In the area of image recognition, the so-called deep neural networks belong to the most promising models these days. They often achieve considerably better results than traditional techniques even without the necessity of any excessive task-oriented preprocessing. This thesis is devoted to the study and analysis of three basic variants of deep neural networks-namely the neocognitron, convolutional neural networks, and deep belief networks. Based on extensive testing of the described models on the standard task of handwritten digit recognition, the convolutional neural networks seem to be most suitable for the recognition of general image data. Therefore, we have used them also to classify images from two very large data sets-CIFAR-10 and ImageNet. In order to optimize the architecture of the applied networks, we have proposed a new pruning algorithm based on the Principal Component Analysis. Powered by TCPDF (www.tcpdf.org)

Der Einfluss des Menschen verändert die Erdoberfläche in gravierendem Maße. Die Anwendung von Landnutzungsmodellen ist etabliert, um derartige Prozesse zu analysieren und um Handlungsempfehlungen für Entscheidungsträger zu geben. Landnutzungsmodelle stehen unter dem Einfluss von Unsicherheiten, welche beim Interpretieren der Ergebnisse berücksichtigt werden müssen. Dennoch gibt es wenige Ansätze, die unterschiedliche Unsicherheitsquellen mit ihren Interdependenzen untersuchen und ihre Auswirkungen auf die projizierte Änderung der Landschaft analysieren. Aus diesem Grund ist das erste Ziel dieser Arbeit einen systematischen Ansatz zu entwickeln, der wesentliche Unsicherheitsquellen analysiert und ihre Fortentwicklung zur resultierenden Änderungskarte untersucht. Eine andere Herausforderung in der Landnutzungsmodellierung ist es, die Eignung von Projektionen abzuschätzen wenn keine Referenzdaten vorliegen. Bayes’sche Netze wurden als eine vielseitige Methode identifiziert, um das erste Ziel zu erreichen. Darüber hinaus wurden die Modellierungsschritte „Definition der Modellstruktur“, „Auswahl der Eingangsdaten“ und „Weiterverarbeitung der Daten“ als wesentliche Unsicherheitsquellen identifiziert. Um das zweite Ziel zu adressieren wurde eine Auswahl an Maßzahlen entwickelt. Diese quantifizieren Unsicherheit mit Hilfe einer projizierten Änderungskarte und ohne den Vergleich mit Referenzdaten. Mit diesen Maßzahlen ist es zusätzlich möglich zwischen quantitativer und räumlicher Unsicherheit zu unterscheiden. Vor allem in räumlichen Anwendungen wie der Landnutzungsmodellierung ist diese Möglichkeit wertvoll. Dennoch kann auch ein absolut sicheres Modell gleichzeitig ein falsches und nutzloses Modell sein. Deswegen wird ein Ansatz empfohlen, der die Beziehung zwischen Unsicherheit und Genauigkeit in bekannten Zeitschritten schätzt. Die entwickelten Ansätze geben wichtige Informationen um die Eignung von modellierten Entwicklungspfaden der Landnutzung zu verstehen. / Human influence has led to substantial changes to the Earth’s surface. Land change models are widely applied to analyze land change processes and to give recommendations for decision-making. Land change models are affected by uncertainties which have to be taken into account when interpreting their results. However, approaches which examine different sources of uncertainty with regard to their interdependencies and their influence on projected land change are rarely applied. The first objective of this thesis is therefore to develop a systematic approach which identifies major sources of uncertainty and the propagation to the resulting land change map. Another challenge in land change modeling is the estimation of the reliability of land change predictions when no reference data are available. Bayesian Belief Networks were identified as a useful technique to reach the first objective. Moreover, the modeling steps of “model structure definition”, “data selection” and “data preprocessing” were detected as relevant sources of uncertainty. To address the second research objective, a set of measures based on probabilities were developed. They quantify uncertainty by means of a single predicted land change map without using a reference map. It is additionally possible to differentiate uncertainty into its spatial and quantitative components by means of these measures. This is especially useful in spatial applications such as land change modeling. However, even a certain model can be wrong and therefore useless. Therefore, an approach is suggested which estimates the relationship between disagreement and uncertainty in known time steps to use this relationship in future time steps. The approaches give important information for understanding the reliability of modeled future development paths of land change.

