Le coeur de cette thèse porte sur la modélisation géométrique et introduit une méthode robuste d'extraction de formes tubulaires à partir de nuages de points. Nous avons choisi de tester nos méthodes dans le contexte applicatif de la foresterie pour mettre en valeur la robustesse de nos algorithmes.Nos méthodes intègrent les normales aux points, il est donc nécessaire de les pré-calculer. Notre premier développement a alors consisté à présenter une méthode rapide d'estimation de normales. Pour ce faire nous avons approximé localement la géométrie du nuage de points en utilisant des "patchs" lisses dont la taille s'adapte à la complexité locale des nuages de points.Nos travaux se sont ensuite concentrés sur l’extraction robuste de formes tubulaires dans des nuages de points occlus, bruités et de densité inhomogène. Nous avons développé une variante de la transformée de Hough que nous avons couplé à une proposition de contours actifs indépendants de leur paramétrisation. Notre méthode a été validée en environnement forestier pour reconstruire des troncs d'arbre afin d'en relever les qualités par comparaison à des méthodes existantes.La reconstruction de troncs d'arbre ouvre d'autres questions dont la segmentation des arbres d'une placette forestière. Nous proposons également une méthode de segmentation pour isoler les différents objets d'un jeu de données.Durant nos travaux nous avons utilisé des approches de modélisation pour répondre à des questions géométriques, et nous les avons appliqué à des problématiques forestières. Il en résulte un pipeline de traitements cohérent qui, bien qu'illustré sur des données forestières, est applicable dans des contextes variés. / The core of this thesis concerns geometric modelling and introduces a fast and robust method for the extraction of tubular shapes from point clouds. We chose to test our method in the difficult applicative context of forestry in order to highlight the robustness of our algorithms.Our methods integrate normal vectors, thus they have to be pre-computed. Our first development consisted in the development of a fast normal estimation method on point cloud. To do so, we locally approximated the point cloud geometry using smooth "patches" of points which size adapts to the local complexity of the point cloud geometry.We then focused our work on the robust extraction of tubular shapes from dense, occluded, noisy point clouds suffering from non-homogeneous sampling density. We developed a variant of the Hough transform and combined this research with a new definition of parametrisation-invariant active contours. We validated our method in complex forest environments with the reconstruction of tree stems to emphasize its advantages and compare it to existing methods.Tree stem reconstruction also opens new perspectives halfway in between forestry and geometry such as the segmentation of trees from a forest plot. Therefore we propose a segmentation approach capable of isolating objects inside a point cloud.During our work we used modelling approaches to answer geometric questions and we applied our methods to forestry problems. Therefore, our studies result in a processing pipeline adapted to forest point cloud analyses, but the general geometric algorithms we propose can also be applied in various contexts.

Les capacités des technologies de télédétection ont augmenté exponentiellement au cours des dernières années : de nouveaux scanners fournissent maintenant une représentation géométrique de leur environnement sous la forme de nuage de points avec une précision jusqu'ici inégalée. Le traitement de nuages de points est donc devenu une discipline à part entière avec ses problématiques propres et de nombreux défis à relever. Le coeur de cette thèse porte sur la modélisation géométrique et introduit une méthode robuste d'extraction de formes tubulaires à partir de nuages de points. Nous avons choisi de tester nos méthodes dans le contexte applicatif difficile de la foresterie pour mettre en valeur la robustesse de nos algorithmes et leur application à des données volumineuses. Nos méthodes intègrent les normales aux points comme information supplémentaire pour atteindre les objectifs de performance nécessaire au traitement de nuages de points volumineux.Cependant, ces normales ne sont généralement pas fournies par les capteurs, il est donc nécessaire de les pré-calculer.Pour préserver la rapidité d'exécution, notre premier développement a donc consisté à présenter une méthode rapide d'estimation de normales. Pour ce faire nous avons approximé localement la géométrie du nuage de points en utilisant des "patchs" lisses dont la taille s'adapte à la complexité locale des nuages de points. Nos travaux se sont ensuite concentrés sur l’extraction robuste de formes tubulaires dans des nuages de points denses, occlus, bruités et de densité inhomogène. Dans cette optique, nous avons développé une variante de la transformée de Hough dont la complexité est réduite grâce aux normales calculées. Nous avons ensuite couplé ces travaux à une proposition de contours actifs indépendants de leur paramétrisation. Cette combinaison assure la cohérence interne des formes reconstruites et s’affranchit ainsi des problèmes liés à l'occlusion, au bruit et aux variations de densité. Notre