Baigiamajame magistro darbe yra nagrinėjamas bekontaktis pulso matavimo metodas. Darbo tikslams pasiekti naudota Logitech C310 internetinė vaizdo kamera. Įrašomo vaizdo dydis yra 640X480 pikselių. Filmavimo sparta – 15 kadrų per sekundę. Vaizdo įrašo ilgis – 30 sekundžių. Tiriamieji buvo filmuojami apie 0,5 m atstumu nuo kameros. Tiriamųjų amžius nuo 24 iki 64 metų. Vaizdas buvo įrašomas, esant įvairiam apšvietimui: tiek dienos metu, tiek šviečiant skirtingo galingumo lempoms. Rezultatams palyginti buvo naudojamas ant riešo uždedamas pulso matavimo prietaisas ReliOn, kurio veikimas pagrįstas kraujagyslėse pulsuojančio kraujo spaudimo kitimu. Išgautam pulso signalui apdoroti, palyginimui buvo panaudoti du nepriklausomų komponenčių analizės algoritmai: Fast ICA bei stSobi. Eksperimentams atlikti buvo naudojama C# programavimo kalba ir Matlab 2008 matematinis skaičiavimo paketas. / The thesis analyses the non-contact cardiac pulse measurement method. To achieve work main goals Logitech C310 web camera was used. Video resolution was 640X480 pixels. Video capture speed was 15 frames per second. Video length was 30 seconds. Distance from web camera to human face was ~ 0,5 m. Participant age varied from 24 to 64 years old. Video was captured with different light sources: sun, lamps with different power. For results comparison ReliOn handy pulse measurement device was used. Pulse signal was filtered using two independent component analysis algorithms: Fast ICA and stSobi. Experiments have been made using C# programming language and Matlab 2008 mathematical package.

Le développement de techniques à haut débit fournit de nombreuses données sur le fonctionnement de réseaux de régulation. Il devient donc de plus en plus important de développer des techniques qui permettent de déduire la topologie et le fonctionnement des réseaux de régulation à partir des données expérimentales. La plupart des études dans ce domaine se focalisent sur la reconstruction de l'architecture locale du réseau de régulation et la détermination des paramètres qui relient les composants du réseau. Cependant, les réseaux biologiques ne sont jamais entièrement connus. L'absence d'un noeud important dans le réseau de régulation peut facilement conduire à de mauvaises prédictions de la structure du réseau ou des paramètres d'interactions. Dans cette thèse, nous proposons une méthode qui permet d'inférer l'existence, le profile d'expression et la connexion au reste du réseau d'un gène (ou de gènes) manquant. Pour résoudre ce problème difficile, nous devons simplifier la description du réseau de régulation. Nous faisons l'hypothèse communément acceptée que les interactions dans le réseau sont décrites par des fonctions de Hill. Nous approximons ces fonctions trop compliquées par des fonctions de puissance et nous montrons que cette simplification préserve la dynamique du réseau. En prenant le logarithme du système d'équations nous convertissons le système non-linéaire en un système linéaire. De nombreux outils sont disponibles pour analyser des systèmes linéaires. Nous utilisons l'analyse factorielle (FA) et l'analyse de composants indépendants (ICA) pour extraire le profil d'expression du gène inconnu à partir des profils d'expression des parties connues du réseau de régulation. Après avoir estimé le pattern d'expression du gène inconnu, nous explorons les différentes possibilités de connecter ce gène au reste du réseau. Une recherche exhaustive est trop coûteuse pour des grands réseaux de régulation. Nos proposons donc un algorithme de réduction de l'espace de recherche pour diminuer le nombre de calculs nécessaires. L'algorithme proposé est robuste au bruit expérimental et le profil d'expression du gène inconnu est retrouvé avec une probabilité de 80% dans des réseaux de petite taille et avec une probabilité de 60% pour des grands réseaux. FA est plus efficace que ICA pour extraire le profile du gène inconnu. L'algorithme est finalement appliqué à un réseau biologique réel: le réseau de régulation de la transcription du gène acs d'Escherichia coli. Nous prédisons qu'il y a un gène manquant dans ce réseau et les deux méthodes d'extraction du signal trouvent un profil d'expression très similaire pour le gène inconnu. De plus, ce profil d'expression est identique dans trois contextes expérimentaux différents : la souche sauvage, la souche dont l'adénylate cyclase a été délété et cette même souche complémentée par des l'AMPc ajouté au milieu de croissance. Puisque le profil d'expression du gène inconnu reste le mŘme dans les trois souches nous pouvons conclure que ce gène est indépendant de l'AMPc. Les deux méthodes d'extraction du profil d'expression prédisent deux structures différentes du réseau complet. FA prédit que le gène manquant contrôle l'expression de fis, tandis que ICA prédit que le gène inconnu contrôle d'expression de crp. / With the development of hight-throughput technologies, the investigation of the topologies and the functioning of genetic regulatory networks have become an important research topic in recent years. Most of the studies concentrate on reconstructing the local architecture of genetic regulatory networks and the determination of the corresponding interaction parameters. The preferred data sources are time series expression data. However, inevitably one or more important members of the regulatory network will remain unknown. The absence of important members of the genetic circuit leads to incorrectly inferred network topologies and control mechanisms. In this thesis we propose a method to infer the connection and expression pattern of these “missing genes”. In order to make the problem tractable, we have to make further simplifying assumptions. We assume that the interactions within the network are described by Hill-functions. We then approximate these functions by power-law functions. We show that this simplification still captures the dynamic regulatory behaviors of the network. The genetic control system can now be converted to linear model by using a logarithm transformation. In another word, we can analyze the genetic regulatory networks by linear approaches. In the logarithmic space, we propose a procedure for extracting the expression profile of a missing gene within the otherwise defined genetic regulatory network. The algorithm also determines the regulatory connections of this missing gene to the rest of the regulation network. The inference algorithm is based on Factor Analysis, a well-developed multivariate statistical analysis approach that is used to investigate unknown, underlying features of an ensemble of data, in our case the promoter activities and intracellular concentrations of the known genes. We also explore a second blind sources separation method, “Independent Component Analysis”, which is also commonly used to estimate hidden signals. Once the expression profile of the missing gene has been derived, we investigate possible connections of this gene to the remaining network by methods of search space reduction. The proposed method of inferring the expression profile of a missing gene and connecting it to a known network structure is applied to artificial genetic regulatory networks, as well as a real biologicial network studied in the laboratory: the acs regulatory network of Escherichia coli. In these applications we confirm that power-law functions are a good approximation of Hill-functions. Factor Analysis predicts the expression profiles of missing genes with a high accuracy of 80% in small artificial genetic regulatory networks. The accuracy of Factor Analysis of predicting the expression profiles of missing genes of large artificial genetic regulatory networks is 60%. In contrast, Independent Component Analysis is less powerful than Factor Analysis in extracting the expression profiles of missing components in small, as well as large, artificial genetic regulatory networks. Both Factor Analysis and Independent Component suggest that only one missing gene is sufficient to explain the observed expression profiles of Acs, Fis and Crp. The expression profiles of the missing genes in the △cya strain and in the △cya strain supplemented with cAMP estimated by Factor Analysis and Independent Component Analysis are very similar. Factor Analysis suggests that fis is regulated by the missing genes, while Independent Component Analysis suggests that crp is controlled by the missing gene.

