The thesis argues that there is a pressure on the authorship concept since the emergence of collections of facts, anthologies, and adaptations of pre-existing works. These works were the reason that Judges offered various interpretations to authorship and originality, as some Judges lessened the requirement of originality to obtain copyrightability for these works and some raised it. This led to make the protection granted by copyright law to intellectual works vague and uncertain. This became apparent in conflicts in courts decisions on copyright subsistence in works. This subsequently led to confusion around the criteria of interpretation that should be adopted and the theory or justification that copyright law is founded upon. The thesis argues that this vagueness and uncertainty is related not to the authorship concept but to the failure of law to adapt to two separate natures of works, one including authorial, mental and personal contribution and the other only including manually skilful contribution. Those two kinds cannot be subject to same principles or justifications of protection. The inexistence of such differentiation in doctrine, judiciary and legislation led to the distortion of authorship and originality concepts in the attempts to reduce their interpretation to suit those works that actually miss authorial contribution. Alternatively, whole attention was paid to granting ownership to right holders of these works, which led to the prevalence of the ownership concept as being a necessity for the marketability of cultural works over the authorship concept. The thesis finds that this difference in nature can be uncovered by settling on a differentiation between two kinds of skills that are used in creating works: the mental skills, which are authorial skills, on the one side, and manual skills, which are the collecting, combining, performing or executing skills, on the other. Accordingly, this thesis proposes a categorisation of works, that of ‘high, low and non-authorship’ works, which relies on the nature of the works and elements of authorship in the work. The thesis finds that every category of works needs a separate criterion that can suit its nature and constituent authorship elements; also, the protection needs to be graded depending on the level of authorship in the work. This thesis suggests that such a legal proposition be implemented digitally in what it calls the ‘Digital Cultural National Gate’, which decides the category the work should belong to and the correspondent protection, and that through some questionnaires on the work the authorship elements can be recognised.

Understanding the molecular structure of DNA was considered as greatest achievement in modern biology. It helped in understanding fundamental cellular processes such as replication of DNA, nature of the genetic code and transcription. It also led to technological advancements such as DNA sequencing, genetic engineering and gene cloning. The DNA molecule is highly polymorphic in nature and its structure is dependent on environment, base composition and sequence context. B-DNA, A-DNA, Z-DNA and curved or kinked DNA are some of the well characterized double helical polymorphs. B-DNA is the most prevalent structure in vivo and it can undergo small local variations and global variations. In this thesis we refer to distinct structural property of any particular DNA sequence as deviation from fibre model B-DNA structural parameters or random sequence DNA. Structural properties of DNA are an outcome of the linear arrangement of the 4 chemically different nucleotide bases and the characteristic features of the two grooves (minor and major) arising due to the asymmetric position of glycosidic bonds of base pairs. DNA structure and properties are expected to vary along its length. Several structural features have been defined for DNA duplex, while DNA stability, bendability and intrinsic curvature are well studied and found to be biologically relevant. These three sequence dependent properties differ in their nature and information content and can be studied both at local and global levels, depending on the length of DNA fragment being examined. Majority of the work in this thesis focuses on the analysis of these three DNA structural features in promoter regions of different eukaryotic systems and their relationship with gene expression. The thesis work is divided in to five sections briefly described below. The sections discuss prevalence of the three structural features, DNA stability, bendability and intrinsic curvature in the promoter regions of six eukaryotic systems namely S. cerevisiae, D. melanogaster, C. elegans, zebrafish, mouse and human. The relationship between DNA structural features of promoter regions of S. cerevisiae with gene expression variability is discussed, followed by application of the structure-based promoter prediction algorithm ‘PromPredict’ in annotating promoter regions of six different eukaryotes. Finally, an analysis of structural features of the flanking sequences of transcription factor binding sites (TFBSs) of six transcription factors and their relationship with the DNA binding affinity is discussed. Each of the projects described below will appear as a separate chapters in the thesis. An overview of the eukaryotic transcription machinery, promoter elements and different DNA structural properties are discussed in the introduction of the thesis (chapter 1). The structural properties of DNA in the promoter regions of eukaryotic genes (chapter 2)Earlier studies in the lab reported that, apart from sequence motifs, promoter re- gions have distinct structural properties, such as lower