méthode a été validée en environnement complexe forestier pour reconstruire des troncs d'arbre afin d'en relever les qualités par comparaison à des méthodes existantes. La reconstruction de troncs d'arbre ouvre d'autres questions à mi-chemin entre foresterie et géométrie. La segmentation des arbres d'une placette forestière est l'une d’entre elles. C'est pourquoi nous proposons également une méthode de segmentation conçue pour contourner les défauts des nuages de points forestiers et isoler les différents objets d'un jeu de données. Durant nos travaux nous avons utilisé des approches de modélisation pour répondre à des questions géométriques, et nous les avons appliqué à des problématiques forestières.Il en résulte un pipeline de traitements cohérent qui, bien qu'illustré sur des données forestières, est applicable dans des contextes variés. / Abstract : The potential of remote sensing technologies has recently increased exponentially: new sensors now provide a geometric representation of their environment in the form of point clouds with unrivalled accuracy. Point cloud processing hence became a full discipline, including specific problems and many challenges to face. The core of this thesis concerns geometric modelling and introduces a fast and robust method for the extraction of tubular shapes from point clouds. We hence chose to test our method in the difficult applicative context of forestry in order to highlight the robustness of our algorithms and their application to large data sets. Our methods integrate normal vectors as a supplementary geometric information in order to achieve the performance goal necessary for large point cloud processing. However, remote sensing techniques do not commonly provide normal vectors, thus they have to be computed. Our first development hence consisted in the development of a fast normal estimation method on point cloud in order to reduce the computing time on large point clouds. To do so, we locally approximated the point cloud geometry using smooth ''patches`` of points which size adapts to the local complexity of the point cloud geometry. We then focused our work on the robust extraction of tubular shapes from dense, occluded, noisy point clouds suffering from non-homogeneous sampling density. For this objective, we developed a variant of the Hough transform which complexity is reduced thanks to the computed normal vectors. We then combined this research with a new definition of parametrisation-invariant active contours. This combination ensures the internal coherence of the reconstructed shapes and alleviates issues related to occlusion, noise and variation of sampling density. We validated our method in complex forest environments with the reconstruction of tree stems to emphasize its advantages and compare it to existing methods. Tree stem reconstruction also opens new perspectives halfway in between forestry and geometry. One of them is the segmentation of trees from a forest plot. Therefore we also propose a segmentation approach designed to overcome the defects of forest point clouds and capable of isolating objects inside a point cloud. During our work we used modelling approaches to answer geometric questions and we applied our methods to forestry problems. Therefore, our studies result in a processing pipeline adapted to forest point cloud analyses, but the general geometric algorithms we propose can also be applied in various contexts.

La définition de formes par ces procédés itératifs génère des structures avec des propriétésspécifiques intéressantes : rugosité, lacunarité. . . . Cependant, les modèles géométriques classiquesne sont pas adaptés à la description de ces formes.Dans le but de développer un modeleur itératif pour concevoir des objets fractals décrits à l’aide duBCIFS, nous avons développé un ensemble d’outils et d’algorithmes génériques qui nous permettentd’évaluer, de caractériser et d’analyser les différentes propriétés géométriques (la localisation, lecalcul de l’enveloppe convexe, de la distance à partir d’un point, etc) de fractals. Nous avons identifiéles propriétés des opérations standards (intersection, union, offset, . . . ) permettant de calculer uneapproximation d’image des fractales et de plus d’optimiser ces algorithmes d’approximation.Dans certains cas, il est possible de construire un CIFS avec l’opérateur de HUTCHINSON généralisédont l’attracteur est suffisamment proche du résultat de l’opération par rapport à la métrique deHausdorff. Nous avons développé un algorithme générique pour calculer ces CIFS pour une précisiondonnée. Nous avons défini la propriété d’auto-similarité de l’opération, qui définie un ensemble detransformations utilisé dans un système itératif résultant.Pour construire un CIFS exact de l’image, si il existe, il faut prouver tous les similitudes nécessairesmanuellement. Nous explicitons également la condition de l’opération, quand le résultat peut êtrereprésenté par un IFS avec un opérateur de HUTCHINSON généralisé. Dans ce cas, il n’est que cettecondition à prouver manuellement / Defining shapes by iteration allows us to generate new structures with specific properties (roughness,lacunarity), which cannot be achieved with classic modelling.For developing an iterative modeller to design fractals described