Doctor of Philosophy / This thesis is based on research on financial time series analysis using pattern recognition methods. The first part of this research focuses on univariate time series analysis using different pattern recognition methods. First, probabilities of basic patterns are used to represent the features of a section of time series. This feature can remove noise from the time series by statistical probability. It is experimentally proven that this feature is successful for pattern repeated time series. Second, a multiscale Gaussian gravity as a pattern relationship measurement which can describe the direction of the pattern relationship is introduced to pattern clustering. By searching for the Gaussian-gravity-guided nearest neighbour of each pattern, this clustering method can easily determine the boundaries of the clusters. Third, a method that unsupervised pattern classification can be transformed into multiscale supervised pattern classification by multiscale supervisory time series or multiscale filtered time series is presented. The second part of this research focuses on multivariate time series analysis using pattern recognition. A systematic method is proposed to find the independent variables of a group of share prices by time series clustering, principal component analysis, independent component analysis, and object recognition. The number of dependent variables is reduced and the multivariate time series analysis is simplified by time series clustering and principal component analysis. Independent component analysis aims to find the ideal independent variables of the group of shares. Object recognition is expected to recognize those independent variables which are similar to the independent components. This method provides a new clue to understanding the stock market and to modelling a large time series database.

This dissertation deals with dynamics of engineering structures and principally discusses the identification of the modal parameters (i.e., natural frequencies, damping ratios and vibration modes) using output-only information, the excitation sources being considered as unknown and unmeasurable. To solve these kind of problems, a quite large selection of techniques is available in the scientific literature, each of them possessing its own features, advantages and limitations. One common limitation of most of the methods concerns the post-processing procedures that have proved to be delicate and time consuming in some cases, and usually require good users expertise. The constant concern of this work is thus the simplification of the result interpretation in order to minimize the influence of this ungovernable parameter. A new modal parameter estimation approach is developed in this work. The proposed methodology is based on the so-called Blind Source Separation techniques, that aim at reducing large data set to reveal its essential structure. The theoretical developments demonstrate a one-to-one relationship between the so-called mixing matrix and the vibration modes. Two separation algorithms, namely the Independent Component Analysis and the Second-Order Blind Identification, are considered. Their performances are compared, and, due to intrinsic features, one of them is finally identified as more suitable for modal identification problems. For the purpose of comparison, numerous academic case studies are considered to evaluate the influence of parameters such as damping, noise and nondeterministic excitations. Finally, realistic examples dealing with a large number of active modes, typical impact hammer modal testing and operational testing conditions, are studied to demonstrate the applicability of the proposed methodology for practical applications.