stability, lesser bendability and more curvature compared to other genomic regions. But those studies were on small datasets and few model systems. Advancement in high-throughput tech- niques has made availability of transcription start site information for many model systems. This work was initiated with the aim of investigating the structural fea- tures in different eukaryotic systems belonging to different domains of life. The quantitative analysis of three different structural features of promoter regions of six different model systems S. cerevisiae, C. elegans, D. melanogaster, zebrafish, mouse and human has been carried out. Further, the composition of different k-mers (k=3, 4 and 6) A-tracts and G-quadruplexes has been studied. The analysis allowed us to understand the similarities and differences in struc- tural features of promoter sequences in different model systems. The core promoter sequences of S. cerevisiae, C. elegans, D. melanogaster, zebra fish, mouse and hu- man have been observed to be less stable and have lower preference for nucleosome formation. S. cerevisiae, C. elegans and D. melanogaster promoter sequences have been shown to be less bendable whereas zebrafish, mouse and human promoter se- quences are flexible in terms of bendability towards major groove as predicted fDNase 1 sensitivity model. S. cerevisiae, C. elegans, D. melanogaster core promoter regions have AT rich oligomers, whereas mouse and human core promoter regions have GC rich oligomers and G-quadruplex motifs. DNA structural features of TATA-containing andTATA-less promoters (chapter 3)Eukaryotic genes can be broadly classified as TATA-containing and TATA-less based on the presence of TATA-box in their promoter sequences. Experiments on both classes of genes have reported that, they have differences in regulation of gene ex- pression and cellular functions. In this chapter, the differences in compositional and structural features of TATA-containing and TATA-less promoters in the above mentioned model systems are discussed. The results suggested that DNA structural features of TATA-containing and TATA-less promoters are distinctly different in all eukaryotes. The TATA-containing promoters are less stable, more flexible and more curved compared to TATA-less promoters in lower eukaryotes. In mouse and hu- man genes, DNA duplex stability and G-quadruplex motifs are very distinguishing features in the two classes of promoters. DNA structural properties of eukaryotic promoter regions and gene expression variability (chapter 4) Gene expression is regulated by various external (environment and evolution) and internal (genetic) factors. Presence of sequence motifs, such as TFBSs and TATA- box, as well as DNA methylation has been implicated in the regulation of expression of some genes in vertebrates, but a large number of genes lack these sequences. Ear- lier analyses (described in previous sections) in S. cerevisiae, have shown that their promoter sequences have special structural properties, such as low stability, less bendability and more curvature compared to other genomic regions. These strutural features may play a role in transcription initiation and regulation of gene expression. This project was carried out to understand 1. What is the relationship between DNA structural features and gene expres- sion? 2. What is the relationship between gene expression and bidirectionality of a pro- moter region? For this purpose, the information of seven different gene expression variability measures, stochastic noise, responsiveness, stress response, trans variability, mu- tational variance, interstrain variation and expression divergence have been com- pared with structural features in the promoter regions. It is observed that a few of the variability measures of gene expression are linked to DNA structural prop- erties, along with nucleosome occupancy, TATA-box presence and bidirectionality of promoter regions. Interestingly, gene responsiveness is shown to be most, inti- mately correlated with DNA structural features and promoter architecture. The study highlights the importance of sequence dependent structural features in gene regulation. Promoter prediction in eukaryotes using DNA duplex stability (chapter 5) Structural property-based algorithms can discriminate promoter sequences from non-promoter sequences and are far better than sequence motif-based predictors. Compared to other structural features, low stability is found to be the most preva- lent feature in promoter regions. “PromPredict” (in-house algorithm) uses the din- ucleotide free energy values obtained from differential melting stability of DNA du- plexes as a predictor of promoters and has been successfully used earlier to annotate promoter sequences in prokaryotes and rice. Comprehensive analysis of the perfor- mance of PromPredict in S. cerevisiae, D. melanogaster, C. elegans, zebrafish, mouse and human as well as TATA-containing and TATA-less promoter regions of S. cere visiae with TSS data and 48 eukaryotic systems with translation start site (TLS) data revealed that differential stability is a good criterion for promoter prediction. DNA structure in flanking sequences of consensus motifs modulate transcription factor binding (chapter 6) Sequence specific DNA-protein interactions are essential for specific expression pat- terns during the development. There are several factors contribute to DNA-binding specificities of transcription factors (TFs). They include structure and flexibility of TFs, cofactors, chromatin environment and DNA sequence. Along with actual tran- scription factor binding sites (TFBSs), their sequence context (flanking sequences) is also shown