Techniques for establishing a person7#8217;s identity are characterized by several shortcomings. Identification documents may be forged or altered, signatures are difficult to authenticate and verify, tokens or seals may be stolen or counterfeited, and physical descriptions are difficult to quantitatively assess. Establishing identity continues to be important for the same purpose it has been for centuries – for banking transactions, establishing legal presence, entering contracts, gaining entry to secured premises, identifying fugitives, etc. Recently biometrics – the science of recognizing an individual based on his physiological or behavioral traits – has gained increasing acceptance as a legitimate method for these tasks. Currently, most deployed biometric systems are unimodal – they rely on a single feature to identify a person. Although these features, such as face, iris, ear, fingerprint, signature, or voice, may be sufficiently unique, systems must still contend with a variety of problems such as noisy data, intra-class variations, restricted degrees of freedom, non-universality, spoof attacks, and unacceptable error rates. Some of these issues may be eliminated and system accuracy increased through a multimodal biometric system.
In this dissertation we formulate and investigate a method for processing multimodal biometric data – collected from a single source – to extract multiple biometric features from a sample and subsequently classify the identity of the sample using multiple biometric methods in such a way that some or all of the identity features may be opportunistically selected. In this context, opportunistic selection is meant to refer to techniques used to identify sporadic but consistently unavailable features which may be used to improve the classification rate of a more reliably present feature. Specifically, we work with three biometrics extracted from a single high-resolution near infrared face image: iris, face, and skin irregularity features. To implement this system, we present an original pupil boundary estimation method for localizing irises, as well as an iris feature encoding technique using the discrete cosine transform. We then describe a face detection scheme which provides motivation for a novel skin irregularity feature recognition component. In this module, skin features are detected using the Speeded-Up Robust Features method and encoded using fully connected skin irregularity feature graphs. Additionally, we describe a face recognition step using two-dimensional principal components analysis. With these components, we fuse the recognition decision results using a voting method. The accuracy of the system, both as separate components and together, is evaluated on the CASIA-Iris-Distance database, a long-range and high-quality face/iris data set. Finally, following the results achieved using structural information features in face component detection and skin irregularity details, we discuss future work for flexible, targeted feature extraction.