Age Factors in Biometric Processing
As biometrics-based identification and identity authentication become increasingly widespread in their deployment, it becomes correspondingly important to consider more carefully issues relating to reliability, usability and inclusion. One factor which is particularly important in this context is that of the relationship between the nature of the measurements extracted from a particular biometric modality and the age of the sample donor, and the effect which age has on physiological and behavioural characteristics invoked in a biometric transaction. In Age Factors in Biometric Processing an international panel of experts explore the implications of ageing on biometric technologies, and how such factors can be managed in practical situations. Topics include understanding the impact of ageing on biometric measurements; age factors as barriersopportunities in relation to performance; modality-related approaches to management of age factors; implications for practical application; and future trends and research challenges. Age Factors in Biometric Processing provides an outstanding overview of this topic for the rapidly expanding community of stakeholders in biometricsbased identification solutions in academia, industry and government.
Inspec keywords: biometrics (access control); age issues; face recognition; digital forensics; digital signatures; ethical aspects
Other keywords: ethical considerations; ageing; biometric processing; human inspection; fingerprints; age factors; age prediction; online signature; social considerations; face image; forensics; face recognition
Subjects: General electrical engineering topics; Optical, image and video signal processing; Computer vision and image processing techniques; Economic, social and political aspects of computing; Data security; General and management topics
- Book DOI: 10.1049/PBSP010E
- Chapter DOI: 10.1049/PBSP010E
- ISBN: 9781849195027
- e-ISBN: 9781849195034
- Page count: 376
- Format: PDF
-
Front Matter
- + Show details - Hide details
-
p.
(1)
-
Section 1: Introduction and basic issues
1 Ageing and biometrics: an introduction
- + Show details - Hide details
-
p.
3
–16
(14)
The chapters of this book which follow will discuss and analyse a wide variety of issues which describe the effects of human ageing on the way in which biometric systems are designed and deployed. This is a particularly important issue since, of course, biometrics is an area which is intrinsically concerned with the measurement of human characteristics which are subject to change as a consequence of ageing. It is appropriate, therefore, to begin by briefly examining why the ageing issue is potentially a matter of concern from a general societal perspective, before introducing the nature of the problem when biometric systems are considered. This initial chapter will identify the principal features of this general landscape, setting the scene for the more detailed studies to be reported subsequently.
2 Review of ageing with respect to biometrics and diverse modalities
- + Show details - Hide details
-
p.
17
–36
(20)
Concluding this chapter we can safely state that ageing affects biometric templates due to a number of reasons which might be different for each biometric trait. As a result of ageing effects the false rejection rate usually increases and the long-term performance of a biometric system decreases.
3 Biometrics and ageing: social and ethical considerations
- + Show details - Hide details
-
p.
37
–62
(26)
This chapter discusses the major societal and ethical issues raised by biometrics in an ageing society. There are various well-known problems associated with biometrics, like privacy concerns, which certainly affect older people, but which in fact pertain to all users, regardless of age. And there are various ethical and societal issues arising in relation to biometrics and older people, e.g. issues of inclusion and exclusion, which also arise with other non-biometric technologies.
4 Using age to enhance performance in biometrics
- + Show details - Hide details
-
p.
63
–89
(27)
This chapter will discuss both the impact of the use of age information on overall performance, complexity and usability in the design of biometric-based systems as well as the predictive capabilities of biometric systems in relation to age. Furthermore, because of the nature of the data in this specific category (population data characteristics) and as a consequence of a lower level analysis, it is possible to point to more specific strategies both for potentially improving error-rate performance achievable in a biometric system while also increasing the flexibility with which specific implementation configurations can be designed, and for considering a range of solutions of different levels of complexity.
-
Section 2: Modality-related approaches to management of age factors
5 Human face ageing: a perspective analysis from anthropometry and biometrics
- + Show details - Hide details
-
p.
93
–116
(24)
This chapter has provided an overview of the mechanisms of ageing and an update on the current state of the art for synthetic ageing and for the template ageing problem. Although there have been some clear advancements in all areas of face ageing, a solution is still elusive. Yet, while the early years of this research area were starved of data, there are now massive datasets with longitudinal images available to the research community, with new datasets available online each year. As a result, young and seasoned researchers alike will be motivated to forge ahead to develop comprehensive solutions to the set of challenges inherent in face ageing.