to play a major role in gene regulation. Most of the studies have ad- dressed the sequence context at global level but very little is understood about the role of sequences flanking TFBSs in binding of transcription factors. This project was initiated with the aim of understanding the effect of flanking sequences of TFBSs in transcription factor binding affinity. In vitro DNA binding information of six different transcription factors (with three types of DNA bind- ing domains, Zinc finger (GATA4), home domain (AbdA, AbdB and Ubx) and bZIP (fos-jun and Nfil3)) was provided by Aseem Ansari’s lab. The compositional and structural features (minor groove width, propeller twist, wedge and free energy) are compared with the DNA binding profiles of 12mers (or 8mers) of six different transcription factors. It has been observed that some of the DNA structural proper- ties of flanking sequences are strongly correlated with binding affinity. For GATA4 sequences, binding affinity is negatively correlated to GC content or minor groove width at their 5′ -flanking region, showing the significance of narrow minor groove at 5′ -region. On the other hand, the binding affinity of bZIP proteins is negatively correlated to wedge angles, whereas in case of homeodomain proteins, it is posi- tively correlated to propeller twist and GC content. Thus, this study highlights the differential preference for flanking sequences outside the core binding motifs of six different TFs, which interact with DNA through α-helix. ‘The relationship between transcription pre-initiation complexes and gene ex- pression variability in S. cerevisiae’ is briefly described in the appendix section of the thesis. General conclusion Overall, the results presented in this thesis indicate that DNA sequence based structural features are unique to promoter regions and play an important role in gene regulation. Local structural features of flanking sequences of transcription factor binding sites are also instrumental in determining the DNA binding affinity of transcription factors.

This thesis is called "Teacher personality in a Classic School and Alternative School". The thesis is based on a theoretical-empirical approach. The aim of the thesis is to depict differences in personalities of teachers at regular school and alternative school. Theoretical part concerns about characteristics of alternative schools and their distinctions in comparison with regular schools. Teacher's personality, its structure and basic terminology are also concerns of a theoretical part, including teacher typology and their typical features. Separate part of the thesis is about teacher competences and qualities of a good teacher. At the end of the theoretical part are described requirements on a teacher of regular and alternative school. Practical part of the thesis was realized by questionnaire research. The goal of the thesis and research assumptions are there. The research is supplemented by evaluation of those assumptions. It contains tables and graphical illustrations. The summary of the thesis sums up acquired findings about teachers personality at regular and alternative school.

Submitted by Ulisses Nehmi (ulisses@nehmi.com.br) on 2017-12-22T17:42:27Z No. of bitstreams: 1 Dissert_MPE_Ulisses_2017.12.pdf: 2808696 bytes, checksum: fe4dbd86f710e7d09cbb1705853aba42 (MD5) / Approved for entry into archive by Thais Oliveira (thais.oliveira@fgv.br) on 2017-12-22T17:51:45Z (GMT) No. of bitstreams: 1 Dissert_MPE_Ulisses_2017.12.pdf: 2808696 bytes, checksum: fe4dbd86f710e7d09cbb1705853aba42 (MD5) / Made available in DSpace on 2017-12-23T15:56:54Z (GMT). No. of bitstreams: 1 Dissert_MPE_Ulisses_2017.12.pdf: 2808696 bytes, checksum: fe4dbd86f710e7d09cbb1705853aba42 (MD5) Previous issue date: 2017-12-15 / Muitos estudos sobre a Estrutura a Termo das Taxas de Juros (ETTJ) focam na análise de um único país, geralmente uma economia desenvolvida. São raros os estudos que avaliam as características das curvas de juros para um conjunto de países desenvolvidos, e ainda mais raros os estudos que avaliam essas características para países emergentes. Este estudo parametrizou a ETTJ de 19 economias por um período de 10 anos, divididas entre economias desenvolvidas e emergentes, identificando as principais características que definem cada grupo, algumas das quais se revelaram contraintuitivas. A parametrização das curvas de juros também foi utilizada para remover o ruído dos dados originais, o que permitiu uma análise mais precisa dos fatores que explicam suas variâncias. Com isso, foram encontradas evidências de diferenças relevantes no peso dos fatores nível, inclinação e curvatura na explicação das variações na ETTJ para os países desenvolvidos em relação aos países emergentes. / Many studies on Term Structure of Interest Rates (TSIR) focus on the analysis of a single country, usually a developed economy. Seldom do studies evaluate the features of yield curves for a set of developed countries, and even more rarely do studies evaluate these features for emerging countries. The present study evaluates the parametric TSIR of 19 economies over a period of 10 years, grouped into two distinct sets: developed and emerging economies. It identifies the main features, some of which have proved counterintuitive, that define each group. The parameterization of the yield curves was also used to removed noise from the original data, which allowed for a more accurate analysis of the factors that explain its variances. Evidence of relevant differences in weights for the level, slope and curvature factors were found, which explain the variations in the TSIR of developed countries relative to emerging countries.