Graphical techniques for modeling the dependencies of randomvariables have been explored in a variety of different areas includingstatistics, statistical physics, artificial intelligence, speech recognition, image processing, and genetics.Formalisms for manipulating these models have been developedrelatively independently in these research communities. In this paper weexplore hidden Markov models (HMMs) and related structures within the general framework of probabilistic independencenetworks (PINs). The paper contains a self-contained review of the basic principles of PINs.It is shown that the well-known forward-backward (F-B) and Viterbialgorithms for HMMs are special cases of more general inference algorithms forarbitrary PINs. Furthermore, the existence of inference and estimationalgorithms for more general graphical models provides a set of analysistools for HMM practitioners who wish to explore a richer class of HMMstructures.Examples of relatively complex models to handle sensorfusion and coarticulationin speech recognitionare introduced and treated within the graphical model framework toillustrate the advantages of the general approach.

Usually, human beings are able to quickly distinguish between different places, solely from their visual appearance. This is due to the fact that they can organize their space as composed of discrete units. These units, called ''semantic places'', are characterized by their spatial extend and their functional unity. Such a semantic category can thus be used as contextual information which fosters object detection and recognition. Recent works in semantic place recognition seek to endow the robot with similar capabilities. Contrary to classical localization and mapping works, this problem is usually addressed as a supervised learning problem. The question of semantic places recognition in robotics - the ability to recognize the semantic category of a place to which scene belongs to - is therefore a major requirement for the future of autonomous robotics. It is indeed required for an autonomous service robot to be able to recognize the environment in which it lives and to easily learn the organization of this environment in order to operate and interact successfully. To achieve that goal, different methods have been already proposed, some based on the identification of objects as a prerequisite to the recognition of the scenes, and some based on a direct description of the scene characteristics. If we make the hypothesis that objects are more easily recognized when the scene in which they appear is identified, the second approach seems more suitable. It is however strongly dependent on the nature of the image descriptors used, usually empirically derived from general considerations on image coding.Compared to these many proposals, another approach of image coding, based on a more theoretical point of view, has emerged the last few years. Energy-based models of feature extraction based on the principle of minimizing the energy of some function according to the quality of the reconstruction of the image has lead to the Restricted Boltzmann Machines (RBMs) able to code an image as the superposition of a limited number of features taken from a larger alphabet. It has also been shown that this process can be repeated in a deep architecture, leading to a sparse and efficient representation of the initial data in the feature space. A complex problem of classification in the input space is thus transformed into an easier one in the feature space. This approach has been successfully applied to the identification of tiny images from the 80 millions image database of the MIT. In the present work, we demonstrate that semantic place recognition can be achieved on the basis of tiny images instead of conventional Bag-of-Word (BoW) methods and on the use of Deep Belief Networks (DBNs) for image coding. We show that after appropriate coding a softmax regression in the projection space is sufficient to achieve promising classification results. To our knowledge, this approach has not yet been investigated for scene recognition in autonomous robotics. We compare our methods with the state-of-the-art algorithms using a standard database of robot localization. We study the influence of system parameters and compare different conditions on the same dataset. These experiments show that our proposed model, while being very simple, leads to state-of-the-art results on a semantic place recognition task.

This thesis details the development of a model of driver behaviour at railway level crossings that allows the probability of an accident under different conditions and interventions to be calculated. A method for classifying different crossings according to their individual risk levels is also described.

