The study of particle morphology plays an important role in understanding the micromechanical behavior of cohesionless soil. Shear strength and liquefaction characteristics of granular soil depend on various morphological characteristics of soil grains such as their particle size, shape and surface texture. Therefore, accurate characterization and quantification of particle shape is necessary to study the effect of grain shape on mechanical behavior of granular assembly. However, the theoretical and practical developments of quantification of particle morphology and its influence on the mechanical response of granular assemblies has been very limited due to the lack of quantitative information about particle geometries, the experimental and numerical difficulties in characterizing and modeling irregular particle morphology. Motivated by the practical relevance of these challenges, this research presents a comprehensive approach to model irregular particle shape accurately both in two and three dimensions. To facilitate the research goal, a variety of natural and processed sand samples is collected from various locations around the world. A series of experimental and analytical studies are performed following the sample collection effort to characterize and quantify particle shapes of various sand samples by using Fourier shape descriptors. As part of the particle shape quantification and modeling, a methodology is developed to determine an optimum sample size for each sand sample used in the analysis. Recently, Discrete Element Method (DEM) has gained attention to model irregular particle morphology in two and three dimensions. In order to generate and reconstruct particle assemblies of highly irregular geometric shapes of a particular sand sample in the DEM environment, the relationship between grain size and shape is explored and no relationship is found between grain size and shape for the sand samples analyzed. A skeletonization algorithm is developed in this study in order to automate the Overlapping Discrete Element Cluster (ODEC) technique for modeling irregular particle shape in two and three dimensions. Finally, the two-dimensional and three-dimensional particle shapes are implemented within discrete element modeling software, PFC2D and PFC3D, to evaluate the influence of grain shape on shear strength behavior of granular soil by using discrete simulation of direct shear test.

Orchards demand large areas of land, thus they are often situated far from major population centres. As a result it is often difficult to obtain the necessary personnel, limiting both growth and productivity. However, if autonomous robots could be integrated into the operation of the orchard, the manpower demand could be reduced. A key problem for any autonomous robot is localisation; how does the robot know where it is? In agriculture robots, the most common approach is to use GPS positioning. However, in an orchard environment, the dense and tall vegetation restricts the usage to large robots that reach above the surroundings. In order to enable the use of smaller robots, it is instead necessary to use a GPS independent system. However, due to the similarity of the environment and the lack of strong recognisable features, it appears unlikely that typical non-GPS solutions will prove successful. Therefore we present a GPS independent localisation system, specifically aimed for orchards, that utilises the inherent structure of the surroundings. Furthermore, we examine and individually evaluate three related sub-problems. The proposed system utilises a 3D point cloud created from a 2D LIDAR and the robot’s movement. First, we show how the data can be segmented into individual trees using a Hidden Semi-Markov Model. Second, we introduce a set of descriptors for describing the geometric characteristics of the individual trees. Third, we present a robust localisation method based on Hidden Markov Models. Finally, we propose a method for detecting segmentation errors when associating new tree measurements with previously measured trees. Evaluation shows that the proposed segmentation method is accurate and yields very few segmentation errors. Furthermore, the introduced descriptors are determined to be consistent and informative enough to allow localisation. Third, we show that the presented localisation method is robust both to noise and segmentation errors. Finally it is shown that a significant majority of all segmentation errors can be detected without falsely labeling correct segmentations as incorrect. / Eftersom fruktodlingar kräver stora markområden är de ofta belägna långt från större befolkningscentra. Detta gör det svårt att finna tillräckligt med arbetskraft och begränsar expansionsmöjligheterna. Genom att integrera autonoma robotar i drivandet av odlingarna skulle arbetet kunna effektiviseras och behovet av arbetskraft minska. Ett nyckelproblem för alla autonoma robotar är lokalisering; hur vet roboten var den är? I jordbruksrobotar är standardlösningen att använda GPS-positionering. Detta är dock problematiskt i fruktodlingar, då den höga och täta vegetationen begränsar användandet till större robotar som når ovanför omgivningen. För att möjliggöra användandet av mindre robotar är det istället nödvändigt att använda ett GPS-oberoende lokaliseringssystem. Detta problematiseras dock av den likartade omgivningen och bristen på distinkta riktpunkter, varför det framstår som osannolikt att existerande standardlösningar kommer fungera i denna omgivning. Därför presenterar vi ett GPS-oberoende lokaliseringssystem, speciellt riktat mot fruktodlingar, som utnyttjar den naturliga strukturen hos omgivningen.Därutöver undersöker vi och utvärderar tre relaterade delproblem. Det föreslagna systemet använder ett 3D-punktmoln skapat av en 2D-LIDAR och robotens rörelse. Först visas hur en dold semi-markovmodell kan användas för att segmentera datasetet i enskilda träd. Därefter introducerar vi ett antal deskriptorer för att beskriva trädens geometriska form. Vi visar därefter hur detta kan kombineras med en dold markovmodell för att skapa ett robust lokaliseringssystem.Slutligen föreslår vi en metod för att detektera segmenteringsfel när nya mätningar av träd associeras med tidigare uppmätta träd. De föreslagna metoderna utvärderas individuellt och visar på goda resultat. Den föreslagna segmenteringsmetoden visas vara noggrann och ge upphov till få segmenteringsfel. Därutöver visas att de introducerade deskriptorerna är tillräckligt konsistenta och informativa för att möjliggöra lokalisering. Ytterligare visas att den presenterade lokaliseringsmetoden är robust både mot brus och segmenteringsfel. Slutligen visas att en signifikant majoritet av alla segmenteringsfel kan detekteras utan att felaktigt beteckna korrekta segmenteringar som inkorrekta.