Submitted by Eduardo Correa da Silva (correa@correaporto.com.br) on 2018-01-10T20:32:56Z No. of bitstreams: 1 DISSERTAÇÃO - Eduardo Correa da Silva - Após defesa vf.docx: 212519 bytes, checksum: efefa82610170f5811adfccb4aee12d5 (MD5) / Rejected by Thais Oliveira (thais.oliveira@fgv.br), reason: Prezado Eduardo, boa noite. Para que possamos aprovar seu trabalho, é necessário que faça as seguintes alterações: - Na capa, Getulio não tem acento. - Na contracapa, não deve conter o nome da Escola. - O trabalho deve ser postado em PDF. Qualquer dúvida, entre em contato. Att, Thais Oliveira. on 2018-01-10T22:28:03Z (GMT) / Submitted by Eduardo Correa da Silva (correa@correaporto.com.br) on 2018-01-11T00:28:49Z No. of bitstreams: 1 DISSERTAÇÃO - Eduardo Correa da Silva - Após defesa vf.pdf: 1277614 bytes, checksum: 3f48bd24e0b472358e447759da860885 (MD5) / Approved for entry into archive by Thais Oliveira (thais.oliveira@fgv.br) on 2018-01-12T22:49:36Z (GMT) No. of bitstreams: 1 DISSERTAÇÃO - Eduardo Correa da Silva - Após defesa vf.pdf: 1277614 bytes, checksum: 3f48bd24e0b472358e447759da860885 (MD5) / Made available in DSpace on 2018-01-15T12:31:09Z (GMT). No. of bitstreams: 1 DISSERTAÇÃO - Eduardo Correa da Silva - Após defesa vf.pdf: 1277614 bytes, checksum: 3f48bd24e0b472358e447759da860885 (MD5) Previous issue date: 2017-12-04 / Com o intuito de garantir o respeito aos precedentes judiciais positivado pelo Código de Processo Civil/15, bem como visando conferir observância às garantias constitucionais da razoável duração do processo e da isonomia ente os contribuintes, investiga-se a possibilidade e a necessidade de reinterpretação e flexibilização da aplicação do artigo 170-A do Código Tribunal Nacional, a fim de se reconhecer satisfeita a condição para realização de compensação de tributo objeto de controvérsia judicial (“trânsito em julgado”), quando houver a certificação do trânsito em julgado do acórdão paradigma proferido em recurso extraordinário ou especial repetitivos, favorável aos contribuintes e, assim, autorizar o início imediato dos procedimentos de compensação aos contribuintes que estiverem litigando a mesma questão fático-jurídica. / In order to ensure compliance with the judicial precedents set forth by the Code of Civil Procedure, as well as to ensure compliance with the constitutional guarantees of the reasonable duration of the litigation process and isonomy, it is reviewed the possibility and necessity of reinterpretation and flexibility of the application of Article 170- A from National Tax Code, in order to recognize that the condition for the realization of tax compensation subject to judicial litigation is met when the final decision is issued in an extraordinary or special repetitive appeal, favorable to taxpayers and thus authorize the immediate application of the compensation procedures for taxpayers who are litigating the same legal and factual question.

Additive manufacturing (AM) is the name given to the technology of building 3D parts by adding layer-by-layer of materials, including metals, plastics, concrete, etc. Of the different types of AM techniques, electron beam melting (EBM), as a powder bed fusion technology, has been used in this study. EBM is used to build parts by melting metallic powders by using a highly intense electron beam as the energy source. Compared to a conventional process, EBM offers enhanced efficiency for the production of customized and specific parts in aerospace, space, and medical fields. In addition, the EBM process is used to produce complex parts for which other technologies would be either expensive or difficult to apply. This thesis has been divided into three sections, starting from a wider window and proceeding to a smaller one. The first section reveals how the position-related parameters (distance between samples, height from build plate, and sample location on build plate) can affect the microstructural characteristics. It has been found that the gap between the samples and the height from the build plate can have significant effects on the defect content and niobium-rich phase fraction. In the second section, through a deeper investigation, the behavior of Alloy 718 during the EBM process as a function of different geometry-related parameters is examined by building single tracks adjacent to each other (track-by-track) andsingle-wall samples (single tracks on top of each other). In this section, the main focus is to understand the effect of successive thermal cycling on microstructural evolution. In the final section, the correlations between the main machine-related parameters (scanning speed, beam current, and focus offset) and the geometrical (melt pool width, track height, re-melted depth, and contact angle) and microstructural (grain structure, niobium-rich phase fraction, and primary dendrite arm spacing) characteristics of a single track of Alloy 718 have been investigated. It has been found that the most influential machine-related parameters are scanning speed and beam current, which have significant effects on the geometry and the microstructure of the single-melted tracks.