6 Ageing in biometrics: a case study in online signature
- + Show details - Hide details
-
p.
117
–132
(16)
Due to the fact that biometrics, as an automatic means of human recognition, constitutes a relatively novel field of research (Jain et al., 2006), to date most efforts undertaken by the different parties involved in the development of this technology (researchers, industry, evaluators, etc.) have been mainly focused on the improvement of its performance (i.e. finding novel methods to obtain lower error rates) (Wayman et al., 2005; Cappelli et al., 2006). As a consequence, other important aspects closely related to this type of system such as the performance degradation effect known as ageing have been left partially uncovered (Jain et al., 2008).
7 Ageing in iris biometrics
- + Show details - Hide details
-
p.
133
–152
(20)
This chapter is organized as follows: we start with a brief introduction to the eye and the iris. It is important to stop to see what we are measuring and try to understand from the anatomical organ what we can expect to happen with time. Following this, we will briefly describe the different components which form an iris biometric system as well as each component's functionality. This will help us to understand how age influences each part of the system. After all these basic concepts, we will move to the studies developed by different researchers on this topic, observing that different researchers have focused their efforts on different parts of the iris biometric system.
8 Ageing effects in fingerprint biometrics
- + Show details - Hide details
-
p.
153
–170
(18)
This chapter strives to investigate fingerprint ageing effects from a new full-hand image ageing dataset highlighting template-ageing effects and a separate child vs. adult dataset targeting age-group performance.
9 The impact of ageing on speech-based biometric systems
- + Show details - Hide details
-
p.
171
–184
(14)
Biometric authentication relies on the use of biometric templates. A person seeking to access a biometrically controlled system must first identify himself/herself and the system will then verify a match with his/her stored template, created during an enrolment procedure. Such systems are intended for long-term deployment. Hence invariance to human ageing is an important element of any biometric system. The majority of human biometrics such as face, fingerprint, handwriting, iris and voice are subject to ageing effects. Other chapters in this book discuss many of these modalities at length. Here we will focus on speech. Speech is the least intrusive biometric from a user perspective. It can be used independently, or combined with other modalities such as face in multimodal systems.
10 An analysis of biometric performance change over time: a multimodal perspective
- + Show details - Hide details
-
p.
185
–202
(18)
In this study, we attempt to gain an understanding of whether or not a change in performance of a biometric system is subject-dependent. We will not attempt to explain the cause of performance change; we know that it is not only age-related, but is also dependent on biometric sample quality and habituation. What will become clear is that measuring the change in biometric performance is challenging, let alone explaining the causes of this change. However, we would challenge the common assertion that the performance of a biometric system systematically degrades overtime.
-
Section 3: Applications and implications for practical applications
11 An industrial perspective on biometric age factors
- + Show details - Hide details
-
p.
205
–220
(16)
This chapter draws on the author's practical experience gained through delivering trial and production biometric systems to address a fundamental question - how is an understanding of biometric ageing relevant to the world of industry, which includes governmental use for immigration and identity management, and commerce? The following discussion and conclusions are therefore intended to help to set biometric ageing in an industrial context, and may be of interest to those engaged in the topic, regardless of their relative alignment to academia, industrial research and development, or commercial delivery. The paper approaches the topic of biometric ageing in the broadest sense, considering what impact it might have on today's industrial and commercial systems. Following this initial `sanity check', some of the key factors that might tend to exacerbate the impact of biometric ageing are considered and present an approach to identify those systems most exposed to this impact. Mitigation approaches that might serve to reduce the impact that biometric ageing presents, and how these might be put into practice, are considered.
12 Fingerprints and human inspection: a forensics perspective
- + Show details - Hide details
-
p.