The increase in the use of digital video/image has showed the need for modeling and querying the semantic content in them. Using manual annotation techniques for defining the semantic content is both costly in time and have limitations on querying capabilities. So, the need for content based information retrieval in multimedia domain is to extract the semantic content in an automatic way. The semantic content is usually defined with the objects in images/videos. In this thesis, a Genetic Algorithm based object extraction and classification mechanism is proposed for extracting the content of the videos and images. The object extraction is defined as a classification problem and a Genetic Algorithm based classifier is proposed for classification. Candidate objects are extracted from videos/images by using Normalized-cut segmentation and sent to the classifier for classification. Objects are defined with the Best Representative and Discriminative Feature (BRDF) model, where features are MPEG-7 descriptors. The decisions of the classifier are calculated by using these features and BRDF model. The classifier improves itself in time, with the genetic operations of GA. In addition to these, the system supports fuzziness by making multiple categorization and giving fuzzy decisions on the objects. Externally from the base model, a statistical feature importance determination method is proposed to generate BRDF model of the categories automatically. In the thesis, a platform independent application for the proposed system is also implemented.

In this thesis, the performance of four state-of-the-art feature detectors along with SIFT and SURF descriptors in matching object features of mid-wave infrared, long-wave infrared and visual-band images is evaluated across viewpoints and changing distance conditions. The utilized feature detectors are Scale Invariant Feature Transform (SIFT), multiscale Harris-Laplace, multiscale Hessian-Laplace and Speeded Up Robust Features (SURF) detectors, all of which are invariant to image scale and rotation. Features on different blackbodies, human face and vehicle images are extracted and performance of reliable matching is explored between different views of these objects each in their own category. All of these feature detectors provide good matching performance results in infrared-band images compared with visual-band images. The comparison of matching performance for mid-wave and long-wave infrared images is also explored in this study and it is observed that long-wave infrared images provide good matching performance for objects at lower temperatures, whereas mid-wave infrared-band images provide good matching performance for objects at higher temperatures. The matching performance of SURF detector and descriptor for human face images in long-wave infrared-band is found to be outperforming than other detectors and descriptors.

Novel methods of object recognition that form a bridge between today&rsquo / s local feature frameworks and previous decade&rsquo / s strong but deserted geometric invariance field are presented in this dissertation. The rationale behind this effort is to complement the lowered discriminative capacity of local features, by the invariant geometric descriptions. Similar to our predecessors, we first start with constrained cases and then extend the applicability of our methods to more general scenarios. Local features approach, on which our methods are established, is reviewed in three parts / namely, detectors, descriptors and the methods of object recognition that employ them. Next, a novel planar object recognition framework that lifts the requirement for exact appearance-based local feature matching is presented. This method enables matching of groups of features by utilizing both appearance information and group geometric descriptions. An under investigated area, scene logo recognition, is selected for real life application of this method. Finally, we present a novel method for three-dimensional (3D) object recognition, which utilizes well-known local features in a more efficient way without any reliance on partial or global planarity. Geometrically consistent local features, which form the crucial basis for object recognition, are identified using affine 3D geometric invariants. The utilization of 3D geometric invariants replaces the classical 2D affine transform estimation /verification step, and provides the ability to directly verify 3D geometric consistency. The accuracy and robustness of the proposed method in highly cluttered scenes with no prior segmentation or post 3D reconstruction requirements, are presented during the experiments.

