過去的研究指出，具有威脅訊息的情緒臉孔會比正向情緒臉孔快速地吸引個體的注意力。視覺搜尋研究發現情緒臉孔的突顯程度會引發由下而上的注意力攫取，而由下而上的注意力攫取是否能夠被由上而下的因素所凌駕，一直是一個很重要的論爭，因此，本研究同時呈現兩種情緒臉孔，並以文字線索引發由上而下注意力選擇，以探討由上而下的因素是否能夠凌駕情緒臉孔之注意力攫取，並比較不同情緒臉孔攫取注意力的強度。實驗一探討文字線索是否能夠幫助注意力搜尋目標情緒臉孔。實驗二使用一致效果當作注意力攫取指標，探討當臉孔情緒與作業相關時，由上而下因素是否能夠凌駕干擾情緒臉孔為生氣臉孔或快樂臉孔之注意力攫取，以及比較生氣與快樂臉孔攫取注意力的強度。實驗三延續實驗二的研究，將臉孔情緒改為與作業無關，比較生氣臉孔與快樂臉孔攫取注意力的強度。實驗四與實驗二實驗程序相同，但增加倒立中性臉孔與生氣臉孔或快樂臉孔同時呈現情況，如此在以倒立中性臉孔為目標或干擾臉孔情況，可以比較生氣臉孔與快樂臉孔之由上而下因素注意力選擇與由下而上注意力攫取強度。本研究結果在大部分情況下並未獲得一致效果，顯示由上而下之注意力選擇會影響由下而上之情緒臉孔攫取注意力，使得干擾臉孔未能吸引注意力於其上。然而由下而上注意力攫取仍影響搜尋情緒臉孔的反應時間，結果顯示生氣臉孔與快樂臉孔競爭下，生氣臉孔的突顯程度大於快樂臉孔，說明生氣臉孔較快樂臉孔吸引注意力。 / Previous studies using visual search task showed that the saliency of the emotional faces was sufficient to induce the bottom-up attentional capture. It is a critical debate that whether top-down factors can override bottom-up attentional capture. To address this issue, we presented two emotional faces at the same display while using word cues to induce top-down attentional selection. Experiment 1 investigated the effect of word cues on the target face searching. Using congruency effect as an index of attentional capture, Experiment 2 investigated whether top-down factors could override attentional capture of the emotional distractors when emotion is task-relevant and the degree of attentional capture of angry faces was compared with that of happy faces. In Experiment 3, face emotion was changed to be task-irrelevant rather than task-relevant, and the degree of attentional capture of angry and happy faces was also contrasted. In Experiment 4, the invert neutral face was included additionally as another target or distractor. We can use the invert neutral face as the common basis to dissociate the components of top-down attentional selection and bottom-up attentional capture, and contrast the attention effects between angry and happy faces in these two aspects independently. The results showed that congruency effect could not be found in most conditions. This implies that top-down attentional selection can influence the bottom-up attentional capture of emotional faces. But reaction time for searching target was still found to be influenced by bottom-up process. When an angry face competes with a happy face, the saliency of the angry face is higher than the happy face.

The purpose of this master thesis was to study different content-aware video retargeting techniques, concentrating on a generalization of seam carving for video. Focus have also been put on the possibility to combine different techniques to achieve better retargeting of both multi-shot video and single-shot video. This also involved significant studies of automatic cut detection and different measures of video content. The work resulted in a prototype application for semi-automatic video retargeting, developed in Matlab. Three different retargeting techniques, seam carving, automated pan & scan and subsampling using bi-cubic interpolation, have been implemented in the prototype. The techniques have been evaluated and compared to each other from a content preservation perspective and a perceived quality perspective. / Syftet med examensarbetet har varit att studera tekniker för ändring av bredd/höjd-förhållandet i videosekvenser, där hänsyn tas till innehållet i bilderna. Fokus har lagts på en generalisering av "seam carving" för video och möjligheterna att kombinera olika tekniker för att nå bättre kvalitet både för videosekvenser som består av endast ett, eller flera, klipp. Detta innefattade således också omfattande studier av automatisk klippdetektering och olika mått av videoinnehåll. Arbetet har resulterat i en prototypapplikation utvecklad i Matlab för halvautomatisk förändring av bildförhållande där hänsyn tas till innehållet i sekvenserna. I prototypen finns tre metoder implementerade, "seam carving", automatiserad "pan & scan" och nedsampling med bi-kubisk interpolering. Dessa metoder har utvärderats och jämförts med varandra från ett innehållsbevarande perspektiv och ett kvalitetsperspektiv.

