Novel Applications of Multimodal Biometrics

This chapter presents the original idea of using social networks and context information in multimodal biometric for increased system security. A recently investigated study’s outcomes is presented, which showcase this idea as a new step in multi-biometric research. Since this method does not degrade the performance of the system and is not computationally expensive, it can be used in any biometric framework. However, as the amount of improvement depends on how distinctive and predictable people are in terms of their behavioral patterns, the method is most suitable for the predictable environments with some predefined behavioral routines. Fine tuning the system for each environment to find the most suitable parameters based on the behavioral patterns of that specific environment can result in better performance. This research is validated on example of gait recognition.

Robotics and Multimodal Biometrics

This chapter presents a review on a new subfield of security research which transforms and expands the domain of biometrics beyond biological entities to include virtual reality entities, such as avatars, which are rapidly becoming a part of society. Artimetrics research at Cybersecurity Lab, University of Louisville, USA, and Biometric Technologies Lab, University of Calgary, Canada, builds on and expands such diverse fields of science as forensics, robotics, virtual worlds, computer graphics, biometrics, and security. Analyzing the visual properties and behavioral profiling can ensure verification and recognition of avatars. This chapter introduces a multimodal system for artificial entities recognition, simultaneously profiling multiple independent physical and behavioral characteristic of an entity, and creating a new generation multimodal system capable of authenticating both biological (human being) and non-biological (avatars) entities. At the end, this chapter focuses on some future research directions by discussing robotic biometrics beyond images and text-based communication to intelligent software agents that can emulate human intelligence. As artificial intelligence and virtual reality domains evolve, they will in turn give rise to new generation security solutions to identity management spanning both human and artificial entity worlds.

Rank Level Fusion

Rank level fusion is one of the after matching fusion methods used in multibiometric systems. The problem of rank information aggregation has been raised before in various fields. This chapter extensively discusses the rank level fusion methodology, starting with existing literature from the last decade in different application scenarios. Several approaches of existing biometric rank level fusion methods, such as plurality voting method, highest rank method, Borda count method, logistic regression method, and quality-based rank fusion method, are discussed along with their advantages and disadvantages in the context of the current state-of-the-art in the discipline.

Introduction

This chapter presents an introductory overview of the application of computational intelligence in biometrics. Starting with the historical background on artificial intelligence, the chapter proceeds to the evolutionary computing and neural networks. Evolutionary computing is an ability of a computer system to learn and evolve over time in a manner similar to humans. The chapter discusses swarm intelligence, which is an example of evolutionary computing, as well as chaotic neural network, which is another aspect of intelligent computing. At the end, special concentration is given to a particular application of computational intelligence—biometric security.

Chaotic Neural Networks and Multi-Modal Biometrics

Neural network is a collection of interconnected neurons with the ability to derive conclusion from imprecise data that can be used to both identify and learn patterns. This chapter presents the concept of neural network as an intelligent learning tool for biometric security systems. Neural networks have been extensively used in a variety of computational and optimization problems. In the first half of this chapter, focus is given to a specific topic—chaos in neural network. A detailed description of an on-demand chaotic noise injection method recently developed to deal with a common drawback of non-autonomous methods—their blind noise injecting strategy—is presented. The second part of the chapter discusses the issue of high-dimensionality in the context of a complex biometric security system. The amount of data and its complexity can be overwhelming, and one way of dealing with this issue is to use the dimensionality reduction techniques, which are typically based on clustering or transformations from one space to another. The reduced dimensionality vector can be then used in the energy model for an associative memory, which will learn the data patterns. The benefit is that this is a learner system that converges the given set of vectors to the stored pattern in a network, which can be later used for biometric recognition and also for identifying the most significant biometric patterns. At the end of this chapter, some examples are presented showing the feasibility of using such approach in biometric domain—both for single and multi-modal biometric.

Overview of Biometrics and Biometrics Systems

Recent security threats increase the necessity to establish the identity of every person. Biometric authentication is a solution to person authentication by analyzing physiological or behavioral characteristics. In this chapter, various biometric notions and terms are reviewed, along with typical biometric system components and different functionalities and performance parameters. The design and development of a biometric system, depending on a particular application scenario, is covered. This chapter also focuses on the inherent issues associated with biometric data and system performance through introducing radically new methods based on intelligent information fusion and intelligent pattern recognition, thus creating a notion of intelligent security systems. At the end of the chapter, the potential drawbacks of biometric unimodal systems, which serves as the motivation to introduce the concept of multimodal biometric system in the context of intelligent security systems, is discussed.

Fuzzy Fusion for Multimodal Biometric

Fuzzy logic is a mathematical tool that can provide a simple way to derive a conclusion with the presence of noisy input information. It is a powerful intelligent tool and used heavily in many cognitive and decision-making systems. In this chapter, fuzzy logic-based fusion approach for multimodal biometric system is discussed. After discussing the basics of fuzzy logic, the fuzzy fusion mechanism in the context of a multimodal biometric system is illustrated. A brief discussion on the research conducted for fuzzy logic-based fusion in different application domains is also presented. The biggest advantage of the system is that instead of binary “Yes”/“No” decision, the probability of a match and confidence level can be obtained. A fuzzy fusion-based biometric system can be easily adjusted by controlling weight assignment and fuzzy rules to fit changing conditions. Some results of experimentations conducted in a recent research investigation on two virtual multimodal databases are presented. The discussion on the effect of incorporating soft biometric information with the fuzzy fusion method to make the system more accurate and robust is also included.

Multimodal Biometric System and Information Fusion

Integrating different information originating from different sources, known as information fusion, is one of the main factors of designing a biometric system involving more than one biometric source. In this chapter, various information fusion techniques in the context of multimodal biometric systems are discussed. Usually, the information in a multimodal biometric system can be combined in senor level, feature extraction level, match score level, rank level, and decision level. There is also another emerging fusion method, which is becoming popular—the fuzzy fusion. Fuzzy fusion deals with the quality of the inputs or with the quality of any system components. This chapter discusses the associated challenges related to making the choice of appropriate fusion method for the application domain, to balance between fully automated versus user defined operational parameters of the system and to take the decision on governing rules and weight assignment for fuzzy fusion.

Biometric Image Processing

In most biometric-based security systems, images of the associated biometric identifiers are used as the input to that system. This chapter discusses various image processing methods and algorithms commonly used for biometric pattern recognition. Efficient and reliable processing of images is essential to achieve good performance of biometric systems. Different appearance-based methods, such as eigenimage and fisherimage, and topological feature-based methods, such as Voronoi diagram-based recognition, are discussed in the context of face, ear, and fingerprint application frameworks. Utilizing cognitive intelligence and adaptive learning methods in both physical and behavioral biometrics are some emerging new directions of biometric pattern recognition. As such, neural networks, fuzzy logic, and cognitive architectures would play a more important role in biometric domain of research. The chapter concludes with discussion of the importance of context-based recognition for behavioral biometrics.

Markov Chain for Multimodal Biometric Rank Fusion

Markov chain is a mathematical model used to represent a stochastic process. In this chapter, Markov chain-based rank level fusion method for multimodal biometric authentication system is discussed. Due to some inherent problems associated with existing biometric rank fusion methods, Markov chain-based biometric rank fusion has recently emerged in biometric context. The notion of Markov chain and its construction mechanisms are presented along with discussion on some early research conducted on Markov chain in other rank aggregation frameworks. This chapter also presents a detailed description of recent experimentations conducted to evaluate the performance of Markov chain-based biometric rank fusion method in a face, ear, and iris-based application framework.