This study reports on the quantitative evaluation of a set of state-of-the-art feature detectors and descriptors in the context of repeat photography. Unlike most related work, the proposed study assesses the performance of feature detectors when intra-pair variations are uncontrolled and due to a variety of factors (landscape change, weather conditions, different acquisition sensors). There is no systematic way to model the factors inducing image change. The proposed evaluation is performed in the context of image matching, i.e. in conjunction with a descriptor and matching strategy. Thus, beyond just comparing the performance of these detectors and descriptors, we also examine the feasibility of feature-based matching on repeat photography. Our dataset consists of a set of repeat and historic images pairs that are representative for the database created by the Mountain Legacy Project www.mountainlegacy.ca. / Graduate

In order to meet the needs of people with hearing loss today hearing aids are custom designed. Increasingly accurate 3-D scanning technology has contributed to the transition from conventional production scenarios to software based processes. Nonetheless, there is a tremendous amount of manual work involved to transform an input 3-D surface mesh of the outer ear into a final hearing aid shape. This manual work is often cumbersome and requires lots of experience which is why automatic solutions are of high practical relevance. This work is concerned with the recognition of 3-D surface meshes of ear implants. In particular we present a semantic part-labeling framework which significantly outperforms existing approaches for this task. We make at least three contributions which may also be found useful for other classes of 3-D meshes. Firstly, we validate the discriminative performance of several local descriptors and show that the majority of them performs poorly on our data except for 3-D shape contexts. The reason for this is that many local descriptor schemas are not rich enough to capture subtle variations in form of bends which is typical for organic shapes. Secondly, based on the observation that the left and the right outer ear of an individual look very similar we raised the question how similar the ear shapes among arbitrary individuals are? In this work, we define a notion of distance between ear shapes as building block of a non-parametric shape model of the ear to better handle the anatomical variability in ear implant labeling. Thirdly, we introduce a conditional random field model with a variety of label priors to facilitate the semantic part-labeling of 3-D meshes of ear implants. In particular we introduce the concept of a global parametric transition prior to enforce transition boundaries between adjacent object parts with an a priori known parametric form. In this way we were able to overcome the issue of inadequate geometric cues (e.g., ridges, bumps, concavities) as natural indicators for the presence of part boundaries. The last part of this work offers an outlook to possible extensions of our methods, in particular the development of 3-D descriptors that are fast to compute whilst at the same time rich enough to capture the characteristic differences between objects residing in the same class.

The aerodynamic integration of an aero-engine intake system with the airframe can pose some notable challenges. This is particularly so for many military air- craft and is likely to become a more pressing issue for both new military systems with highly embedded engines as well as for novel civil aircraft configurations. During the late 1960s with the advent of turbo-fan engines, industry became in- creasingly aware of issues which arise due to inlet total pressure distortion. Since then, inlet-engine compatibility assessments have become a key aspect of any new development. In addition to total temperature and total pressure distortions, flow angularity and the associated swirl distortion are also known to be of notable con- cern. The importance of developing a rigorous methodology to understand the effects of swirl distortion on turbo-machinery has also become one of the major concerns of current design programmes. The goal of this doctoral research was to further the current knowledge on swirl distortion, and its adverse effects on engine performance, focusing on the turbo-machinery components (i.e. fans or compressors). This was achieved by looking into appropriate swirl flow descriptors and by correlating them against the compressor performance parameters (e.g loss in stability pressure ratios). To that end, a number of high-fidelity three-dimensional Computational Fluid Dynamics (CFD) models have been developed using two sets of transonic rotors (i.e. NASA Rotor 67 and 37), and a stator (NASA Stator 67B). For the numerical purpose, a boundary condition methodology for the definition of swirl distortion patterns at the inlet has been developed. Various swirl distortion numerical parametric studies have been performed using the modelled rotor configurations. Two types of swirl distortion pattern were investigated in the research, i.e. the pure bulk swirl and the tightly-wound vortex. Numerical simulations suggested that the vortex core location, polarity, size and strength greatly affect the compressor performance. The bulk swirl simula- tions also showed the dependency on swirl strength and polarity. This empha- sized the importance of quantifying these swirl components in the flow distortion descriptors. For this, a methodology have been developed for the inlet-engine compatibility assessment using different types of flow descriptors. A number of correlations have been proposed for the two types of swirl distortion investigated in the study.