Recent developments in social media and web technologies offer new opportunities to access, analyze and process ever-increasing amounts of fashion-related data. In the appealing context of design and fashion, our main goal is to automatically suggest fashionable outfits based on the preferences extracted from real-world data provided either by individual users or gathered from the internet. In our case, the clothing items have the form of 2D-images. Especially for visual data processing tasks, recent models of deep neural networks are known to surpass human performance. This fact inspired us to apply the idea of transfer learning to understand the actual variability in clothing items. The principle of transfer learning consists in extracting the internal representa- tions formed in large convolutional networks pre-trained on general datasets, e.g., ImageNet, and visualizing its (similarity) structure. Together with transfer learn- ing, clustering algorithms and the image color schemes can be, namely, utilized when searching for related outfit items. Viable means applicable to generating new out- fits include deep belief networks and genetic algorithms enhanced by a convolutional network that models the outfit fitness. Although fashion-related recommendations remain highly subjective, the results we have achieved...

Les convertisseurs DC-DC suscitent un intérêt considérable en raison de leur puissance élevée et de leurs bonnes performances. Ils sont particulièrement utiles dans les systèmes multisources de production d'énergie électrique. Toutefois, en raison du grand nombre de composants sensibles utilisés dans ces circuits et comprenant des semi-conducteurs de puissance, des bobines et des condensateurs, une probabilité non négligeable de défaillance des composants doit être prise en compte. Cette thèse considère l'un des convertisseurs DC-DC les plus prometteurs - le convertisseur ZVS à pont isolé de type Buck. Une approche en deux étapes est présentée pour détecter et isoler les défauts en circuit ouvert dans les semi-conducteurs de puissance des convertisseurs DC-DC. La première étape concerne la détection et la localisation des défauts dans un convertisseur donne. La seconde étape concerne sur les systèmes munis de plusieurs convertisseurs DC-DC. Les méthodes proposées sont basées sur les réseaux Bayesiens (BBN). Les signaux utilisés dans ces méthodes sont ceux des entrées de mesure du système de commande et aucune mesure supplémentaire n'est requise. Un convertisseur expérimental ZVS à pont isolé de type Buck a été conçu et construit pour valider la détection et la localisation des défauts Sur un seul convertisseur. Ces méthodes peuvent être étendues à d'autres types de convertisseurs DC-DC. / DC-DC converters have received significant interest recently as a result of their high power capabilities and good power quality. They are of particular interest in systems with multiple sources of energy. However due to the large number of sensitive components including power semiconductor devices, coils, and capacitors used in such circuits there is a high likelihood of component failure. This thesis considers one of the most promising DC-DC converters—the ZVS full bridge isolated Buck converter. An approach with two stages is presented to detect and isolate opencircuit faults in the power semiconductor devices in systems with DC-DC converters. The first stage is the fault detection and isolation for a single DC-DC converter, while the second stage works on a system with multiple DC-DC converters. The proposed methods are based on Bayesian Belief Network (BBN). The signals used in the proposed methods are already available as measurement inputs to control system and no additional measurements are required. An experimental ZVS full bridge isolated Buck converter has been designed and built to validate the fault detection and isolation method on a single converter. The methods can be used with other DC-DC converter typologies employing similar analysis and principals.

