based on the Denoising Diffusion Probabilistic Model (DDPM). The system is trained only on bona fide samples and makes decisions via score level fusion of two deep similarity measures, yielding robustness to unseen attacks without access to attack labels. We further provide a Jacobian based theoretical analysis that explains why diffusion reconstructions are well suited to PAD. Using Frobenius and spectral norms together with a log determinant surrogate, we show that the iterative refinement in diffusion models preserves ridge aligned structure while contracting off manifold artifacts, thereby accounting for the observed empirical gains over single shot alternatives. In addition, we introduce a new fingerphoto presentation attack dataset comprising 101 subjects with 6,060 bona fide and 1,818 attack images captured on recent smartphones under diverse conditions. Benchmarking across four datasets shows that the proposed method is effective in practice, improving PAD performance and outperforming strong supervised baselines and recent state of the art methods in unseen attack scenarios. We report uncertainty estimates and non parametric statistical summaries across datasets and PAIs (Presentation Attack Instruments), which indicate consistent advantages without significant regressions, supporting the suitability of diffusion based PAD for deployment.

State-of-the-art spoof detection methods tend to overfit to the spoof types seen during training and fail to generalize to unknown spoof types. Given that face anti-spoofing is inherently a local task, we propose a face anti-spoofing framework, namely Self-Supervised Regional Fully Convolutional Network (SSR-FCN), that is trained to learn local discriminative cues from a face image in a self-supervised manner. The proposed framework improves generalizability while maintaining the computational efficiency of holistic face anti-spoofing approaches (< 4 ms on a Nvidia GTX 1080Ti GPU). The proposed method is interpretable since it localizes which parts of the face are labeled as spoofs. Experimental results show that SSR-FCN can achieve TDR = 65% @ 2.0% FDR when evaluated on a dataset comprising of 13 different spoof types under unknown attacks while achieving competitive performances under standard benchmark datasets (Oulu-NPU, CASIA-MFSD, and Replay-Attack).

The main scope of this chapter is to serve as an introduction to face presentation attack detection, including key resources and advances in the field in the last few years. The next pages present the different presentation attacks that a face recognition system can confront, in which an attacker presents to the sensor, mainly a camera, a Presentation Attack Instrument (PAI), that is generally a photograph, a video, or a mask, with the target to impersonate a genuine user or to hide the actual identity of the attacker via obfuscation. First, we make an introduction of the current status of face recognition, its level of deployment, and its challenges. In addition, we present the vulnerabilities and the possible attacks that a face recognition system may be exposed to, showing that way the high importance of presentation attack detection methods. We review different types of presentation attack methods, from simpler to more complex ones, and in which cases they could be effective. Then, we summarize the most popular presentation attack detection methods to deal with these attacks. Finally, we introduce public datasets used by the research community for exploring vulnerabilities of face biometrics to presentation attacks and developing effective countermeasures against known PAIs.

Face recognition systems face real challenges from various presentation attacks. New, more sophisticated methods of presentation attacks are becoming more difficult to detect using traditional face recognition systems. Thermal infrared imaging offers specific physical properties that may boost presentation attack detection capabilities. The aim of this paper is to present outcomes of investigations on the detection of various face presentation attacks in thermal infrared in various conditions including thermal heating of masks and various states of subjects. A thorough analysis of presentation attacks using printed and displayed facial photographs, 3D-printed, custom flexible 3D-latex and silicone masks is provided. The paper presents the intensity analysis of thermal energy distribution for specific facial landmarks during long-lasting experiments. Thermalization impact, as well as varying the subject’s state due to physical effort on presentation attack detection are investigated. A new thermal face spoofing dataset is introduced. Finally, a two-step deep learning-based method for the detection of presentation attacks is presented. Validation results of a set of deep learning methods across various presentation attack instruments are presented.

Biometric systems are vulnerable to presentation attacks (PAs) performed using various PA instruments (PAIs). Even though there are numerous PA detection (PAD) techniques based on both deep learning and hand-crafted features, the generalization of PAD for unknown PAI is still a challenging problem. In this work, we empirically prove that the initialization of the PAD model is a crucial factor for generalization, which is rarely discussed in the community. Based on such observation, we proposed a self-supervised learning-based method, denoted as DF-DM. Specifically, DF-DM is based on a global-local view coupled with de-folding and de-mixing to derive the task-specific representation for PAD. During de-folding, the proposed technique will learn region-specific features to represent samples in a local pattern by explicitly minimizing the generative loss. While de-mixing drives detectors to obtain the instance-specific features with global information for more comprehensive representation by minimizing the interpolation-based consistency. Extensive experimental results show that the proposed method can achieve significant improvements in terms of both face and fingerprint PAD in more complicated and hybrid datasets when compared with the state-of-the-art methods. When training in CASIA-FASD and Idiap Replay-Attack, the proposed method can achieve an 18.60% equal error rate (EER) in OULU-NPU and MSU-MFSD, exceeding the baseline performance by 9.54%. The source code of the proposed technique is available at https://github.com/kongzhecn/dfdm.