Le niveau d'exigence minimum pour tout système 3D (images stéréoscopiques) est de garantir le confort visuel des utilisateurs. Le confort visuel est un des trois axes perceptuels de la qualité d'expérience (QoE) 3D qui peut être directement lié aux paramètres techniques du système 3D. Par conséquent, le but de cette thèse est de caractériser objectivement l'impact de ces paramètres sur la perception humaine afin de contrôler la qualité stéréoscopique. La première partie de la thèse examine l'intérêt de prendre en compte l'attention visuelle des spectateurs dans la conception d'une mesure objective de qualité 3D. Premièrement, l'attention visuelle en 2D et 3D sont comparées en utilisant des stimuli simples. Les conclusions de cette première expérience sont validées en utilisant des scènes complexes avec des disparités croisées et décroisées. De plus, nous explorons l'impact de l'inconfort visuel causé par des disparités excessives sur l'attention visuelle. La seconde partie de la thèse est dédiée à la conception d'un modèle objectif de QoE pour des vidéos 3D, basé sur les seuils perceptuels humains et le niveau d'acceptabilité. De plus nous explorons la possibilité d'utiliser la modèle proposé comme une nouvelle échelle subjective. Pour la validation de ce modèle, des expériences subjectives sont conduites présentant aux sujets des images stéréoscopiques fixes et animées avec différents niveaux d'asymétrie. La performance est évaluée en comparant des prédictions objectives avec des notes subjectives pour différents niveaux d'asymétrie qui pourraient provoquer un inconfort visuel. / The minimum requirement for any 3D (stereoscopic images) system is to guarantee visual comfort of viewers. Visual comfort is one of the three primary perceptual attributes of 3D QoE, which can be linked directly with technical parameters of a 3D system. Therefore, the goal of this thesis is to characterize objectively the impact of these parameters on human perception for stereoscopic quality monitoring. The first part of the thesis investigates whether visual attention of the viewers should be considered when designing an objective 3D quality metrics. First, the visual attention in 2D and 3D is compared using simple test patterns. The conclusions of this first experiment are validated using complex stimuli with crossed and uncrossed disparities. In addition, we explore the impact of visual discomfort caused by excessive disparities on visual attention. The second part of the thesis is dedicated to the design of an objective model of 3D video QoE, which is based on human perceptual thresholds and acceptability level. Additionally we explore the possibility to use the proposed model as a new subjective scale. For the validation of proposed model, subjective experiments with fully controlled still and moving stereoscopic images with different types of view asymmetries are conducted. The performance is evaluated by comparing objective predictions with subjective scores for various levels of view discrepancies which might provoke visual discomfort.