We live in a dynamic world full of noises and interferences. The conventional acoustic echo cancellation (AEC) framework based on the least mean square (LMS) algorithm by itself lacks the ability to handle many secondary signals that interfere with the adaptive filtering process, e.g., local speech and background noise. In this dissertation, we build a foundation for what we refer to as the system approach to signal enhancement as we focus on the AEC problem. We first propose the residual echo enhancement (REE) technique that utilizes the error recovery nonlinearity (ERN) to "enhances" the filter estimation error prior to the filter adaptation. The single-channel AEC problem can be viewed as a special case of semi-blind source separation (SBSS) where one of the source signals is partially known, i.e., the far-end microphone signal that generates the near-end acoustic echo. SBSS optimized via independent component analysis (ICA) leads to the system combination of the LMS algorithm with the ERN that allows for continuous and stable adaptation even during double talk. Second, we extend the system perspective to the decorrelation problem for AEC, where we show that the REE procedure can be applied effectively in a multi-channel AEC (MCAEC) setting to indirectly assist the recovery of lost AEC performance due to inter-channel correlation, known generally as the "non-uniqueness" problem. We develop a novel, computationally efficient technique of frequency-domain resampling (FDR) that effectively alleviates the non-uniqueness problem directly while introducing minimal distortion to signal quality and statistics. We also apply the system approach to the multi-delay filter (MDF) that suffers from the inter-block correlation problem. Finally, we generalize the MCAEC problem in the SBSS framework and discuss many issues related to the implementation of an SBSS system. We propose a constrained batch-online implementation of SBSS that stabilizes the convergence behavior even in the worst case scenario of a single far-end talker along with the non-uniqueness condition on the far-end mixing system. The proposed techniques are developed from a pragmatic standpoint, motivated by real-world problems in acoustic and audio signal processing. Generalization of the orthogonality principle to the system level of an AEC problem allows us to relate AEC to source separation that seeks to maximize the independence, hence implicitly the orthogonality, not only between the error signal and the far-end signal, but rather, among all signals involved. The system approach, for which the REE paradigm is just one realization, enables the encompassing of many traditional signal enhancement techniques in analytically consistent yet practically effective manner for solving the enhancement problem in a very noisy and disruptive acoustic mixing environment.

The task of automatically recognizing targets in IR imagery has a history of approximately 25 years of research and development. ATR is an application of pattern recognition and scene analysis in the field of defense industry and it is still one of the challenging problems. This thesis may be viewed as an exploratory study of ATR problem with encouraging recognition algorithms implemented in the area. The examined algorithms are among the solutions to the ATR problem, which are reported to have good performance in the literature. Throughout the study, PCA, subspace LDA, ICA, nearest mean classifier, K nearest neighbors classifier, nearest neighbor classifier, LVQ classifier are implemented and their performances are compared in the aspect of recognition rate. According to the simulation results, the system, which uses the ICA as the feature extractor and LVQ as the classifier, has the best performing results. The good performance of this system is due to the higher order statistics of the data and the success of LVQ in modifying the decision boundaries.

Doctor of Philosophy / This thesis is based on research on financial time series analysis using pattern recognition methods. The first part of this research focuses on univariate time series analysis using different pattern recognition methods. First, probabilities of basic patterns are used to represent the features of a section of time series. This feature can remove noise from the time series by statistical probability. It is experimentally proven that this feature is successful for pattern repeated time series. Second, a multiscale Gaussian gravity as a pattern relationship measurement which can describe the direction of the pattern relationship is introduced to pattern clustering. By searching for the Gaussian-gravity-guided nearest neighbour of each pattern, this clustering method can easily determine the boundaries of the clusters. Third, a method that unsupervised pattern classification can be transformed into multiscale supervised pattern classification by multiscale supervisory time series or multiscale filtered time series is presented. The second part of this research focuses on multivariate time series analysis using pattern recognition. A systematic method is proposed to find the independent variables of a group of share prices by time series clustering, principal component analysis, independent component analysis, and object recognition. The number of dependent variables is reduced and the multivariate time series analysis is simplified by time series clustering and principal component analysis. Independent component analysis aims to find the ideal independent variables of the group of shares. Object recognition is expected to recognize those independent variables which are similar to the independent components. This method provides a new clue to understanding the stock market and to modelling a large time series database.