L’utilisation de techniques informatiques pour la reconnaissance et l'identification des poissons est devenue assez populaire parmi les chercheurs. Ces nouvelles approches sont importantes, puisque les informations extraites sur les poissons telles que leurs trajectoires, leurs positions ou leurs couleurs, permettent de déterminer si les poissons sont en bonne santé ou en état de stress. Les méthodes existantes ne sont pas assez précises notamment lorsque des éléments tels que les bulles ou des zones éclairées peuvent être identifiées comme étant des poissons. De plus, les taux de reconnaissance et d'identification des systèmes existants peuvent encore être améliorés afin d’obtenir des résultats à la fois meilleurs et plus précis. Afin d’obtenir de meilleurs taux de reconnaissance et d'identification, un système amélioré a été construit en combinant plusieurs méthodes de détection et d’analyse. Tout d'abord, la première étape a consisté à proposer une méthode de suivi d'objets dans le but de localiser en temps réel la position des poissons à partir de vidéos. Celle-ci inclut le suivi automatisé multi-cibles de poissons dans un aquarium. Les performances en termes de détection et d’identification risquaient d’être faibles notamment en raison du processus de suivi dans un environnement temps réel. Une méthode de suivi des poissons plus précise est donc proposée ainsi qu'une méthode complète pour identifier et détecter les modèles de nage des poissons. Dans ces travaux, nous proposons, pour le suivi des poissons, une amélioration de l’algorithme du filtre particulaire en l’associant à un algorithme de détection de mouvement. Un système doté de deux caméras est également proposé afin d'obtenir un meilleur taux de détection. La seconde étape comprend la conception et le développement d'une méthode améliorée pour le recadrage et la segmentation dynamique des images dans un environnement temps réel. Ce procédé est proposé pour extraire de la vidéo les images représentant les poissons en éliminant les éléments provenant de l’arrière-plan. La troisième étape consiste à caractériser les objets (les poissons). La méthode proposée est basée sur des descripteurs utilisant la couleur pour caractériser les poissons. Ces descripteurs sont ensuite utilisés dans la suite des traitements. Dans nos travaux, les descripteurs couleurs généralisés de Fourier (GCFD : Generalized Color Fourier Descriptor) sont utilisés et une adaptation basée sur la détection de l’environnement est proposée afin d’obtenir une identification plus précise des poissons. Une méthode de mise en correspondance basée sur un calcul de distance est utilisée pour comparer les vecteurs de caractéristiques des images segmentées afin de classifier les poissons présents dans la vidéo. Un prototype dont le but est de modéliser les profils de nage des poissons a été développé. Celui-ci intègre toutes les méthodes proposées et a permis d’évaluer la validité de notre approche. Les résultats montrent que les méthodes proposées améliorent la reconnaissance et l’identification en temps réel des poissons. La méthode de suivi proposée montre une amélioration par rapport au procédé basé sur le filtre particulaire classique. Le recadrage dynamique et la méthode de segmentation temps-réel présentent en termes de précision un pourcentage moyen de 84,71%. La méthode de caractérisation des objets développée pour reconnaitre et identifier en temps réel les poissons montre également une amélioration par rapport aux descripteurs couleurs classiques. Le travail réalisé peut trouver une application directe auprès des aquaculteurs afin de suivre en temps réel et de manière automatique le comportement des poissons et éviter ainsi un suivi « visuel » tel qu’il est réalisé actuellement. / Recognition and identification of fish using computational methods have increasingly become a popular research endeavour among researchers. The methods are important as the information displayed by the fish such as trajectory patterns, location and colour could determine whether the fish are healthy or under stress. Current methods are not accurate especially when there exist thresholds such as bubbles and some lighted areas which might be identified as fish. Besides, the recognition and identification rate of the existing systems can still be improved to obtain better and more accurate results. In order to achieve a better recognition and identification rate, an improved scheme consisting of a combination of several methods is constructed. First of all, the first approach is to propose an object tracking method for the purpose of locating the position of fish for real-time videos. This includes the consideration of tracking multiple fish in a single tank in an automated way. The detection and identification rate may be slow due to the on-going tracking process especially in a real-time environment. A more accurate fish tracking method is proposed as well as a systematic method to identify and detect fish swimming patterns. In this research, the particle filter algorithm is enhanced and further combined with the motion detection algorithm for fish tracking. A dual camera system is also proposed to obtain better detection rate. The second approach includes the design and development of an enhanced method for dynamically cropping and segmenting images in real-time environment. This method is proposed to extract each image of the fish from every successive video frame to reduce the tendency of detecting the background as an object. The third approach includes an adapted object characterisation method which utilises colour feature descriptors to represent the fish in a computational form for further processing. In this study, an object characterisation method, GCFD (Generalized Colour Fourier Descriptor) is adapted to suit the environment for more accurate identification of the fish. A feature matching method based on distance matching is used to match the feature vectors of the segmented images for classifying the specific fish in the recorded video. In addition, a real-time prototype system which models the fish swimming pattern incorporating all the proposed methods is developed to evaluate the methods proposed in this study. Based on the results, the proposed methods show improvements which result in a better real-time fish recognition and identification system. The proposed object tracking method shows improvement over the original particle filter method. Based on the average percentage in terms of the accuracy for the dynamic cropping and segmentation method in real time, an acceptable value of 84.71% was recorded. The object characterisation method which is adapted for fish recognition and identification in real time shows an improvement over existing colour feature descriptors. As a whole, the main output of this research could be used by aquaculturist to track and monitor fish in the water computationally in real-time instead of manually.