by a BCIFS, we developed a set oftools and algorithms that permits one to evaluate, to characterize and to analyse different geometricproperties (localisation, convex hull, volume, fractal dimension) of fractals. We identified properties ofstandard CAD operations (intersection, union, offset, . . . ) allowing us to approximate them for fractalsand also to optimize these approximation algorithms.In some cases, it is possible to construct a CIFS with generalised HUTCHINSON operator, whoseattractor is close enough to the operation result with respect to the HAUSDORFF metric.We introduceda generic algorithm to compute such CIFS for a given accuracy.We defined the self-similarity propertyof the operation defining a set of transformations, which are used in the output iterative system.In order to construct an exact CIFS of the image, if it exists, we must prove all the necessarysimilarities manually. We explicit also the condition of the operation to be represented by an IFS witha generalised HUTCHINSON operator. In this case, only this condition should be proved manually

The importance of perceived quality within the automotive industry has been rapidly increasing these years. Since judgmentsconcerning perceived quality is a highly subjective process, eye-tracking technology is one of the best approaches to extractcustomers’ subconscious visual activity during interaction with the product. This thesis aims to find an appropriate solution forrepresenting 3D eye-tracking data for further improvements in the validity and verification efficiency of perceived qualityanalysis, attempting to answer the question:How can eye-tracking data be presented and integrated into 3D automobile design workflow as a material that allows designersto understand their customers better?In the study, a prototype system was built for car-interior inspection in the virtual reality (VR) showroom through an explorativeresearch process including investigations in the acquisition of gaze data in VR, classification of eye movement from thecollected gaze data, and the visualizations for the classified eye movements. The prototype system was then evaluated throughcomparisons between algorithms and feedbacks from the engineers who participated in the pilot study.As a result, a method combining I-VT (identification with velocity threshold) and DBSCAN (density-based spatial clusteringof application with noise) was implemented as the optimum algorithm for eye movement classification. A modified heat map,a cluster plot, a convex hull plot, together with textual information, were used to construct the complete visualization of theeye-tracking data. The prototype system has enabled car designers and engineers to examine both the customers’ and their ownvisual behavior in the 3D virtual showroom during a car inspection, followed by the extraction and visualization of the collectedgaze data. This paper presents the research process, including the introduction to relevant theory, the implementation of theprototype system, and its results. Eventually, strengths and weaknesses, as well as the future work in both the prototype solutionitself and potential experimental use cases, are discussed. / Betydelsen av upplevd kvalitet inom bilindustrin har ökat kraftigt dessa år. Eftersom uppfattningar om upplevd kvalitet är en mycket subjektivt är ögonspårningsteknik en av de bästa metoderna för att extrahera kundernas undermedvetna visuella aktivitet under interaktion med produkten. Denna avhandling syftar till att hitta en lämplig lösning för att representera 3Dögonspårningsdata för ytterligare förbättringar av validitets- och verifieringseffektiviteten hos upplevd kvalitetsanalys, och försöker svara på frågan: Hur kan ögonspårningsdata presenteras och integreras i 3D-arbetsflödet för bildesign som ett material som gör det möjligt för designers att bättre förstå sina kunder? I studien byggdes ett prototypsystem för bilinteriörinspektion i showroomet för virtuell verklighet (VR) genom en explorativ forskningsprocess inklusive undersökningar i förvärv av blickdata i VR, klassificering av ögonrörelse från insamlad blicksdata och visualiseringar för de klassificerade ögonrörelserna. Prototypsystemet utvärderades sedan genom jämförelser mellan algoritmer och återkopplingar från ingenjörerna som deltog i pilotstudien. Följaktligen implementerades en metod som kombinerar I-VT (identifiering med hastighetströskel) och DBSCAN (densitetsbaserad spatial gruppering av applikation med brus) som den optimala algoritmen för ögonrörelseklassificering. En modifierad värmekarta, ett klusterdiagram, en konvex skrovdiagram, tillsammans med textinformation, användes för att konstruera den fullständiga visualiseringen av ögonspårningsdata. Prototypsystemet har gjort det möjligt för bilkonstruktörer och ingenjörer att undersöka både kundernas och deras visuella beteende i det virtuella 3D-utställningsrummet under en bilinspektion, följt av utvinning och visualisering av den insamlade blicken. Denna uppsats presenterar forskningsprocessen, inklusive introduktion till relevant teori, implementeringen av prototypsystemet och dess resultat. Så småningom diskuteras styrkor och svagheter, liksom det framtida arbetet i både prototyplösningen och potentiella experimentella användningsfall.