Σκοπός της παρούσας διπλωματικής εργασίας είναι η μελέτη του αντίστροφου καθορισμού πηγής (inverse source localization problem) και του ρυθμού μ (mu). Έχοντας ως δεδομένο το σήμα του ΗΕΓ γίνεται προσπάθεια µέσω της εφαρμογής της μεθόδου Ανάλυσης Ανεξάρτητων συνιστωσών (ICA) να προσδιοριστούν οι συνιστώσες οι οποίες σχετίζονται με τις περιοχές του εγκεφάλου που ενεργοποιούνται από την κίνηση των χεριών. Με βάση τη λειτουργία της αισθητηριοκινητικής περιοχής του εγκεφάλου και τις ιδιότητες του ρυθμού μ, γίνεται μια μελέτη πάνω στις συνιστώσες που προκύπτουν από την ICA τόσο σε δεδομένα από πραγματική κίνηση, όσο και σε δεδομένα από νοερή κίνηση, καθώς και στην εφαρμογή που μπορεί να υπάρχει σε συστήματα Διεπαφής Εγκεφάλου – Υπολογιστή. / The subject of this diploma thesis is the study of the inverse source localization problem and the mu rhythm. Performing Independent Component Analysis (ICA) on EEG data, we try to specify the components that are related to the brain areas activated by hand movement. By focusing on the function of the somatosensory brain area and the properties or mu rhythm, we study the components resulting from Independent Component Analysis on data of both real and imaginary movement, as well as the possible implementations on Brain – Computer Interface systems.

Το θέμα της κατασκευής ενός συστήματος αναγνώρισης προτύπων για τα ηχητικά σήματα ενός ανθρώπου που κοιμάται είναι ένα από τα ανοιχτά ζητήματα της Βιοιατρικής. Στην παρούσα διπλωματική εξετάζουμε την εξαγωγή ερμηνεύσιμων σημάτων που αντιστοιχούν στον καρδιακό ρυθμό, την αναπνοή και το ροχαλητό. Χρησιμοποιούμε μεθόδους Ανάλυσης σε Ανεξάρτητες Συνιστώσες και μεθόδους Τυφλού Διαχωρισμού που εκμεταλεύονται Στατιστικές Δεύτερης Τάξης. Συμπεραίνουμε ότι οι δεύτερες είναι οι πλέον κατάλληλες όταν συνοδεύονται από ένα στάδιο προεπεξεργασίας που αφορά ανάλυση σε ζώνες συχνοτήτων. / The design of a non-intrusive Pattern Recognition System to estimate the sleep sounds is an open problem of Bioengineering. We use recordings from body-sensors to estimate the heart beat, the breathing and the snoring. In this thesis we examine the effectiveness of Independent Component Analysis for this Blind Source Separation Problem and we compare it with methods that perform Source Separation using Second Order Statistics. We take into account the temporal structure of the sources as well as the presence of noise. Our system is greatly improved by a preprocessing stage of targeted subband decomposition which uses a priori knowledge about the sources. We propose an efficient solution to this problem which is confirmed by medical data.

Σκοπός της παρούσας διπλωματικής εργασίας είναι η δημιουργία ενός συστήματος μίξης ηχητικών σημάτων και προσπάθεια διαχωρισμού τους με βάση τις μεθόδους τυφλού διαχωρισμού σημάτων. Έχοντας ως δεδομένα τα αρχικά σήματα των πηγών γίνεται προσπάθεια, αρχικά μέσω της εφαρμογής της μεθόδου Ανάλυσης Ανεξάρτητων Συνιστωσών (ICA) για την περίπτωση της στιγμιαίας μίξης και στη συνέχεια μέσω της χρήσης αλγορίθμων που στηρίζονται στο μοντέλο παράλληλου παράγοντα (PARAFAC) για την περίπτωση της συνελικτικής μίξης, να προσδιοριστούν τα σήματα των πηγών από τα σήματα μίξης. Επιπλέον, τροποποιώντας τις παραμέτρους του συστήματος που μελετάμε σε κάθε περίπτωση, προσπαθούμε να πετύχουμε τη βέλτιστη απόδοση του διαχωρισμού. / The subject of this diploma thesis is the creation of a mixing system of speech signals and the attempt of their separation using the methods of blind source separation (BSS). Considering the original source signals known, we attempt, firstly by using independent component analysis for instantaneous mixtures and then by using PARAFAC model for convolutive mixtures, to extract the original source signals from the mixing signals. Moreover, by modifying the parameters of the system we make an effort to achieve the best performance of the separation.