Le but de la thèse est de déterminer l’historique de chargement d’une pièce rompue ou fissurée en service, par analyse des surfaces de fissuration. Pour les analyses de défaillances, les enjeux sont de définir le rôle du chargement dans la défaillance et de déterminer la fraction de durée de vie atteinte lors de la détection de la fissure. Un enjeu complémentaire est d’enrichir la base de données de chargements en service. L’organe choisi pour cette étude est l’essieu-axe ferroviaire. Ainsi, la sollicitation étudiée est en flexion rotative et les matériaux considérés sont les aciers A4T (25CrMo4) et A1N (C40). Suite à une revue bibliographique, quatre méthodes d’analyse quantitative des surfaces de fissuration ont été ciblées et testées, deux ont été retenues et développées. La première est basée sur la quantification des faciès de rupture (stries de fatigue, cupules, fissures secondaires…). La seconde s’appuie sur l’analyse des contraintes résiduelles (fractographie X) en profondeur pour déterminer l’épaisseur de matériau plastifié sous la surface de rupture. Cette profondeur correspond au sillage plastique induit par la propagation de la fissure et est fonction du chargement recherché. / The aim of the thesis is to determine the loading history of a broken or cracked piece in service, by analysing the cracking surfaces. For failure analyses, the issues are to define the role of the loading in the failure and to determine the fraction of lifespan reached when the crack is detected. An additional issue is to expand the in-service loadings database. The component chosen for this study is the railway axle. Thus, the studied stress is in alternate bending and the considered materials are the steels A4T (25CrMo4) and A1N (C40). Following a bibliographic review, four methods for quantitative analysis of cracking surfaces were targeted and tested, two of which were selected and developed. The first one is based on the quantification of fractographic features (fatigue striations, dimples, secondary cracks...). The second is based on the in-depth analysis of the residual stresses (X-ray fractography) to determine the thickness of plasticized material below the fracture surface. This depth corresponds to the plastic wake induced by the crack propagation and is a function of the sought loading.

La reconnaissance faciale (RF) est un domaine de recherche très actif en raison de ses nombreuses applications dans le domaine de la vision par ordinateur en général et en biométrie en particulier. Cet intérêt est motivé par plusieurs raisons. D’abord, le visage est universel. Ensuite, il est le moyen le plus naturel par les êtres humains de s’identifier les uns des autres. Enfin, le visage en tant que modalité biométrique est présente un caractère non intrusif, ce qui le distingue d’autres modalités biométriques comme l’iris ou l’emprunte digitale. La RF représente aussi des défis scientifiques importants. D’abord parce que tous les visages humains ont des configurations similaires. Ensuite, avec les images faciales 2D que l’on peut acquérir facilement, la variation intra-classe, due à des facteurs comme le changement de poses et de conditions d’éclairage, les variations d’expressions faciales, le vieillissement, est bien plus importante que la variation inter-classe.Avec l’arrivée des systèmes d’acquisition 3D capables de capturer la profondeur d’objets, la reconnaissance faciale 3D (RF 3D) a émergé comme une voie prometteuse pour traiter les deux problèmes non résolus en 2D, à savoir les variations de pose et d’éclairage. En effet, les caméras 3D délivrent généralement les scans 3D de visages avec leurs images de texture alignées. Une solution en RF 3D peut donc tirer parti d’une fusion avisée d’informations de forme en 3D et celles de texture en 2D. En effet, étant donné que les scans 3D de visage offrent à la fois les surfaces faciales pour la modalité 3D pure et les images de texture 2D alignées, le nombre de possibilités de fusion pour optimiser le taux de reconnaissance est donc considérable. L’optimisation de stratégies de fusion pour une meilleure RF 3D est l’objectif principal