Various real-life data such as time series and multi-sensor recordings can be represented by vectors and matrices, which are one-way and two-way arrays of numerical values, respectively. Valuable information can be extracted from these measured data matrices by means of matrix factorizations in a broad range of applications within signal processing, data mining, and machine learning. While matrix-based methods are powerful and well-known tools for various applications, they are limited to single-mode variations, making them ill-suited to tackle multi-way data without loss of information. Higher-order tensors are a natural extension of vectors (first order) and matrices (second order), enabling us to represent multi-way arrays of numerical values, which have become ubiquitous in signal processing and data mining applications. By leveraging the powerful utitilies offered by tensor decompositions such as compression and uniqueness properties, we can extract more information from multi-way ...

Gliomas represent about 80% of all malignant primary brain tumors. Despite recent advancements in glioma research, patient outcome remains poor. The 5 year survival rate of the most common and most malignant subtype, i.e. glioblastoma, is about 5%. Magnetic resonance imaging (MRI) has become the imaging modality of choice in the management of brain tumor patients. Conventional MRI (cMRI) provides excellent soft tissue contrast without exposing the patient to potentially harmful ionizing radiation. Over the past decade, advanced MRI modalities, such as perfusion-weighted imaging (PWI), diffusion-weighted imaging (DWI) and magnetic resonance spectroscopic imaging (MRSI) have gained interest in the clinical field, and their added value regarding brain tumor diagnosis, treatment planning and follow-up has been recognized. Tumor segmentation involves the imaging-based delineation of a tumor and its subcompartments. In gliomas, segmentation plays an important role in treatment planning as well ...

Reducing interference noise in a noisy speech recording has been a challenging task for many years yet has a variety of applications, for example, in handsfree mobile communications, in speech recognition, and in hearing aids. Traditional single-channel noise reduction schemes, such as Wiener filtering, do not work satisfactorily in the presence of non-stationary background noise. Alternatively, supervised approaches, where the noise type is known in advance, lead to higher-quality enhanced speech signals. This dissertation proposes supervised and unsupervised single-channel noise reduction algorithms. We consider two classes of methods for this purpose: approaches based on nonnegative matrix factorization (NMF) and methods based on hidden Markov models (HMM). The contributions of this dissertation can be divided into three main (overlapping) parts. First, we propose NMF-based enhancement approaches that use temporal dependencies of the speech signals. In a standard NMF, the important temporal ...

Sound source separation refers to the task of estimating the signals produced by individual sound sources from a complex acoustic mixture. It has several applications, since monophonic signals can be processed more efficiently and flexibly than polyphonic mixtures. This thesis deals with the separation of monaural, or, one-channel music recordings. We concentrate on separation methods, where the sources to be separated are not known beforehand. Instead, the separation is enabled by utilizing the common properties of real-world sound sources, which are their continuity, sparseness, and repetition in time and frequency, and their harmonic spectral structures. One of the separation approaches taken here use unsupervised learning and the other uses model-based inference based on sinusoidal modeling. Most of the existing unsupervised separation algorithms are based on a linear instantaneous signal model, where each frame of the input mixture signal is modeled ...

Modern society is undergoing a fundamental change in the way we interact with technology. More and more devices are becoming "smart" by gaining advanced computation capabilities and communication interfaces, from household appliances over transportation systems to large-scale networks like the power grid. Recording, processing, and exchanging digital information is thus becoming increasingly important. As a growing share of devices is nowadays mobile and hence battery-powered, a particular interest in efficient digital signal processing techniques emerges. This thesis contributes to this goal by demonstrating methods for finding efficient algebraic solutions to various applications of multi-channel digital signal processing. These may not always result in the best possible system performance. However, they often come close while being significantly simpler to describe and to implement. The simpler description facilitates a thorough analysis of their performance which is crucial to design robust and reliable ...

Today’s society is characterized by an abundance of data that is generated at an unprecedented velocity. However, much of this data is immediately thrown away by compression or information extraction. In a compressed sensing (CS) setting the inherent sparsity in many datasets is exploited by avoiding the acquisition of superfluous data in the first place. We combine this technique with tensors, or multiway arrays of numerical values, which are higher-order generalizations of vectors and matrices. As the number of entries scales exponentially in the order, tensor problems are often large-scale. We show that the combination of simple, low-rank tensor decompositions with CS effectively alleviates or even breaks the so-called curse of dimensionality. After discussing the larger data fusion optimization framework for coupled and constrained tensor decompositions, we investigate three categories of CS type algorithms to deal with large-scale problems. First, ...

The spectral signatures of the materials contained in hyperspectral images, also called endmembers (EMs), can be significantly affected by variations in atmospheric, illumination or environmental conditions typically occurring within an image. Traditional spectral unmixing (SU) algorithms neglect the spectral variability of the endmembers, what propagates significant mismodeling errors throughout the whole unmixing process and compromises the quality of the estimated abundances. Therefore, significant effort have been recently dedicated to mitigate the effects of spectral variability in SU. However, many challenges still remain in how to best explore a priori information about the problem in order to improve the quality, the robustness and the efficiency of SU algorithms that account for spectral variability. In this thesis, new strategies are developed to address spectral variability in SU. First, an (over)-segmentation-based multiscale regularization strategy is proposed to explore spatial information about the abundance ...

