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 ...

When the noise affecting time series is colored with unknown statistics, a difficulty for periodic signal detection is to control the true significance level at which the detection tests are conducted. This thesis investigates the possibility of using training datasets of the noise to improve this control. Specifically, for the case of regularly sampled observations, we analyze the performances of various detectors applied to periodograms standardized using the noise training datasets. Emphasis is put on sparse detection in the Fourier domain and on the limitation posed by the necessary finite size of the training sets available in practice. We study the resulting false alarm and detection rates and show that the proposed standardization leads, in some cases, to powerful constant false alarm rate tests. Although analytical results are derived in an asymptotic regime, numerical results show that the theory accurately ...

Circumstellars environments observation is a key for the comprehension of planet formation. If the very large telescopes allow the resolution of these environments, their observation is difficult due to the high contrast between the environment and their host stars. In fact the host stars are 1000 to 10 000 times brighter than the environment, even 10 000 000 times brighter for exoplanets. When images of these circumstellar environnements are acquired in direct imaging, the signal of the environnements mixed to star light residuals. Yet, the light of the environment is partially linearly polarized while the light od the star is unpolarized. The instrument Infrared Dual-band Imaging and Spectroscopy (IRDIS) of the European Southern Observatory’s (ESO) Spectro-Polarimeter High-contrast Expolanet REsearch (SPHERE) instrument, installed at one of the four Very Large Telescopes (VLT) in Atacama in Chile, acquires datasets where the polarization ...

The characterization of Clock and Ephemeris error of the Global Navigation Satellite Systems is a key element to validate the assumptions for the integrity analysis of GNSS Safety of Life (SoL) applications. Specifically, the performance metrics of SoL applications require the characterization of the nominal User Range Errors (UREs) as well as the knowledge of the probability of a satellite, Psat or a constellation fault, Pconst, i.e. when one or more satellites are not in the nominal mode. We will focus on Advanced Autonomous Integrity Monitoring (ARAIM). The present dissertation carries-out an end-to-end characterization and analysis of Galileo and GPS satellites for ARAIM. It involves two main targets. First, the characterization of Galileo and GPS broadcast ephemeris and clock errors, to determine the fault probabilities Psat and Pconst, and the determination on an upper bound of the nominal satellite ranging ...

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 ...

This work has, as its objective, the development of non-invasive and low-cost systems for monitoring and automatic diagnosing specific neonatal diseases by means of the analysis of suitable video signals. We focus on monitoring infants potentially at risk of diseases characterized by the presence or absence of rhythmic movements of one or more body parts. Seizures and respiratory diseases are specifically considered, but the approach is general. Seizures are defined as sudden neurological and behavioural alterations. They are age-dependent phenomena and the most common sign of central nervous system dysfunction. Neonatal seizures have onset within the 28th day of life in newborns at term and within the 44th week of conceptional age in preterm infants. Their main causes are hypoxic-ischaemic encephalopathy, intracranial haemorrhage, and sepsis. Studies indicate an incidence rate of neonatal seizures of 2‰ live births, 11‰ for preterm ...

It is often assumed that interference or noise signals are Gaussian stochastic processes. Gaussian noise models are appealing as they usually result in noise suppression algorithms that are simple: i.e. linear and closed form. However, such linear techniques may be sub-optimal when the noise process is either a non-Gaussian stochastic process or a chaotic deterministic process. In the event of encountering such noise processes, improvements in noise suppression, relative to the performance of linear methods, may be achievable using nonlinear signal processing techniques. The application of interest for this thesis is maritime surveillance radar, where the main source of interference, termed sea clutter, is widely accepted to be a non-Gaussian stochastic process at high resolutions and/or at low grazing angles. However, evidence has been presented during the last decade which suggests that sea clutter may be better modelled as a ...

