Accounting for uncertainty in three-dimensional (3D) shapes is important in a large number of scientific and engineering areas, such as biometrics, biomedical imaging, and data mining. It is well known that 3D polar shaped objects can be represented by Fourier descriptors such as spherical harmonics and double Fourier series. However, the statistics of these spectral shape models have not been widely explored. This thesis studies several areas involved in 3D shape modeling, including random field models for statistical shape modeling, optimal shape filtering, parametric active contours for object segmentation and surface reconstruction. It also investigates multi-modal image registration with respect to tumor activity quantification. Spherical harmonic expansions over the unit sphere not only provide a low dimensional polarimetric parameterization of stochastic shape, but also correspond to the Karhunen-Lo´eve (K-L) expansion of any isotropic random field on the unit sphere. Spherical ...

High dimensional time series data are observed in many complex systems. In networked data, some of the time series are influenced by other time series. Identifying these relations encoded in a graph structure or topology among the time series is of paramount interest in certain applications since the identifi ed structure can provide insights about the underlying system and can assist in inference tasks. In practice, the underlying topology is usually sparse, that is, not all the participating time series influence each other. The goal of this dissertation pertains to study the problem of sparse topology identi fication under various settings. Topology identi fication from time series is a challenging task. The first major challenge in topology identi fication is that the assumption of static topology does not hold always in practice since most of the practical systems are evolving ...

The subject of this PhD thesis is the efficient and robust processing and analysis of the audio recordings that are derived from a call center. The thesis is comprised of two parts. The first part is dedicated to dialogue/non-dialogue detection and to speaker segmentation. The systems that are developed are prerequisite for detecting (i) the audio segments that actually contain a dialogue between the system and the call center customer and (ii) the change points between the system and the customer. This way the volume of the audio recordings that need to be processed is significantly reduced, while the system is automated. To detect the presence of a dialogue several systems are developed. This is the first effort found in the international literature that the audio channel is exclusively exploited. Also, it is the first time that the speaker utterance ...

Preterm birth is a major risk factor for neurodevelopment impairments often only appearing later in life. The brain is still at a high rate of development during adolescence, making this a promising window for intervention. It is thus crucial to understand the mechanisms of altered brain function in this population. The aim of this thesis is to investigate how the brain dynamically reconfigures its own organisation over time in preterm-born young adolescents. Research to date has mainly focused on structural disturbances or in static features of brain function in this population. However, recent studies have shown that brain activity is highly dynamic, both spontaneously and during performance of a task, and that small disruptions in its complex architecture may interfere with normal behaviour and cognitive abilities. This thesis explores the dynamic nature of brain function in preterm-born adolescents in three ...

In direction finding (DF) applications, there are several factors affecting the estimation accuracy of the direction-of-arrivals (DOA) of unknown source locations. The major distortions in the estimation process are due to the array imperfections, model mismatches and multipath. The array imperfections usually exist in practical applications due to the nonidealities in the antenna array such as mutual coupling (MC) and gain/phase uncertainties. The model mismatches usually occur when the model of the received signal differs from the signal model used in the processing stage of the DF system. Another distortion is due to multipath signals. In the multipath scenario, the antenna array receives the transmitted signal from more than one path with different directions and the array covariance matrix is rank-deficient. In this thesis, three new methods are proposed for the problems in DF applications in the presence of array ...

As biometrics gains popularity and proliferates into the daily life, there is an increased concern over the loss of privacy and potential misuse of biometric data held in central repositories. The major concerns are about i) the use of biometrics to track people, ii) non-revocability of biometrics (eg. if a fingerprint is compromised it can not be canceled or reissued), and iii) disclosure of sensitive information such as race, gender and health problems which may be revealed by biometric traits. The straightforward suggestion of keeping the biometric data in a user owned token (eg. smart cards) does not completely solve the problem, since malicious users can claim that their token is broken to avoid biometric verification altogether. Put together, these concerns brought the need for privacy preserving biometric authentication methods in the recent years. In this dissertation, we survey existing ...

The abundance of large and heterogeneous systems is rendering contemporary data more pervasive, intricate, and with a non-regular structure. With classical techniques facing troubles to deal with the irregular (non-Euclidean) domain where the signals are defined, a popular approach at the heart of graph signal processing (GSP) is to: (i) represent the underlying support via a graph and (ii) exploit the topology of this graph to process the signals at hand. In addition to the irregular structure of the signals, another critical limitation is that the observed data is prone to the presence of perturbations, which, in the context of GSP, may affect not only the observed signals but also the topology of the supporting graph. Ignoring the presence of perturbations, along with the couplings between the errors in the signal and the errors in their support, can drastically hinder ...

