Hand-based biometrics identifies users by unique features in their hands, such as fingerprints, palmprints, hand geometry, and finger and palm vein patterns. This book explores the range of technologies and methods under development and in use for handbased biometrics, with evaluations of the advantages and performance of each. The inclusion of significant material on the relevant aspects of the physiology of the hand is a particularly useful and innovative feature. Topics covered in this book include inner and outer hand physiology and diseases; nail structure and common disorders; fingerprint recognition; synthetic fingerprints; finger vein recognition; palm vein biometrics; hand shape recognition and palm print recognition; 3D hand shape recognition; and spoofing and anti-spoofing methods. With contributions from an international panel of experts in this field, Hand-Based Biometrics is essential reading for researchers, students and engineers working in biometrics and security.
Inspec keywords: biometrics (access control)
Other keywords: hand-based biometrics; biometric processing
Subjects: General and management topics; Data security
This chapter includes the overview of all described technologies in this whole book. At the beginning, we address medical point of view to the hand, i.e., inner and outer hand physiology, including nail structure. We continue with very well-known fingerprint recognition, continued by palmprint recognition, recognition of hand and finger veins and finished by 2D and 3D hand geometry recognition. Because of lack of interest and availability, we neglect nail structure recognition for biometric purposes and recognition of thermal images of hand and finger. However, we discuss a very important topic, which is liveness detection, i.e., spoofing and antispoofing methods for various hand-based biometric characteristics, especially fingerprints. ABC systems, watchlists for e-boarders and use of electronic travel documents (e-passports) play an important role for biometric systems based on recognition of hand features, especially for fingerprints because they are used in biometric epassports. This topic is discussed in the second section of this chapter.
For description of anatomical structures in the hand, it is important to know that the anatomical position of the hand is hanging alongside the trunk with fingers pointing downward and the palm pointing forward (Figure 2.1). Hand represents the utmost peripheral part of the upper extremity. In humans, it accomplishes fine movements incomparable with other species, like monkeys, which use their hands for handling items and performing simple movements. In contrast, human hand is able to perform the most intricate movements-for example see piano players, whose hands smoothly and quickly change among all sorts of the movements in numerous joints forming their hands: flexion, extension, abduction, adduction, circumduction, pronation and supination. It is a concert of movements parallel to the concert of sounds. Bones, bound together by joints and operated by muscles, represent the structural conditions for performing these movements. Proper function is ensured by appropriate blood flow in the hand muscles. Blood flow in the hand skin plays moreover a crucial role in control of body temperature.
This chapter includes description of outer hand, its physiology, and changes when skin disease is present, everything from medical point of view. The complete skin anatomy of the hand will be described, together with histopathological changes. The skin diseases localized on this special part of the body and their consequences to fingerprint recognition will be mentioned and discussed.
The nail is a complex structure with many important functions; several significant conditions may affect the nail and signs of serious underlying systemic disease may be seen in the nail.
This chapter deals with fingerprint recognition technology. Each of us has papillary lines (elevated skin reliefs, called often ridges and the gaps between ridges are called valleys) that are uniquely shaped for each person so that they can be distinguished on the basis of their passing through each other on the surface of their fingers (hands and feet). Today, it is the most widespread technology that can be met in almost everyday life.
To make Automatic Fingerprint Identification Systems (AFIS) capable of searching across several millions of fingerprints in a few seconds, very powerful (and expensive) distributed computing architectures are typically used. The recent improvement of algorithms and the availability of powerful CPUs and GPUs now make it possible to deploy large-scale fingerprint recognition on low-cost hardware, thus approaching a larger number of applications (e.g., welfare benefits in poor countries). This chapter discusses architectural design, algorithms, and hardware optimization to speed up fingerprint recognition on large databases.
This chapter includes a description of fingerprints which are common in a population, however, cause big troubles in automatic processing and recognition. In the past, we did a lot of acquisition tests of problematic fingerprints, including acquirement of a database with diseased fingerprints.
Synthetic fingerprint generation (SFinGe) techniques and associated tools (e.g., SFinGe) were introduced more than 15 years ago [1]. The main aim was to generate large databases for performance evaluation without allocating huge amount of resources for acquisition campaigns and, at the same time, to conform with the privacy directives that in many countries limit the exchange of biometric data. While the original scope remains central today, since the generation of very large synthetic dataset is crucial to predict accuracy on very large scenarios, new security needs (such as detecting altered fingerprints) and algorithms improvements (supervised learning approaches) are continuously renewing interest in the generation of synthetic fingerprints.
One of the big issues in biometric recognition is robustness of recognition accuracy against sample signal quality degradation. The performance of a biometric recognition system is usually heavily affected by sample signal quality. A wide variety of factors potentially influence the quality of acquired biometric samples. The different types of features that can be extracted from biometric samples influence the impact of quality degradations on recognition performance in various ways. Moreover, there is interplay among different types of feature extraction and acquisition technology/conditions such that it is not clear a priori which type of feature extraction is favourable under which conditions. Therefore, it is essential to provide reliable methodology to comparatively assess biometric recognition robustness under varying conditions.