Advances in unsupervised learning and deep neural networks have led to increased performance in a number of domains, and to the ability to draw strong comparisons between the biological method of self-organization conducted by the brain and computational mechanisms. This thesis aims to use real-world data to tackle two areas in the domain of computer vision which have biological equivalents: feature detection and motion tracking. The aforementioned advances have allowed efficient learning of feature representations directly from large sets of unlabeled data instead of using traditional handcrafted features. The first part of this thesis evaluates such representations by comparing regularization and preprocessing methods which incorporate local neighbouring information during training on a single-layer neural network. The networks are trained and tested on the Hollywood2 video dataset, as well as the static CIFAR-10, STL-10, COIL-100, and MNIST image datasets. The induction of topography or simple image blurring via Gaussian filters during training produces better discriminative features as evidenced by the consistent and notable increase in classification results that they produce. In the visual domain, invariant features are desirable such that objects can be classified despite transformations. It is found that most of the compared methods produce more invariant features, however, classification accuracy does not correlate to invariance. The second, and paramount, contribution of this thesis is a biologically-inspired model to explain the emergence of motion tracking behaviour in early development using unsupervised learning. The model’s self-organization is biased by an original concept called retinal constancy, which measures how similar visual contents are between successive frames. In the proposed two-layer deep network, when exposed to real-world video, the first layer learns to encode visual motion, and the second layer learns to relate that motion to gaze movements, which it perceives and creates through bi-directional nodes. This is unique because it uses general machine learning algorithms, and their inherent generative properties, to learn from real-world data. It also implements a biological theory and learns in a fully unsupervised manner. An analysis of its parameters and limitations is conducted, and its tracking performance is evaluated. Results show that this model is able to successfully follow targets in real-world video, despite being trained without supervision on real-world video.

V dnešní době je kladen vysoký nárok na účinnost elektrických zařízení a to jak ze strany provozovatele, tak i legislativy. Nejlepší účinnosti dosahují synchronní motory s permanentními magnety umístěnými na povrchu rotoru (SMPM), se kterými lze i u malých motorů dosáhnout účinnosti nad 90%. Nicméně tyto motory jsou z důvodů použití magnetů ze vzácných zemin, např. NdFeB, drahé a jsou schopny provozu pouze s frekvenčním měničem. Z cenových důvodů jsou hledány levnější alternativy k SMPM motorům. Jedním z typů motorů, kterým lze SMPM nahradit je synchronní reluktanční motor s permanentními magnety (PMASR). Tento motor je cenově výhodnější, protože používá menší množství magnetů, při zachování podobných, mnohdy i lepších vlastností, nicméně neodpadá potřeba použití frekvenčního měniče. Navíc je zde možnost použití levnějších feritových magnetů a tím ještě výrazněji snížit cenu motoru. V této práci bude PMASR topologie popsána důkladněji včetně elektromagnetického návrhu metodou konečných prvků. Bude provedena i mechanická analýza zvoleného optimálního modelu. Výsledky dosažené metodou konečných prvků budou následně porovnány s analytickým modelem. Z navrženého modelu bude vyroben prototyp a naměřené výsledky budou porovnány s výpočty.

Cette thèse est motivée par le diagnostic et l’évaluation des maladies neuro-dégénératives et dans le but de diagnostique sur la base de l’attention visuelle.Néanmoins, le dépistage à grande échelle de la population n’est possible que si des modèles de prédiction automatique suffisamment robustes peuvent être construits. Dans ce contexte nous nous intéressons `a la conception et le développement des modèles de prédiction automatique pour un contenu visuel spécifique à utiliser dans l’expérience psycho-visuelle impliquant des patients atteints des maladies neuro-dégénératives. La difficulté d’une telle prédiction réside dans une très faible quantité de données d’entraînement. Les modèles de saillance visuelle ne peuvent pas être fondés sur les caractérisitiques “bottom-up” uniquement, comme le suggère la théorie de l’intégration des caractéristiques. La composante “top-down” de l’attention visuelle humaine devient prépondérante au fur et à mesure d’observation de la scène visuelle. L’attention visuelle peut-être prédite en se basant sur les scènes déjà observées. Les réseaux de convolution profonds (CNN) se sont révèlés être un outil puissant pour prédire les zones saillantes dans les images statiques.Dans le but de construire un modèle de prédiction automatique pour les zones saillantes dans les vidéos naturels et intentionnellement dégradées, nous avons conçu une architecture spécifique de CNN profond. Pour surmonter le manque de données d’apprentissage,nous avons conçu un système d’apprentissage par transfert dérivé de la méthode de Bengio.Nous mesurons ses performances lors de la prédiction de régions saillantes. Les r´esultatsobtenus sont int´eressants concernant