Visual information, in the form of digital images and videos, has become so omnipresent in computer databases and repositories, that it can no longer be considered a “second class citizen”, eclipsed by textual information. In that scenario, image classification has become a critical task. In particular, the pursuit of automatic identification of complex semantical concepts represented in images, such as scenes or objects, has motivated researchers in areas as diverse as Information Retrieval, Computer Vision, Image Processing and Artificial Intelligence. Nevertheless, in contrast to text documents, whose words carry semantic, images consist of pixels that have no semanticinformation by themselves, making the task very challenging. In this dissertation, we have addressed the problem of representing images based on their visual information. Our aim is content-based concept detection in images and videos, with a novel representation that enriches the Bag-of-Words model. ...

H.264 Advanced Video Coding has become the dominant video coding standard in the market, within a few years after the first version of the standard was completed by the ISO/IEC MPEG and the ITU-T VCEG groups in May 2003. That happened mainly due to the great coding efficiency of H.264. Compared to MPEG-2, the previous dominant standard, the H.264 compression ratio is about twice as higher for the same video quality. That makes H.264 ideal for a numerous of applications, such as video broadcasting, video streaming and video conferencing. However, the H.264 efficiency is achieved at the expense of the codec¢s complexity. H.264 complexity is about four times that of MPEG-2. As a consequence, many video coding issues, which have been addressed in previous standards, need to be re-considered. For example the H.264 encoding of a video in real time ...

Device-to-Device (D2D) is one of the important proposed solutions to increase the capacity, offload the traffic, and improve the energy effciency in next generation cellular networks. D2D communication is known as a direct communication between two users without using cellular infrastructure networks. Despite of large expected bene fits in terms of capacity in D2D, the coexistence of D2D and cellular networks in the same spectrum creates new challenges in interference management and network design. To limit the interference power control schemes on cellular networks and D2D networks are typically adopted. Even though power control is introduced to limit the interference level, it does not prevent cellular and D2D users from experiencing coverage limitation when sharing the same radio resources. Therefore, the design of such networks requires the availability of suitable methods able to properly model the eff ect of interference ...

Using a hearing device is one of the most successful approaches to partially restore the degraded functionality of an impaired auditory system. However, due to the complex structure of the human auditory system, hearing impairment can manifest itself in different ways and, therefore, its compensation can be achieved through different classes of hearing devices. Although the majority of hearing devices consists of conventional hearing aids (HAs), several other classes of hearing devices have been developed. For instance, bone-conduction devices (BCDs) and cochlear implants (CIs) have successfully been used for more than thirty years. More recently, other classes of implantable devices have been developed such as middle ear implants (MEIs), implantable BCDs, and direct acoustic cochlear implants (DACIs). Most of these different classes of hearing devices rely on a sound processor running different algorithms able to compensate for the hearing impairment. ...

Recent years have witnessed a significant interest in interpretable machine learning (IML) research that develops techniques to analyse machine learning (ML) models. Understanding ML models is essential to gain trust in their predictions and to improve datasets, model architectures and training techniques. The majority of effort in IML research has been in analysing models that classify images or structured data and comparatively less work exists that analyses models for other domains. This research focuses on developing novel IML methods and on extending existing methods to understand machine listening models that analyse audio. In particular, this thesis reports the results of three studies that apply three different IML methods to analyse five singing voice detection (SVD) models that predict singing voice activity in musical audio excerpts. The first study introduces SoundLIME (SLIME), a method to generate temporal, spectral or time-frequency explanations ...

It takes more time to think of a silent scene, action or event than finding one that emanates sound. Not only speaking or playing music but almost everything that happens is accompanied with or results in one or more sounds mixed together. This makes acoustic event detection (AED) one of the most researched topics in audio signal processing nowadays and it will probably not see a decline anywhere in the near future. This is due to the thirst for understanding and digitally abstracting more and more events in life via the enormous amount of recorded audio through thousands of applications in our daily routine. But it is also a result of two intrinsic properties of audio: it doesn’t need a direct sight to be perceived and is less intrusive to record when compared to image or video. Many applications such ...

This Thesis is focused on the combination of multiple biometric traits for automatic person authentication, in what is called a multimodal biometric system. More generally, any type of biometric information can be combined in what is called a multibiometric system. The information sources in multibiometrics include not only multiple biometric traits but also multiple sensors, multiple biometric instances (e.g., different fingers in fingerprint verification), repeated instances, and multiple algorithms. Most of the approaches found in the literature for combining these various information sources are based on the combination of the matching scores provided by individual systems built on the different biometric evidences. The combination schemes following this architecture are typically based on combination rules or trained pattern classifiers, and most of them assume that the score level fusion function is fixed at verification time. This Thesis considers the problem of ...

A counting sequence of length n is a list of all 2^n binary n-tuples (binary codewords of length n). The number of bit positions where two codewords differ is called the Hamming distance of these two codewords. The average Hamming distance of a counting sequence of length n is defined as the average Hamming distance between the 2^n pairs of successive codewords, including the pair of the last and the first codeword. A counting sequence of length n which has average Hamming distance equal to n-1/2 is called a maximum counting sequence. The number of bit changes in bit position i, in a counting sequence of length n is called the transition count of bit position i. If a counting sequence of length n has the property that the difference between any two bit positions is at most 2, the ...