Today, face biometric systems are becoming widely accepted as a standard method for identity authentication in many security settings. For example, their deployment in automated border control gates plays a crucial role in accurate document authentication and reduced traveler flow rates in congested border zones. The proliferation of such systems is further spurred by the advent of portable devices. On the one hand, modern smartphone and tablet cameras have in-built user authentication applications while on the other hand, their displays are being consistently exploited for face spoofing. Similar to biometric systems of other physiological biometric identifiers, face biometric systems have their own unique set of potential vulnerabilities. In this work, these vulnerabilities (presentation attacks) are being explored via a biologically-inspired presentation attack detection model which is termed “BIOPAD.” Our model employs Gabor features in a feedforward hierarchical structure of layers that progressively process and train from visual information of people's faces, along with their presentation attacks, in the visible and near-infrared spectral regions. BIOPAD's performance is directly compared with other popular biologically-inspired layered models such as the “Hierarchical Model And X” (HMAX) that applies similar handcrafted features, and Convolutional Neural Networks (CNN) that discover low-level features through stochastic descent training. BIOPAD shows superior performance to both HMAX and CNN in all of the three presentation attack databases examined and these results were consistent in two different classifiers (Support Vector Machine and k-nearest neighbor). In certain cases, our findings have shown that BIOPAD can produce authentication rates with 99% accuracy. Finally, we further introduce a new presentation attack database with visible and near-infrared information for direct comparisons. Overall, BIOPAD's operation, which is to fuse information from different spectral bands at both feature and score levels for the purpose of face presentation attack detection, has never been attempted before with a biologically-inspired algorithm. Obtained detection rates are promising and confirm that near-infrared visual information significantly assists in overcoming presentation attacks.

A fingerphoto is obtained by imaging a human finger using a basic smartphone camera. Although impressive advances have been made to accurately match fingerphotos, this technology is vulnerable to presentation attacks (PAs). These algorithms do not generalize well in the presence of new presentation attacks. While previous research on this issue is limited, this paper systematically evaluates fingerphoto presentation attack detection (PAD) algorithms under unknown attacks. The proposed assessment compares different Convolutional Neural Networks (CNNs) on the IIITD Smartphone Fingerphoto database with spoof data including printout and various display attacks. These images used for the experiments were acquired indoors and subjected to background (i.e., white or natural) and capture device (i.e., Nokia or OPO) variations. Preliminary results show that the PAD based on AlexNet is robust under most types of replica unseen during the training of the detector.

Face Recognition Systems (FRS) are vulnerable to spoofing attacks (a.k.a presentation attacks), which can be carried out by presenting a printed photo (print-photo), displaying a photo (display-photo), or displaying a video (replay-video). The issue of presentation attacks can be alleviated by algorithms known as presentation attack detection (PAD) mechanisms. In this paper, we propose a novel framework based on Hierarchical Cosine/Spherical Linear Interpolation of deep learning feature vectors followed by training a Linear SVM for PAD Classification. The deep learning feature vectors are extracted from existing networks trained on the Imagenet dataset. Our proposed approach hierarchically interpolates the extracted feature vectors using Cosine/Spherically Linear Interpolation, followed by using a Linear SVM for classification, and sum-rule fusion for generating final scores. We show our results on cross-dataset PAD for the classifier trained on OULU P1 Dataset and tested on Replay Mobile Dataset. We compare it with the current state-of-the-art (SOTA) algorithms published in the literature and achieve considerably lower detection error-rate (D-EER). The extraction of features from pre-trained networks makes our approach simple to use, apart from it, giving highly accurate results, which are much better than current SOTA.

Light Field Focal Stack (LFFS) can be efficiently rendered from a light field (LF) image captured by plenoptic cameras. Differences in the 3D surface and texture of biometric samples are internally reflected in the defocus blur and local patterns between the rendered slices of LFFS. This unique property makes LFFS quite appropriate to differentiate presentation attack instruments (PAIs) from bona fide samples. A patch-based dual-view network (PDVN) is proposed in this paper to leverage the merits of LFFS for face presentation attack detection (PAD). First, original LFFS data are divided into various local patches along spatial dimensions, which distracts the model from learning the useless facial semantics and greatly relieve the problem of insufficient samples. The strategy of dual-view branches is innovatively proposed, wherein the original view and microscopic view can simultaneously contribute to liveness detection. Separable 3D convolution on the focal dimension is verified to be more effective than vanilla 3D convolution for extracting discriminative features from LFFS data. The voting mechanism on predictions of patch LFFS samples further strengthens the robustness of the proposed framework. PDVN is compared with other face PAD methods on IST LLFFSD dataset and achieves perfect performance, i.e., ACER drops to 0.