la r´eaction des sujets t´emoins normaux contre leszones d´egrad´ees dans les vid´eos. La comparaison de la carte de saillance pr´edite des vid´eosintentionnellement d´egrad´ees avec des cartes de densit´e de fixation du regard et d’autresmod`eles de r´ef´erence montre l’int´erˆet du mod`ele d´evelopp´e. / This thesis is motivated by the diagnosis and the evaluation of the dementia diseasesand with the aim of predicting if a new recorded gaze presents a complaint of thesediseases. Nevertheless, large-scale population screening is only possible if robust predictionmodels can be constructed. In this context, we are interested in the design and thedevelopment of automatic prediction models for specific visual content to be used in thepsycho-visual experience involving patients with dementia (PwD). The difficulty of sucha prediction lies in a very small amount of training data.Visual saliency models cannot be founded only on bottom-up features, as suggested byfeature integration theory. The top-down component of human visual attention becomesprevalent as human observers explore the visual scene. Visual saliency can be predictedon the basis of seen data. Deep Convolutional Neural Networks (CNN) have proven tobe a powerful tool for prediction of salient areas in static images. In order to constructan automatic prediction model for the salient areas in natural and intentionally degradedvideos, we have designed a specific CNN architecture. To overcome the lack of learningdata we designed a transfer learning scheme derived from bengio’s method. We measureits performances when predicting salient regions. The obtained results are interestingregarding the reaction of normal control subjects against degraded areas in videos. Thepredicted saliency map of intentionally degraded videos gives an interesting results comparedto gaze fixation density maps and other reference models.

Since the turn of the millennium, a large number of computational models of visual salience have been put forward. How best to evaluate a given model's ability to predict where human observers fixate in images of real-world scenes remains an open research question. Assessing the role of spatial biases is a challenging issue; this is particularly true when we consider the tendency for high-salience items to appear in the image center, combined with a tendency to look straight ahead (“central bias”). This problem is further exacerbated in the context of model comparisons, because some—but not all—models implicitly or explicitly incorporate a center preference to improve performance. To address this and other issues, we propose to combine a-priori parcellation of scenes with generalized linear mixed models (GLMM), building upon previous work. With this method, we can explicitly model the central bias of fixation by including a central-bias predictor in the GLMM. A second predictor captures how well the saliency model predicts human fixations, above and beyond the central bias. By-subject and by-item random effects account for individual differences and differences across scene items, respectively. Moreover, we can directly assess whether a given saliency model performs significantly better than others. In this article, we describe the data processing steps required by our analysis approach. In addition, we demonstrate the GLMM analyses by evaluating the performance of different saliency models on a new eye-tracking corpus. To facilitate the application of our method, we make the open-source Python toolbox “GridFix” available.

The Synchronous Reluctance Motor (SynRM) has been studied. A suitable machine vector modelhas been derived. The influence of the major parameters on the motor performance has beentheoretically determined.Due to the complex rotor geometry in the SynRM, a suitable and simple combined theoretical(analytical) and finite element method has been developed to overcome the high number ofinvolved parameters by identifying some classified, meaningful, macroscopic parameters.Reducing the number of parameters effectively was one of the main goals. For this purpose,attempt has been made to find and classify different parameters and variables, based on availableliteratures and studies. Thus a literature study has been conducted to find all useful ideas andconcepts regarding the SynRM. The findings have been used to develop a simple, general, finiteelement aided and fast rotor design procedure. By this method rotor design can be suitablyachieved by related and simplified finite element sensitivity analysis.The procedure have been tested and confirmed. Then it is used to optimize a special rotor for aparticular induction machine (IM) stator. This optimization is mainly focused on the torquemaximization for a certain current. Torque ripple is also minimized to a practically acceptablevalue. The procedure can also be used to optimize the rotor geometry by considering the othermachine performance parameters as constrains.Finally full geometrical parameter sensitivity analysis is also done to investigate the influence ofthe main involved design parameters on the machine performance.Some main characteristics like magnetization inductances, power factor, efficiency, overloadcapacity, iron losses, torque and torque ripple are calculated for the final designs and in differentmachine load conditions.Effects of ribs, air gap length and number of barriers have been investigated by means of suitableFEM based method sensitivity analysis.