This thesis describes a visualization pipeline for sonar profiling data that show reflections of multiple sediments in the sea bottom and that cover huge survey areas with many gaps. Visualizing such data is not trivial, because they may be noisy and because data sets may be very large. The developed techniques are: (1) Quadtree interpolation for estimating new sediment reflections, at all gaps in the longitude-latitude plane. The quadtree is used for guiding the 3D interpolation process: gaps become small at low spatial resolutions, where they can be filled by interpolating between available reflections. In the interpolation, the reflection data are cross correlated in order to construct continuity of multiple, sloping reflections. (2) Segmentation and boundary refinement in an octree in order to detect sediments in the sonar data. In the refinement, coarse boundaries are reclassified by filtering the data ...

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 ...

The main objective of $\gamma$ spectrometry is to characterize the radioactive elements of an unknown source by studying the energy of the emitted $\gamma$ photons. When a photon interacts with a detector, its photonic energy is converted into an electrical pulse, whose integral energy is measured. The histogram obtained by collecting the energies can be used to identify radionucleides and measure their activity. However, at high counting rates, perturbations which are due to the stochastic aspect of the temporal signal can cripple the identification of the radioactive elements. More specifically, since the detector has a finite resolution, close arrival times of photons which can be modeled as an homogeneous Poisson process cause pileups of individual pulses. This phenomenon distorts energy spectra by introducing multiple fake spikes and prolonging artificially the Compton continuum, which can mask spikes of low intensity. The ...

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 ...

This thesis contributes to motion characterization through inertial and physiological signals captured by wearable devices and analyzed using signal processing and deep learning techniques. This research leverages the possibilities of motion analysis for three main applications: to know what physical activity a person is performing (Human Activity Recognition), to identify who is performing that motion (user identification) or know how the movement is being performed (motor anomaly detection). Most previous research has addressed human motion modeling using invasive sensors in contact with the user or intrusive sensors that modify the user’s behavior while performing an action (cameras or microphones). In this sense, wearable devices such as smartphones and smartwatches can collect motion signals from users during their daily lives in a less invasive or intrusive way. Recently, there has been an exponential increase in research focused on inertial-signal processing to ...

Automated analysis of facial expressions has been an active area of study due to its potential applications not only for intelligent human-computer interfaces but also for human facial behavior research. To advance automatic expression analysis, this thesis proposes and empirically proves two hypotheses: (i) 3D face data is a better data modality than conventional 2D camera images, not only for being much less disturbed by illumination and head pose effects but also for capturing true facial surface information. (ii) It is possible to perform detailed face registration without resorting to any face modeling. This means that data-driven methods in automatic expression analysis can compensate for the confounding effects like pose and physiognomy differences, and can process facial features more effectively, without suffering the drawbacks of model-driven analysis. Our study is based upon Facial Action Coding System (FACS) as this paradigm ...

In the field of airborne earth observation there is special attention to compact, cost effective, high resolution imaging sensors. Such sensors are foreseen to play an important role in small-scale remote sensing applications, such as the monitoring of dikes, watercourses, or highways. Furthermore, such sensors are of military interest; reconnaissance tasks could be performed with small unmanned aerial vehicles (UAVs), reducing in this way the risk for one's own troops. In order to be operated from small, even unmanned, aircrafts, such systems must consume little power and be small enough to fulfill the usually strict payload requirements. Moreover, to be of interest for the civil market, cost effectiveness is mandatory. Frequency Modulated Continuous Wave (FMCW) radar systems are generally compact and relatively cheap to purchase and to exploit. They consume little power and, due to the fact that they are ...

Recent improvements in the development of inertial and visual sensors allow building small, lightweight, and cheap motion capture systems, which are becoming a standard feature of smartphones and personal digital assistants. This dissertation describes developments of new motion sensing strategies using the inertial and inertial-visual sensors. The thesis contributions are presented in two parts. The first part focuses mainly on the use of inertial measurement units. First, the problem of sensor calibration is addressed and a low-cost and accurate method to calibrate the accelerometer cluster of this unit is proposed. The method is based on the maximum likelihood estimation framework, which results in a minimum variance unbiased estimator.Then using the inertial measurement unit, a probabilistic user-independent method is proposed for pedestrian activity classification and gait analysis.The work targets two groups of applications including human activity classificationand joint human activity and ...