Sleep is an important part of life as it affects the performance of one's activities during all awake hours. The study of sleep and wakefulness is therefore of great interest, particularly to the clinical and medical fields where sleep disorders are diagnosed. When studying sleep, it is common to talk about different types, or stages, of sleep. A common task in sleep research is to determine the sleep stage of the sleeping subject as a function of time. This process is known as sleep stage scoring. In this study, I seek to determine whether there is any benefit to using unsupervised feature learning in the context of electroencephalogram-based (EEG) sleep scoring. More specifically, the effect of generating and making use of new feature representations for hand-crafted features of sleep data – meta-features – is studied. For this purpose, two scoring algorithms have been implemented and compared. Both scoring algorithms involve segmentation of the EEG signal, feature extraction, feature selection and classification using a support vector machine (SVM). Unsupervised feature learning was implemented in the form of a dimensionality-reducing deep-belief network (DBN) which the feature space was processed through. Both scorers were shown to have a classification accuracy of about 76 %. The application of unsupervised feature learning did not affect the accuracy significantly. It is speculated that with a better choice of parameters for the DBN in a possible future work, the accuracy may improve significantly. / Sömnen är en viktig del av livet eftersom den påverkar ens prestation under alla vakna timmar. Forskning om sömn and vakenhet är därför av stort intresse, i synnerhet för de kliniska och medicinska områdena där sömnbesvär diagnostiseras. I forskning om sömn är det är vanligt att tala om olika typer av sömn, eller sömnstadium. En vanlig uppgift i sömnforskning är att avgöra sömnstadiet av den sovande exemplaret som en funktion av tiden. Den här processen kallas sömnmätning. I den här studien försöker jag avgöra om det finns någon fördel med att använda oövervakad inlärning av särdrag för att utföra elektroencephalogram-baserad (EEG) sömnmätning. Mer specifikt undersöker jag effekten av att generera och använda nya särdragsrepresentationer som härstammar från handgjorda särdrag av sömndata – meta-särdrag. Två sömnmätningsalgoritmer har implementerats och jämförts för det här syftet. Sömnmätningsalgoritmerna involverar segmentering av EEG-signalen, extraktion av särdragen, urval av särdrag och klassificering genom användning av en stödvektormaskin (SVM). Oövervakad inlärning av särdrag implementerades i form av ett dimensionskrympande djuptrosnätverk (DBN) som användes för att bearbetasärdragsrymden. Båda sömnmätarna visades ha en klassificeringsprecision av omkring 76 %. Användningen av oövervakad inlärning av särdrag hade ingen signifikant inverkan på precisionen. Det spekuleras att precisionen skulle kunna höjas med ett mer lämpligt val av parametrar för djuptrosnätverket.

Recent developments in social media and web technologies offer new opportunities to access, analyze and process ever-increasing amounts of fashion-related data. In the appealing context of design and fashion, our main goal is to automatically suggest fashionable outfits based on the preferences extracted from real-world data provided either by individual users or gathered from the internet. In our case, the clothing items have the form of 2D-images. Especially for visual data processing tasks, recent models of deep neural networks are known to surpass human performance. This fact inspired us to apply the idea of transfer learning to understand the actual variability in clothing items. The principle of transfer learning consists in extracting the internal representa- tions formed in large convolutional networks pre-trained on general datasets, e.g., ImageNet, and visualizing its (similarity) structure. Together with transfer learn- ing, clustering algorithms and the image color schemes can be, namely, utilized when searching for related outfit items. Viable means applicable to generating new out- fits include deep belief networks and genetic algorithms enhanced by a convolutional network that models the outfit fitness. Although fashion-related recommendations remain highly subjective, the results we have achieved...