de nos travaux de recherche menés dans cette thèse.Dans l’état d’art, diverses stratégies de fusion ont été proposées pour la reconnaissance de visages 3D, allant de la fusion précoce "early fusion" opérant au niveau de caractéristiques à la fusion tardive "late fusion" sur les sorties de classifieurs, en passant par de nombreuses stratégies intermédiaires. Pour les stratégies de fusion tardive, nous distinguons encore des combinaisons en parallèle, en cascade ou multi-niveaux. Une exploration exhaustive d’un tel espace étant impossible, il faut donc recourir à des solutions heuristiques qui constituent nos démarches de base dans le cadre des travaux de cette thèse.En plus, en s’inscrivant dans un cadre de systèmes biométriques, les critères d’optimalité des stratégies de fusion restent des questions primordiales. En effet, une stratégie de fusion est dite optimisée si elle est capable d’intégrer et de tirer parti des différentes modalités et, plus largement, des différentes informations extraites lors du processus de reconnaissance quelque soit leur niveau d’abstraction et, par conséquent, de difficulté.Pour surmonter toutes ces difficultés et proposer une solution optimisée, notre démarche s’appuie d’une part sur l’apprentissage qui permet de qualifier sur des données d’entrainement les experts 2D ou 3D, selon des critères de performance comme ERR, et d’autre part l’utilisation de stratégie d’optimisation heuristique comme le recuit simulé qui permet d’optimiser les mélanges des experts à fusionner. [...] / Face recognition (FR) was one of the motivations of computer vision for a long time, but only in recent years reliable automatic face recognition has become a realistic target of biometrics research. This interest is motivated by several reasons. First, the face is one of the most preferable biometrics for person identification and verification related applications, because it is natural, non-intrusive, and socially well accepted. The second reason relates to the challenges encountered in the FR domain, in which all human faces are similar to each other and hence offer low distinctiveness as compared with other biometrics, e.g., fingerprints and irises. Furthermore, when employing facial texture images, intra-class variations due to various factors as illumination and pose changes are usually greater than inter-class ones, preventing 2D face recognition systems from being completely reliable in real conditions.Recent, 3D acquisition systems are capable to capture the shape information of objects. Thus, 3D face recognition (3D FR) has been extensively investigated by the research community to deal with the unsolved issues in 2D face recognition, i.e., illumination and pose changes. Indeed, 3D cameras generally deliver the 3D scans of faces with their aligned texture images. 3D FR can benefit from the fusion of 2D texture and 3D shape information.This Ph.D thesis is dedicated to the optimization of fusion strategies based on three dimensional data. However, there are some problems. Indeed, since the 3D face scans provide both the facial surfaces for the 3D model and 2D texture images, the number of fusion method is high.In the literature, many fusion strategies exist that have been proposed for 3D face recognition. We can roughly classify the fusion strategies into two categories: early fusion and late fusion. Some intermediate strategies such as serial fusion and multi-level fusion have been proposed as well. Meanwhile, the search for an optimal fusion scheme remains extraordinarily complex because the cardinality of the space of possible fusion strategies. It is exponentially proportional to the number of competing features and classifiers. Thus, we require fusion technique to efficiently manage all these features and classifiers that constitute our contribution in this work. In addition, the optimality criteria of fusion strategies remain critical issues. By definition, an optimal fusion strategy is able to integrate and take advantage from different data.To overcome all these difficulties and propose an optimized solution, we adopted the following reflection. [...]