Orientadores: Helder Anibal Hermini, Paulo R. G. Kurka / Dissertação (mestrado) - Universidade Estadual de Campinas, Faculdade de Engenharia Mecanica / Made available in DSpace on 2018-08-10T05:53:08Z (GMT). No. of bitstreams: 1 Mendes_EduardoFelippeAguiar_M.pdf: 4299696 bytes, checksum: d73355f383b74bd0e49b3bc2473e5b6b (MD5) Previous issue date: 2007 / Resumo: Este trabalho visou o estudo de manipuladores à semelhança do membro superior humano. Após o estudo do membro superior humano e dos robôs antropomórficos disponíveis tanto na industria quanto no meio acadêmico, utilizaram-se teorias de modelagem geométrica, cinemática direta e cinemática inversa para realizar o modelo de um manipulador robótico antropomórfico. A partir desse modelo desenvolveu-se um software em LabVIEW de Cinemática Direta e Cinemática Inversa de operação em tempo real. Com a intenção de verificar os acionadores mais apropriados disponíveis atualmente, um estudo de músculos artificiais se seguiu, onde se observou a maior viabilidade do músculo artificial de SMA ativado eletricamente. Um protótipo de junta acionada por músculos artificiais foi desenvolvido e controlado via computador. Como resultado deste trabalho conclui-se que há ainda muito para ser desenvolvido na área de manipuladores antropomórficos, principalmente no que diz respeito aos músculos artificiais / Abstract: This work sought the study of manipulators to the similarity of the human superior member. After the study of human superior member, and of anthropomorphics robotics available in the industries and in the academic middle, it was used theories of geometric modelling, direct kinematics modelling and inverse kinematics modelling to make the model of a anthropomorphic robotic manipulator. With that model it grew a software in LabVIEW of real time Direct Kinematics and Inverse Kinematics. With the intention of verifying the available most appropriate actuators, a study of artificial muscles was proceeded, where the largest viability of the artificial muscle of SMA activated electrically was observed. A joint prototype actuated by artificial muscles was developed and controlled through computer. As a result of this work it is ended that there is still a lot to be developed in the area of anthropomorphic manipulators, mainly in what it concerns the artificial muscles / Mestrado / Mecanica dos Sólidos e Projeto Mecanico / Mestre em Engenharia Mecânica

Researchers conceptually recommend inquiry-based learning as a necessary means to alleviate the problems of learning but this study has embarked on practical implementation of inquiry-based facilitation and learning in Euclidean Geometry. Inquiry-based learning is student-centred. Therefore, the teaching or monitoring of inquiry-based learning in this study is referred to as inquiry-based facilitation. The null hypothesis discarded in this study explains that there is no difference between inquiry-based facilitation and traditional axiomatic approach in teaching Euclidean Geometry, that is, H0: μinquiry-based facilitation = μtraditional axiomatic approach. This study emphasises a pragmatist view that constructivism is fundamental to realism, that is, inductive inquiry supplements deductive inquiry in teaching and learning. Participants in this study comprise schools in Tshwane North district that served as experimental group and Tshwane West district schools classified as comparison group. The two districts are in the Gauteng Province of South Africa. The total number of students who participated is 166, that is, 97 students in the experimental group and 69 students in the comparison group. Convenient sampling applied and three experimental and three comparison group schools were sampled. Embedded mixed-method methodology was employed. Quantitative and qualitative methodologies are integrated in collecting data; analysis and interpretation of data. Inquiry-based-facilitation occurred in experimental group when the facilitator probed asking students to research, weigh evidence, explore, share discoveries, allow students to display authentic knowledge and skills and guiding students to apply knowledge and skills to solve problems for the classroom and for the world out of the classroom. In response to inquiry-based facilitation, students engaged in cooperative learning, exploration, self-centred and self-regulated learning in order to acquire knowledge and skills. In the comparison group, teaching progressed as usual. Quantitative data revealed that on average, participant that received intervention through inquiry-based facilitation acquired inquiry-based learning skills and improved (M= -7.773, SE= 0.7146) than those who did not receive intervention (M= -0.221, SE = 0.4429). This difference (-7.547), 95% CI (-8.08, 5.69), was significant at t (10.88), p = 0.0001, p<0.05 and represented a large effect size of 0.55. The large effect size emphasises that inquiry-based facilitation contributed significantly towards improvement in inquiry-based learning and that the framework contributed by this study can be considered as a framework of inquiry-based facilitation in Euclidean Geometry. This study has shown that the traditional axiomatic approach promotes rote learning; passive, deductive and algorithmic learning that obstructs application of knowledge in problem-solving. Therefore, this study asserts that the application of Inquiry-based facilitation to implement inquiry-based learning promotes deeper, authentic, non-algorithmic, self-regulated learning that enhances problem-solving skills in Euclidean Geometry. / Mathematics Education / Ph. D. (Mathematics, Science and Technology Education)