Il est généralement facile pour les humains de distinguer rapidement différents lieux en se basant uniquement sur leur aspect visuel. . Ces catégories sémantiques peuvent être utilisées comme information contextuelle favorisant la détection et la reconnaissance d'objets. Des travaux récents en reconnaissance des lieux visent à doter les robots de capacités similaires. Contrairement aux travaux classiques, portant sur la localisation et la cartographie, cette tâche est généralement traitée comme un problème d'apprentissage supervisé.La reconnaissance de lieux sémantiques - la capacité à reconnaître la catégorie sémantique à laquelle une scène appartient – peut être considérée comme une condition essentielle en robotique autonome. Un robot autonome doit en effet pouvoir apprendre facilement l'organisation sémantique de son environnement pour pouvoir fonctionner et interagir avec succès. Pour atteindre cet objectif, différentes méthodes ont déjà été proposées. Certaines sont basées sur l'identification des objets comme une condition préalable à la reconnaissance des scènes, et d'autres fondées sur une description directe des caractéristiques de la scène. Si nous faisons l'hypothèse que les objets sont plus faciles à reconnaître quand la scène dans laquelle ils apparaissent est bien identifiée, la deuxième approche semble plus appropriée. Elle est cependant fortement dépendante de la nature des descripteurs d'images utilisées qui sont généralement dérivés empiriquement a partir des observations générales sur le codage d'images.En opposition avec ces propositions, une autre approche de codage des images, basée sur un point de vue plus théorique, a émergé ces dernières années. Les modèles d'extraction de caractéristiques fondés sur le principe de la minimisation d'une fonction d'énergie en relation avec un modèle statistique génératif expliquant au mieux les données, ont abouti à l'apparition des Machines de Boltzmann Restreintes (Rectricted Boltzmann Machines : RBMs) capables de coder une image comme la superposition d'un nombre limité de caractéristiques extraites à partir d'un plus grand alphabet. Il a été montré que ce processus peut être répété dans une architecture plus profonde, conduisant à une représentation parcimonieuse et efficace des données initiales dans l'espace des caractéristiques. Le problème complexe de la classification dans l'espace de début est ainsi remplacé par un problème plus simple dans l'espace des caractéristiques.Dans ce travail, nous montrons que la reconnaissance sémantiques des lieux peut être réalisée en considérant des mini-images au lieu d'approches plus classiques de type ''sacs-de-mots'' et par l'utilisation de réseaux profonds pour le codage des images. Après avoir realisé un codage approprié, une régression softmax dans l'espace de projection est suffisante pour obtenir des résultats de classification prometteurs. A notre connaissance, cette approche n'a pas encore été proposée pour la reconnaissance de scène en robotique autonome.Nous avons comparé nos méthodes avec les algorithmes de l'état-de-l'art en utilisant une base de données standard de localisation de robot. Nous avons étudié l'influence des paramètres du système et comparé les différentes conditions sur la même base de données. Les expériences réalisées montrent que le modèle que nous proposons, tout en étant très simple, conduit à des résultats comparables à l'état-de-l'art sur une tâche de reconnaissance de lieux sémantiques. / Usually, human beings are able to quickly distinguish between different places, solely from their visual appearance. This is due to the fact that they can organize their space as composed of discrete units. These units, called ``semantic places'', are characterized by their spatial extend and their functional unity. Such a semantic category can thus be used as contextual information which fosters object detection and recognition. Recent works in semantic place recognition seek to endow the robot with similar capabilities. Contrary to classical localization and mapping works, this problem is usually addressed as a supervised learning problem. The question of semantic places recognition in robotics - the ability to recognize the semantic category of a place to which scene belongs to - is therefore a major requirement for the future of autonomous robotics. It is indeed required for an autonomous service robot to be able to recognize the environment in which it lives and to easily learn the organization of this environment in order to operate and interact successfully. To achieve that goal, different methods have been already proposed, some based on the identification of objects as a prerequisite to the recognition of the scenes, and some based on a direct description of the scene characteristics. If we make the hypothesis that objects are more easily recognized when the scene in which they appear is identified, the second approach seems more suitable. It is however strongly dependent on the nature of the image descriptors used, usually empirically derived from general considerations on image coding.Compared to these many proposals, another approach of image coding, based on a more theoretical point of view, has emerged the last few years. Energy-based models of feature extraction based on the principle of minimizing the energy of some function according to the quality of the reconstruction of the image has lead to the Restricted Boltzmann Machines (RBMs) able to code an image as the superposition of a limited number of features taken from a larger alphabet. It has also been shown that this process can be repeated in a deep architecture, leading to a sparse and efficient representation of the initial data in the feature space. A complex problem of classification in the input space is thus transformed into an easier one in the feature space. This approach has been successfully applied to the identification of tiny images from the 80 millions image database of the MIT. In the present work, we demonstrate that semantic place recognition can be achieved on the basis of tiny images instead of conventional Bag-of-Word (BoW) methods and on the use of Deep Belief Networks (DBNs) for image coding. We show that after appropriate coding a softmax regression in the projection space is sufficient to achieve promising classification results. To our knowledge, this approach has not yet been investigated for scene recognition in autonomous robotics. We compare our methods with the state-of-the-art algorithms using a standard database of robot localization. We study the influence of system parameters and compare different conditions on the same dataset. These experiments show that our proposed model, while being very simple, leads to state-of-the-art results on a semantic place recognition task.