La recherche d’images par contenu (CBIR) est une discipline de l’informatique qui vise à structurer automatiquement les collections d’images selon des critères visuels. Les fonctionnalités proposées couvrent notamment l’accès efficace aux images dans une grande base de données d’images ou l’identification de leur contenu par des outils de détection et de reconnaissance d’objets. Ils ont un impact sur une large gamme de domaines qui manipulent ce genre de données, telles que le multimedia, la culture, la sécurité, la santé, la recherche scientifique, etc.Indexer une image à partir de son contenu visuel nécessite d’abord de produire un résumé visuel de ce contenu pour un usage donné, qui sera l’index de cette image dans la collection. En matière de descripteurs d’images, la littérature est désormais trés riche: plusieurs familles de descripteurs existent, et dans chaque famille de nombreuses approches cohabitent. Bon nombre de descripteurs ne décrivant pas la même information et n’ayant pas les mêmes propriétés d’invariance, il peut être pertinent de les combiner de manière à mieux décrire le contenu de l’image. Cette combinaison peut être mise en oeuvre de différentes manières, selon les descripteurs considérés et le but recherché. Dans cette thése, nous nous concentrons sur la famille des descripteurs locaux, avec pour application la recherche d’images ou d’objets par l’exemple dans une collection d’images. Leurs bonnes propriétés les rendent très populaires pour la recherche, la reconnaissance et la catégorisation d'objets et de scènes. Deux directions de recherche sont étudiées:Combinaison de caractéristiques pour la recherche d’images par l’exemple: Le coeur de la thèse repose sur la proposition d’un modèle pour combiner des descripteurs de bas niveau et génériques afin d’obtenir un descripteur plus riche et adapté à un cas d’utilisation donné tout en conservant la généricité afin d’indexer différents types de contenus visuels. L’application considérée étant la recherche par l’exemple, une autre difficulté majeure est la complexité de la proposition, qui doit correspondre à des temps de récupération réduits, même avec de grands ensembles de données. Pour atteindre ces objectifs, nous proposons une approche basée sur la fusion d'index inversés, ce qui permet de mieux représenter le contenu tout en étant associé à une méthode d’accès efficace.Complémentarité des descripteurs: Nous nous concentrons sur l’évaluation de la complémentarité des descripteurs locaux existant en proposant des critères statistiques d’analyse de leur répartition spatiale dans l'image. Ce travail permet de mettre en évidence une synergie entre certaines de ces techniques lorsqu’elles sont jugées suffisamment complémentaires. Les critères spatiaux sont exploités dans un modèle de prédiction à base de régression linéaire, qui a l'avantage de permettre la sélection de combinaisons de descripteurs optimale pour la base considérée mais surtout pour chaque image de cette base. L'approche est évaluée avec le moteur de recherche multi-index, où il montre sa pertinence et met aussi en lumière le fait que la combinaison optimale de descripteurs peut varier d'une image à l'autre.En outre, nous exploitons les deux propositions précédentes pour traiter le problème de la recherche d'images inter-domaines, correspondant notamment à des vues multi-source et multi-date. Deux applications sont explorées dans cette thèse. La recherche d’images inter-domaines est appliquée aux collections photographiques culturelles numérisées d’un musée, où elle démontre son efficacité pour l’exploration et la valorisation de ces contenus à différents niveaux, depuis leur archivage jusqu’à leur exposition ou ex situ. Ensuite, nous explorons l’application de la localisation basée image entre domaines, où la pose d’une image est estimée à partir d’images géoréférencées, en retrouvant des images géolocalisées visuellement similaires à la requête / Content-Based Image Retrieval (CBIR) is a discipline of Computer Science which aims at automatically structuring image collections according to some visual criteria. The offered functionalities include the efficient access to images in a large database of images, or the identification of their content through object detection and recognition tools. They impact a large range of fields which manipulate this kind of data, such as multimedia, culture, security, health, scientific research, etc.To index an image from its visual content first requires producing a visual summary of this content for a given use, which will be the index of this image in the database. From now on, the literature on image descriptors is very rich; several families of descriptors exist and in each family, a lot of approaches live together. Many descriptors do not describe the same information and do not have the same properties. Therefore it is relevant to combine some of them to better describe the image content. The combination can be implemented differently according to the involved descriptors and to the application. In this thesis, we focus on the family of local descriptors, with application to image and object retrieval by example in a collection of images. Their nice properties make them very popular for retrieval, recognition and categorization of objects and scenes. Two directions of research are investigated:Feature combination applied to query-by-example image retrieval: the core of the thesis rests on the proposal of a model for combining low-level and generic descriptors in order to obtain a descriptor richer and adapted to a given use case while maintaining genericity in order to be able to index different types of visual contents. The considered application being query-by-example, another major difficulty is the complexity of the proposal, which has to meet with reduced retrieval times, even with large datasets. To meet these goals, we propose an approach based on the fusion of inverted indices, which allows to represent the content better while being associated with an efficient access method.Complementarity of the descriptors: We focus on the evaluation of the complementarity of existing local descriptors by proposing statistical criteria of analysis of their spatial distribution. This work allows highlighting a synergy between some of these techniques when judged sufficiently complementary. The