Palmvein recognition (identification or verification) is gaining ground as a biometric system for personal recognition applications, with an intensive investigation during the past few years. Some researchers have devoted attention to this field, but a lot of issues related to palmvein recognition are still open. This paper provides an overview of current palmvein recognition research, describing in particular the current state of biometric palmvein system components. The beginner researchers can find in this paper the more recent approaches at each stage of biometric palmvein recognition. Also, this overview points out current open issues that need to be addressed and investigated.
In this chapter, we review the state of the arts and discuss the main challenges in finger vein recognition. First, the brief introduction of finger vein recognition is given, and then the finger vein image acquisition methods are presented, including imaging principle, imaging devices and open databases. Additionally, we give all kinds of methods in image preprocessing, feature extraction and matching, listing the recognition performance of the state of the arts. Some finger vein involved multifeature and multitrait recognition systems, which are also given. Lastly, the open research issues in finger vein recognition are analyzed.
The current state of palm-print and hand-geometry-based biometrics is presented. The most used methods of acquisition, feature extraction and subsequent identification are described as well as the current state of the commercially available solutions based on these technologies. The limitations of 2D biometrics are discussed along with possible approaches to solving them. 3D scanning techniques utilized in biometrics are introduced and described along with their complimentary nature to the 2D methods. The current development in this area is outlined. Novel multimodal method of biometric image acquisition utilizing the line scanner is introduced, with the experimental results from the proof of concept system demonstrating the capability of hand geometry, fingerprint and palm-print features extraction. The future development direction is outlined by introducing a 3D line scanner and justifying its specification with the required task in mid. The palm-print and hand geometry are two technologies that are often being overlooked in favor of alternatives, despite their advantages. With the advance of cheaper sensing technologies and higher available computing power, some of the shortcomings are being overcome. By expanding these systems into 3D, in a fast and affordable way, these biometrics may very well soon gain in popularity.
With the recent developments in low-cost 3D sensing, the interest in employing 3D biometrics has increased. We have wrapped up the current state of the art in 3D hand biometric recognition. Depending on the representation of the acquired geometry, we are offered different feature extraction methods. Typically, the hand scan is stored as range map. The neighborhood structure is well defined on such surfaces, and it is therefore easy to compute approximate geodesic distances as well as perform curvature analysis. Those are, in fact, the paths that most of the researchers have followed in the past. The features are typically stored as n-D vectors and are well suited to be compared using a specific distance metric. As shown in the article, in order to find the best separation of the subjects, dimensionality reduction and metric learning can be employed to exploit the information better. To encourage future development, we show some unconventional representations of the hand geometry as well. Independently, the field of hand tracking advanced very fast, currently offering fast and precise 3D hand-tracking models. Employing such models in 3D hand biometric recognition is very promising, as it could potentially boost the precision of the hand annotation that would naturally result in better stability and performance of the 3D hand recognition. As the hand-tracking models usually work in real time, new possibilities (e.g., continuous user verification, etc.) come up as well.
Vulnerability of any given security system is an important aspect that needs to be carefully analyzed during design phase. When dealing with biometric systems, one important source of such vulnerability that needs to be especially considered is its tendency to deception by spoofs. Since a high enough quality spoof can become indistinguishable from the original biometric characteristic by human eye, we may need additional methods of spoof recognition. We investigate several approaches used as an antispoofing method, with emphasis on approaches utilizing the liveness detection (antispoofing).
We discuss antispoofing and spoofing methods for hand-based biometrics. Generally, the methods which are based on skin properties could be used not only on fingers, but on the whole hand or even body. The most promising antispoofing technology is very probably the multispectral illumination and sensing. This technology acquires real skin reaction to the illumination with a concrete wavelength, which is hard to simulate for a combination of more wavelengths. Another very promising technology is to use the electrical properties of the human finger, i.e., combination of charging and turbulent flow (induction). However, some fakes have been found for this technology as well. Also, some fingerprint acquisition technologies include automatic antispoofing detection - these are for example, ultrasound scanning and optical tomography. These methods penetrate beneath the skin and acquire the underlying structures of the skin. Therefore if there is any fake glued on the fingertip surface, these technologies show artifacts on the image, which means that a finger(print) fake is in use.
When any technology starts to become pervasive, malicious parties will always search for methods of circumvention. One methos of circumvention in biometric recognition is a presentation attack (or spoofing), where an individual attempts to masquerade as someone else to either avoid being identified or to fraudulently gain access where they have no right to. A presentation attack is defined as “presentation to the biometric data capture subsystem with the goal of interfering with the operation of the biometric system”. This definition includes attempts at avoiding identification, where the individual attempts to portray a nonexistent identity. This may include presentation of synthetic biometric characteristics or the alteration of one's own biometric characteristics in such a way as to portray a new unique identity. To mitigate the risk to biometric recognition technology, automated methods, known as presentation attack detection (PAD), are used to detect such presentation attacks. This chapter provides an overview of presentation attacks, PAD, and associated international standards.