Iris Biometrics

In 1984, a photographer named Steve McCurry traveled to Pakistan in order to document the ordeal of Afghanistan’s refugees, orphaned during the Soviet Union’s bombing of Afghanistan. In the refugee camp Nasir Bagh, which was a sea of tents, he took a photograph of a young girl approximately at the age of 13. McCurry’s portrait turned out to capture emotion quite well and in June 1985 it ran on the cover of National Geographic. The girl’s sea green eyes have captivated the world since then and because no one knew her name she became known as the “Afghan girl” [368].

In past years, the ever-increasing demand on biometric systems entailed continuous proposals of new iris recognition techniques [44]. Still, the processing chain of traditional iris recognition (and other biometric) systems has remained almost unaltered.

Iris biometrics has received remarkable attention in the biometric community due to its unrivaled properties. Possessing epigenetic (not genetically determined) pattern information apart from color, the iris is widely employed as a biometric identifier because of its high universality (almost every person has the characteristic), distinctiveness (high discriminative power due to its entropy), permanence (stability except for pigmentation change over time), and performance (accuracy and speed). While according to the classification by Jain et al. in 2004 [229] there are other modalities with better collectability (the characteristic can well be measured), acceptability (people are willing to provide the characteristic), and circumvention (how easy the system can be fooled) properties, a lot of research effort has been invested to improve iris recognition with respect to these biometric properties.

Biometric systems without active participation of users by means of which people can be identified in surveillance scenarios represent an active research topic. Unconstrained iris recognition is a relatively new branch in iris biometrics.

Traditional iris processing following Daugman’s approach [116] extracts binary features after mapping the textural area between inner pupillary and outer limbic boundary into a doubly dimensionless representation.

Motivated by the growing demand for real-time capable solutions, this chapter presents in detail a multistage approach to iris segmentation [541, 540]

This chapter presents an experimental evaluation of eye detection and iris segmentation techniques discussed in Chaps.4 and6. Evaluation targets all three different quality factors of segmentation software listed in Chap.5, namely accuracy, speed, and usability.

The International Organization for Standardization (ISO) specifies iris biometric data to be recorded and stored in (raw) image form (ISO/IEC 19794-6), rather than in extracted templates (e.g., iris-codes). On the one hand, such deployments benefit from future improvements (e.g., in feature extraction stage) which can be easily incorporated (except sensor improvements), without reenrollment of registered users. On the other hand, since biometric templates may depend on patent-registered algorithms, databases of raw images enable more interoperability and vendor neutrality [121]. Furthermore, the application of low-powered mobile sensors for image acquisition, e.g., mobile phones, raises the need for reducing the amount of transmitted data. These facts motivate detailed investigations of the effect of image compression on iris biometrics in order to provide an efficient storage and rapid transmission of biometric records.

While better segmentation is certainly a highly effective approach to target iris processing for less constrained images, Daugman’s rubbersheet transform model presents a rather simplified anatomical model of pupillary dilation. Therefore, it is likely that two normalized iris images are not perfectly aligned. The majority of feature extraction approaches extracts binary output (iris-codes) from the obtained normalized textures [>44], and employs the fractional HD over different bit shifts between iris-codes in order to determine a degree of similarity at comparison stage. By shifting one of the two iris textures to be compared, or its corresponding iris-code, and calculating a comparison score employing the HD for each shift position, it is possible to account for the optimal alignment of both textures or iris-codes and to achieve so-called rotational invariance, i.e. likely present rotations of the head and consequently iris images are tolerated. However, the polar unwrapping with its simplifying assumptions may cause irrevocable mapping distortions even in case of generalizing from circular or elliptic shapes to AC-based approaches. In this case it is desirable to employ more sophisticated comparison techniques in order to exploit all the available information. Indeed, very few studies have proposed new or compared different binary similarity and distance measures, and it is a common agreement that HD-based comparison as proposed by Daugman [116] is the best method for this task. However, especially for unconstrained imagery more sophisticated comparison techniques are an efficient means to increase recognition accuracy without the necessity of re-enrollment. Another aspect to consider is speed-up of the recognition process. Traditional identification mode assessment involves a 1:n comparison, where n is the number of registered gallery subjects, and consequently does not scale well with respect to the number of enrolled users. Especially the ongoing Aadhaar project in India employing fingerprint and iris biometrics to uniquely identify each Indian citizen, and the necessary de-duplication checks during enrollment to avoid the issue of multiple Aadhaar numbers to a single individual make efficient identification an ultimate goal. Indeed, by employing partial matching and indexing techniques, huge amounts of processing time can be saved [270]. While traditional approaches typically sacrifice recognition accuracy in favor of more efficient comparison, there are methods to increase both accuracy and speed at the same time by effectively re-arranging iris-codes according to bit-reliability [440]. A similar technique may also be used to exploit single-instance multi-algorithm iris biometrics to increase accuracy without extending the total template size [439]. Therefore, the second part of this chapter targets the problem of iris recognition under unideal conditions from a different perspective, namely the application of sophisticated iris comparators and advanced serial iris identification techniques. Table 9.1 lists some of the techniques in a comparative manner accounting for computational cost, accuracy, and needed enrollment samples. Especially the latter is a typical restrictive condition in many application scenarios (e.g., forensics, high-throughput enrollment), where only a single enrollment sample is available.

Experiments with respect to advanced comparators discussed in Chap.9 are conducted in this chapter, employing a common evaluation protocol and setup. Investigated issues comprise the accuracy of each comparator, computational requirements compared to the baseline technique using HD with applied bit-shifting to account for rotational alignment, as well as related questions. More specifically, the distribution of reliable bits or trade-offs between recognition accuracy and speed is evaluated.

With respect to generic biometric systems several different attacks have been encountered to infiltrate these [228, 418]. Recent work systematically identifies security threats against biometric systems and possible countermeasures [449] and discusses man-in-the-middle attacks and BioPhishing against a web-based biometric authentication system [606].

Experimental investigations comprise performance evaluations of different types of iris biometric cryptosystems as well as cancelable iris biometrics. All experiments are performed within a unified experimental scenario, giving an overview of different approaches to iris biometric template protection at a glance.

Biometric cryptosystems and cancelable biometrics offer several advantages over generic biometric systems. The most important advantages are summarized in Table 15.1. These major advantages over conventional biometric systems call for several applications. In order to underline the potential of both technologies two essential use cases are discussed in detail. With respect to the design goals, biometric cryptosystems and cancelable biometrics offer significant advantages to enhance the privacy and security of biometric systems, providing reliable biometric authentication at a high security level. Techniques which provide provable security/privacy, while achieving practical recognition rates have remained elusive (even on small datasets).

In the first section of this chapter, the general relation between watermarking and biometrics is discussed in detail. The second section provides a systematic and critical view of the work done on applying watermarking to enhance biometric systems.