spatial criteria are employed within a regression-based prediction model which has the advantage of selecting the suitable feature combinations globally for a dataset but most importantly for each image. The approach is evaluated within the fusion of inverted indices search engine, where it shows its relevance and also highlights that the optimal combination of features may vary from an image to another.Additionally, we exploit the previous two proposals to address the problem of cross-domain image retrieval, where the images are matched across different domains, including multi-source and multi-date contents. Two applications of cross-domain matching are explored. First, cross-domain image retrieval is applied to the digitized cultural photographic collections of a museum, where it demonstrates its effectiveness for the exploration and promotion of these contents at different levels from their archiving up to their exhibition in or ex-situ. Second, we explore the application of cross-domain image localization, where the pose of a landmark is estimated by retrieving visually similar geo-referenced images to the query images

La détection de structures fines et orientées dans une image peut mener à un très large champ d'applications en particulier dans le domaine de l'imagerie médicale, des sciences des matériaux ou de la télédétection. Les ouvertures et fermetures par chemins sont des opérateurs morphologiques utilisant des chemins orientés et flexibles en guise d'éléments structurants. Ils sont utilisés de la même manière que les opérateurs morphologiques utilisant des segments orientés comme éléments structurants mais sont plus efficaces lorsqu'il s'agit de détecter des structures pouvant être localement non rigides. Récemment, une nouvelle implémentation des opérateurs par chemins a été proposée leur permettant d'être appliqués à des images 2D et 3D de manière très efficace. Cependant, cette implémentation est limitée par le fait qu'elle n'est pas robuste au bruit affectant les structures fines. En effet, pour être efficaces, les opérateurs par chemins doivent être suffisamment longs pour pouvoir correspondre à la longueur des structures à détecter et deviennent de ce fait beaucoup plus sensibles au bruit de l'image. La première partie de ces travaux est dédiée à répondre à ce problème en proposant un algorithme robuste permettant de traiter des images 2D et 3D. Nous avons proposé les opérateurs par chemins robustes, utilisant une famille plus grande d'éléments structurants et qui, donnant une longueur L et un paramètre de robustesse G, vont permettre la propagation du chemin à travers des déconnexions plus petites ou égales à G, rendant le paramètre G indépendant de L. Cette simple proposition mènera à une implémentation plus efficace en terme de complexité de calculs et d'utilisation mémoire que l'état de l'art. Les opérateurs développés ont été comparés avec succès avec d'autres méthodes classiques de la détection des structures curvilinéaires de manière qualitative et quantitative. Ces nouveaux opérateurs ont été par la suite intégrés dans une chaîne complète de traitement d'images et de modélisation pour la caractérisation des matériaux composite renforcés avec des fibres de verres. Notre étude nous a ensuite amenés à nous intéresser à des filtres morphologiques récents basés sur la mesure de caractéristiques géodésiques. Ces filtres sont une bonne alternative aux ouvertures par chemins car ils sont très efficaces lorsqu'il s'agit de détecter des structures présentant de fortes tortuosités ce qui est précisément la limitation majeure des ouvertures par chemins. La combinaison de la robustesse locale des ouvertures par chemins robustes et la capacité des filtres par attributs géodésiques à recouvrer les structures tortueuses nous ont permis de proposer un nouvel algorithme, les ouvertures par chemins robustes et sélectives. / The detection of thin and oriented features in an image leads to a large field of applications specifically in medical imaging, material science or remote sensing. Path openings and closings are efficient morphological operators that use flexible oriented paths as structuring elements. They are employed in a similar way to operators with rotated line segments as structuring elements, but are more effective as they can detect linear structures that are not necessarily locally perfectly straight. While their theory has always allowed paths in arbitrary dimensions, de facto implementations were only proposed in 2D. Recently, a new implementation was proposed enabling the computation of efficient d-dimensional path operators. However this implementation is limited in the sense that it is not robust to noise. Indeed, in practical applications, for path operators to be effective, structuring elements must be sufficiently long so that they correspond to the length of the desired features to be detected. Yet, path operators are increasingly sensitive to noise as their length parameter L increases. The first part of this work is dedicated to cope with this limitation. Thus, we will propose an efficient d-dimensional algorithm, the robust path operators, which use a larger family of flexible structuring elements. Given an arbitrary length parameter G, path propagation is allowed if disconnections between two pixels belonging to a path is less or equal to G and so, render it independent of L. This simple assumption leads to a constant memory bookkeeping and results in a low complexity. The developed operators have been compared qualitatively and quantitatively to other efficient methods for the detection of line-like features. As an application, robust path openings have been integrated into a complete chain of image processing for the modelling and the characterization of glass fibers reinforced polymer. Our study has also led us to focus our interest on recent morphological connected filters based on geodesic measurements. These filters are a good alternative to path operators as they are efficient at detecting the so-called "tortuous" shapes in an image which is precisely the main limitation of path operators. Combining the local robustness of the robust path operators with the ability of geodesic attribute-based filters to recover "tortuous" shapes have enabled us to propose another original algorithm, the selective and robust path operators.