Reverberation consists of a complex acoustic phenomenon that occurs inside rooms. Many audio signal processing methods, addressing source localization, signal enhancement and other tasks, often assume absence of reverberation. Consequently, reverberant environments are considered challenging as state-ofthe-art methods can perform poorly. The acoustics of a room can be described using a variety of mathematical models, among which, physical models are the most complete and accurate. The use of physical models in audio signal processing methods is often non-trivial since it can lead to ill-posed inverse problems. These inverse problems require proper regularization to achieve meaningful results and involve the solution of computationally intensive large-scale optimization problems. Recently, however, sparse regularization has been applied successfully to inverse problems arising in different scientific areas. The increased computational power of modern computers and the development of new efficient optimization algorithms makes it possible ...

Hyperspectral (HS) imaging, which consists of acquiring a same scene in several hundreds of contiguous spectral bands (a three dimensional data cube), has opened a new range of relevant applications, such as target detection [MS02], classification [C.-03] and spectral unmixing [BDPD+12]. However, while HS sensors provide abundant spectral information, their spatial resolution is generally more limited. Thus, fusing the HS image with other highly resolved images of the same scene, such as multispectral (MS) or panchromatic (PAN) images is an interesting problem. The problem of fusing a high spectral and low spatial resolution image with an auxiliary image of higher spatial but lower spectral resolution, also known as multi-resolution image fusion, has been explored for many years [AMV+11]. From an application point of view, this problem is also important as motivated by recent national programs, e.g., the Japanese next-generation space-borne ...

In natural listening environments, speech signals are easily distorted by variousacoustic interference, which reduces the speech quality and intelligibility of human listening; meanwhile, it makes difficult for many speech-related applications, such as automatic speech recognition (ASR). Thus, many speech enhancement (SE) algorithms have been developed in the past decades. However, most current SE algorithms are difficult to capture underlying speech information (e.g., phoneme) in the SE process. This causes it to be challenging to know what specific information is lost or interfered with in the SE process, which limits the application of enhanced speech. For instance, some SE algorithms aimed to improve human listening usually damage the ASR system. The objective of this dissertation is to develop SE algorithms that have the potential to capture various underlying speech representations (information) and improve the quality and intelligibility of noisy speech. This ...

This work is concerned with two image-processing problems, image deconvolution with incomplete observations and data fusion of spectral images, and with some of the algorithms that are used to solve these and related problems. In image-deconvolution problems, the diagonalization of the blurring operator by means of the discrete Fourier transform usually yields very large speedups. When there are incomplete observations (e.g., in the case of unknown boundaries), standard deconvolution techniques normally involve non-diagonalizable operators, resulting in rather slow methods, or, otherwise, use inexact convolution models, resulting in the occurrence of artifacts in the enhanced images. We propose a new deconvolution framework for images with incomplete observations that allows one to work with diagonalizable convolution operators, and therefore is very fast. The framework is also an efficient, high-quality alternative to existing methods of dealing with the image boundaries, such as edge ...

One desideratum in designing cognitive robots is autonomous learning of communication skills, just like humans. The primary step towards this goal is vocabulary acquisition. Being different from the training procedures of the state-of-the-art automatic speech recognition (ASR) systems, vocabulary acquisition cannot rely on prior knowledge of language in the same way. Like what infants do, the acquisition process should be data-driven with multi-level abstraction and coupled with multi-modal inputs. To avoid lengthy training efforts in a word-by-word interactive learning process, a clever learning agent should be able to acquire vocabularies from continuous speech automatically. The work presented in this thesis is entitled \emph{Constrained Non-negative Matrix Factorization for Vocabulary Acquisition from Continuous Speech}. Enlightened by the extensively studied techniques in ASR, we design computational models to discover and represent vocabularies from continuous speech with little prior knowledge of the language to ...

Face recognition is one of the most challenging problems of computer vision and pattern recognition. The difficulty in face recognition arises mainly from facial appearance variations caused by factors, such as expression, illumination, partial face occlusion, and time gap between training and testing data capture. Moreover, the performance of face recognition algorithms heavily depends on prior facial feature localization step. That is, face images need to be aligned very well before they are fed into a face recognition algorithm, which requires precise facial feature localization. This thesis addresses on solving these two main problems -facial appearance variations due to changes in expression, illumination, occlusion, time gap, and imprecise face alignment due to mislocalized facial features- in order to accomplish its goal of building a generic face recognition algorithm that can function reliably under real-world conditions. The proposed face recognition algorithm ...

The exponentially damped sinusoidal (EDS) model arises in numerous signal processing applications. It is therefore of great interest to have methods able to estimate the parameters of such a model in the single-channel as well as in the multi-channel case. Because such a model naturally lends itself to subspace representation, powerful matrix approaches like HTLS in the single-channel case, HTLSstack in the multi-channel case and HTLSDstack in the decimative case have been developed to estimate the parameters of the underlying EDS model. They basically consist in stacking the signal in Hankel (single-channel) or block Hankel (multi- channel) data matrices. Then, the signal subspace is estimated by means of the singular value decomposition (SVD). The parameters of the model, namely the amplitudes, the phases, the damping factors, and the frequencies, are estimated from this subspace. Note that the sample covariance matrix ...

This thesis develops new mathematical theory and presents novel recovery algorithms for discrete linear dynamical systems (LDS) with sparsity constraints on either control inputs or initial state. The recovery problems in this framework manifest as the problem of reconstructing one or more sparse signals from a set of noisy underdetermined linear measurements. The goal of our work is to design algorithms for sparse signal recovery which can exploit the underlying structure in the measurement matrix and the unknown sparse vectors, and to analyze the impact of these structures on the efficacy of the recovery. We answer three fundamental and interconnected questions on sparse signal recovery problems that arise in the context of LDS. First, what are necessary and sufficient conditions for the existence of a sparse solution? Second, given that a sparse solution exists, what are good low-complexity algorithms that ...