221
–230
(10)
This chapter has purposely been written with the lay person in mind. However, we need to be aware that whatever techniques are employed to capture and match fingerprint templates, a human intervention may well be required, which will vary in its approach depending on the age of the donor, or other factors that may influence the success rate of the particular system being used. I believe that we need to examine the effects of human intervention more robustly in terms of the effects of ageing. Whatever system is used and whatever hardware and software we employ, there is always going to be a trade-off between obtaining the best capture and the best chance of an accurate match or non-match. I have pointed out that human friction ridges develop early in the foetus development and persist long after death when compared against other human features. We understand that humans age in much the same way, but that their genetic make up and the physical conditions they are exposed to vary make prediction of ageing effects difficult to forecast in any empirical manner. The varying requirements of human identification using fingerprint matching technologies display differing sensibilities to the effects of ageing, and the `risks' involved also differ according to the function of that system. There are on-going studies relating to the ageing of fingerprints and these must be embraced and performed in a manner that allows current and future vendors and users of systems (AFIS and civil biometric) to understand fully what effects ageing will have on system performance and how best to maximise that performance. But there are obstacles that need to be overcome before we can arrive at a position whereby our data sets are truly deemed to be adequate and fit for the purposes of study. Above all we need to be aware that this is a complex issue, but one related to a simple premise, i.e. whatever we do, we cannot avoid getting older!
13 Age prediction in face images
- + Show details - Hide details
-
p.
231
–251
(21)
The ageing factor is very significant in face images. As shown in Figure 13.1, the facial appearance changes a lot with the person's ageing process. The facial appearance changes could be used to characterize the ages of an individual for the purpose of age prediction. Actually, humans have the capability to perceive the ages of other people by looking at their faces. So, age perception from faces is quite natural for people.
14 Short- and long-time ageing effects in face recognition
- + Show details - Hide details
-
p.
253
–275
(23)
The research described in this chapter indicates the need to provide better models for both short-time and long-time ageing to either predict the age or to compensate for photometric changes in the images due to the time lapse. At the same time it is expected that research in template adaptation and flexible face representations will lead to the development of subject-specific models which take into account the peculiar variabilities of face areas due to transient or permanent changes over time.
-
Section 4: Future trends and research challenges
15 Analysis of ageing effects in biometric systems: difficulties and limitations
- + Show details - Hide details
-
p.
279
–301
(23)
This chapter will explore some of the limitations and difficulties which may be experienced specifically in the investigation and analysis of physical ageing issues in biometric systems. Considerable advances have been made in recent years by researchers seeking to investigate and explore the physical ageing process and its effects on most of the common biometric modalities such as iris, signature, fingerprint, voice and face. However, there are several significant problems which can limit a comprehensive age progression analysis in studying ageing effects, and which indeed can also lead to apparently inconsistent conclusions about the effect of age progression on performance. Thus, in this chapter, some of the most important limitations which challenge research in this area will be investigated and explored, while some suggestions will also be presented about how to manage these issues in the analysis of physical ageing in biometric systems.
16 Usable biometrics for an ageing population
- + Show details - Hide details
-
p.
303
–320
(18)
In this chapter, we examine how ageing affects human interaction with biometric systems. We focus on the ability and willingness of older users to engage with advanced technologies, and use them successfully. Ageing can change physical and behavioural characteristics, and this can impact performance. In this chapter, we focus on the change in general physical and cognitive capabilities, and how this impacts the ability to interact with the systems in which biometric recognition is embedded. If older users find it hard to present their biometrics correctly, or find the behaviour required to present unacceptable, this can lead to exclusion.
17 Ageing effects and implications for biometric template protection
- + Show details - Hide details
-
p.
321
–341
(21)
In practice, biometric template protection schemes require favourable conditions, i.e. low intra-class variability [36]. Biometric template ageing, which has been most studied in the context of face recognition, causes increased intra-class variability [24], i.e. requirements on template protection do not coincide with ageing effects. In this work implications of ageing effects on template protection are discussed for various types of technologies and possible countermeasures are proposed. In a case study on iris biometric fuzzy commitment schemes the effects of ageing are investigated over a time period of two years. Experimental studies confirm recent investigations on iris template ageing effects [11], observing a significant decrease in the performance of fuzzy commitment schemes.
18 Future challenges for systems delivery using ageing tolerant biometrics
- + Show details - Hide details
-
p.
343
–350
(8)
A biometrically enabled transaction token was devised at the UK National Physical Laboratory as long ago as the late 1980s. It was then envisaged that the use of this electronic token in, e.g. payment transactions, would be enabled by automatic personal identification methods. The methods suggested included the measurement of biometric characteristics such as `written signatures, fingerprint and vein scanning'. It was suggested that the PIN be retained as a supplementary check on identity. Particular emphasis was given to the prospects for vein scanning as the basis of a new biometric.
-
Back Matter
- + Show details - Hide details
-
p.
(1)