Although face recognition systems have wide application, they are vulnerable to presentation attack samples (fake samples). Therefore, a presentation attack detection (PAD) method is required to enhance the security level of face recognition systems. Most of the previously proposed PAD methods for face recognition systems have focused on using handcrafted image features, which are designed by expert knowledge of designers, such as Gabor filter, local binary pattern (LBP), local ternary pattern (LTP), and histogram of oriented gradients (HOG). As a result, the extracted features reflect limited aspects of the problem, yielding a detection accuracy that is low and varies with the characteristics of presentation attack face images. The deep learning method has been developed in the computer vision research community, which is proven to be suitable for automatically training a feature extractor that can be used to enhance the ability of handcrafted features. To overcome the limitations of previously proposed PAD methods, we propose a new PAD method that uses a combination of deep and handcrafted features extracted from the images by visible-light camera sensor. Our proposed method uses the convolutional neural network (CNN) method to extract deep image features and the multi-level local binary pattern (MLBP) method to extract skin detail features from face images to discriminate the real and presentation attack face images. By combining the two types of image features, we form a new type of image features, called hybrid features, which has stronger discrimination ability than single image features. Finally, we use the support vector machine (SVM) method to classify the image features into real or presentation attack class. Our experimental results indicate that our proposed method outperforms previous PAD methods by yielding the smallest error rates on the same image databases.

The Presentation attacks on biometric systems have become a major concern in the cyber world. The literature reveals that these systems are vulnerable to spoofing or presentation attacks (PAs) which sometimes result in complete failure of the recognition system. To combat against PAs, the anti-spoofing or Presentation attack detection (PAD) methods have been developed to check the liveness of the fingerprint modality. However, with the emergence of artificial intelligence, several software-based countermeasures have been proposed by the research community. Depending on the type feature extraction mechanism the PAD techniques are categorized into two main types namely; Machine learning (ML) and Deep Learning (DL) based approaches. This paper aims to present a review of existing work on the fingerprint PAD mechanism which has been carried out. The paper discusses the various fingerprint artifacts, state-of-the-art ML and DL based PAD algorithms, and publically available benchmark datasets. Furthermore, we discuss the major research challenges and future work that need to be addressed in this active field of fingerprint liveness detection.

Automated Fingerprint Recognition Systems (AFRSs) have been threatened by Presentation Attack (PA) since its existence. It is thus desirable to develop effective presentation attack detection (PAD) methods. However, the unpredictable PAs make PAD be a challenging problem. This paper proposes a novel One-Class PAD (OCPAD) method for Optical Coherence Technology (OCT) images based fingerprint PA detection. The proposed OCPAD model is learned from a training set only consists of Bonafides (i.e. real fingerprints). The reconstruction error and latent code obtained from the trained auto-encoder network in the proposed model is taken as the basis for the following spoofness score calculation. To get more accurate reconstruction error, we propose an activation map based weighting model to further refine the accuracy of reconstruction error. We test different statistics and distance measures and finally use a decision level fusion to make the final prediction. Our experiments are performed using a dataset with 93200 bonafide scans and 48400 PA scans. The results show that the proposed OCPAD can achieve a True Positive Rate (TPR) of 99.43% when the False Positive Rate (FPR) equals to 10% and a TPR of 96.59% when FPR=5%, which significantly outperformed a feature based approach and a supervised learning based model requiring PAs for training.

Iris recognition technology has attracted an increasing interest in the last decades in which we have witnessed a migration from research laboratories to real-world applications. The deployment of this technology raises questions about the main vulnerabilities and security threats related to these systems. Among these threats, presentation attacks stand out as some of the most relevant and studied. Presentation attacks can be defined as the presentation of human characteristics or artifacts directly to the capture device of a biometric system trying to interfere with its normal operation. In the case of the iris, these attacks include the use of real irises as well as artifacts with different levels of sophistication such as photographs or videos. This chapter introduces iris Presentation Attack Detection (PAD) methods that have been developed to reduce the risk posed by presentation attacks. First, we summarize the most popular types of attacks including the main challenges to address. Second, we present a taxonomy of PAD methods as a brief introduction to this very active research area. Finally, we discuss the integration of these methods into iris recognition systems according to the most important scenarios of practical application.