Cette thèse est consacré à l’évaluation et l’amélioration de la performance de la machine synchrone à réluctance variable pour des applications à vitesse variable en général et pour les applications automobiles en particulier. Les deux axes de développement sont la conception de la machine et l’injection des harmoniques de courants de phase. Le rotor est un élément important dans la conception de la machine, et un intérêt particulier est dédié à la conception et l’évaluation du rotor pour améliorer la performance de la machine. Une méthode analytique est proposée dans la thèse pour concevoir le rotor. Plusieurs éléments tels que les ponts qui maintiennent le rotor mécaniquement résistant, ainsi que le rapport d’isolation d’axe q (rapport air-acier) sont étudiés. Une étude de conception assistée par ordinateur basé sur un problème d’optimisation paramétrique est présentée aussi. Les trois familles des algorithmes d’optimisation sont évaluées pour la procédure d’optimisation: un algorithme à base de gradient (algorithme de Newton Quasi), un algorithme non-évolutionnaire sur la base de non-gradient (Nelder Mead Simplex) et un algorithme évolutif sur la base non-gradient (algorithme génétique). Les designs de machines basées sur la procédure analytique et la procédure d’optimisation sont testés sur un banc d’essai. Le deuxième axe d’études de la thèse est l’injection d’harmoniques dans les courants de phase de la machine à réluctance variable synchrone. L’interaction des harmoniques de courant avec les harmoniques spatiales des inductances est étudiée et formalisée pour une machine à m-phases. Ensuite, le concept d’injection d’harmoniques est évalué dans le cas particulier d’une machine à deux phases. Cette étude montre l’avantage de l’injection d’harmoniques dans la réduction de l’ondulation de couple de la machine. Un design d’une machine est finalement développé pour une application automobile sur la base de l’optimisation paramétrique du stator et du rotor. Cette conception est évaluée pour les spécifications imposées électromagnétiques par une application de traction à puissance moyenne / This thesis is dedicated to the evaluation and the improvement of the synchronous reluctance machine’s performance for variable speed drive applications in general and for automotive applications in particular. The two axes of development are machine design and phase current harmonics injection. The rotor is an important element in the machine design and particular emphasis is placed to the design and evaluation of the rotor for enhancing the machine performance. An analytical procedure is proposed for the rotor design. The rotor elements like the ribs and the bridges that maintain the rotor mechanically strong as well as the q-axis insulation ratio (air-to-steel ratio) are studied. A computer-aided design study based on a parametric optimization problem is presented as well. The main three families of the optimization algorithms are evaluated for the optimization procedure: a gradient-based algorithm (Quasi Newton Algorithm), a non-gradient based non-evolutionary algorithm (Nelder Mead Simplex) and a non-gradient based evolutionary algorithm (Genetic Algorithm). The machine designs based on the analytical procedure and the optimization procedure are both manufactured and tested on a bench. The second axis of study of the thesis is the injection of harmonics in the phase currents of the synchronous reluctance machine. The interaction of current harmonics with the spatial inductance harmonics is studied and formalized for an m-phase machine. Then, the harmonics injection concept is evaluated in the particular case of a 2-phase machine. This study shows the benefi t of harmonics injection in the reduction of the machine torque ripple. A synchronous reluctance machine design is fi nally developed for an automotive application based on parametric optimization of the stator and rotor. This design is evaluated for the electromagnetic specifi cations imposed by a mid-power electric vehicle traction application