Made available in DSpace on 2016-06-02T19:48:22Z (GMT). No. of bitstreams: 1 5003.pdf: 2058659 bytes, checksum: 631c9ff761bcd738bfb1237e51455da8 (MD5) Previous issue date: 2013-02-28 / Some activities need to be stimulated in post-cardiac surgery, such as, mobilization, coughing, deep breathing exercises to avoid complications; however, these activities may be hinder by pain. An assessment of the pain at rest and during activities is needed in order to better deal with this occurrences. The main aim of this study was to assess the perception of pain after cardiac surgery via sternotomy during rest and in five selected activities (coughing, turning aside, deep breathing, sitting or standing up from a chair, and walking); the specific objectives were to identify the location and intensity of pain during rest and activities in postoperative cardiac surgery patients in the 1st, 2nd, 3rd and 6th days; to link the pain intensity with the activities and at rest, considering the postoperative days; to link the pain intensity with variables clinical-surgical; to characterize the pain through pain descriptors. A descriptive study of prospective cohort was carried out. A tool to collect socio-demographic and surgicalclinical data, a Multidimensional Scale for Pain Assessment (EMADOR) that consists of a numeric scale for pain intensity assessment, a body diagram to assess the pain location and an escalation of acute pain descriptors, were utilised. A total of 48 patients who undergone a cardiac surgery via sternotomy participated. All patients complained of pain during one of the activities in at least one of the postoperative days. The pain intensity at rest in the postoperative cardiac surgery was assessed to lessen with day in the following postoperative days. However, during the activities the pain level decreased from the 3rd, excepting for the coughing activity which decreased only in the 6th. The decreasing order of strength, when assessed the pain levels of all days, was: coughing, turning aside, deeply breathing and resting. The sternal region was the most frequently cited location of pain, followed by the epigastric region. The variables gender, age, type and duration of surgery showed weak correlation with the pain level. The keywords that best characterised the pain after cardiac surgery via sternotomy were: strong, intense, terrifying, deep and very severe. The high levels of pain may be contributing to a longer recovery period. The patients considered painful the multidimensionality of the phenomenon when using descriptors to characterize the perceived pain. The study allowed a better understanding of the aspects related to pain in the postoperative cardiac surgery. / Algumas atividades precisam ser estimuladas no pós-operatório de cirurgia cardíaca, como a mobilização, a tosse, os exercícios de respiração profunda para se evitar complicações, no entanto, podem ser prejudicadas pela presença da dor. A avaliação da dor em repouso e durante as atividades é necessária para que haja um melhor manejo deste fenômeno. Este estudo teve como objetivo geral avaliar a percepção da dor em repouso e durante cinco atividades esperadas (ao tossir, ao virar-se de lado, à respiração profunda, ao sentar ou levantar da cadeira e ao deambular) no pós-operatório de cirurgia cardíaca por esternotomia mediana; e específicos identificar a intensidade e a localização de dor durante o repouso e as atividades em sujeitos submetidos à cirurgia cardíaca, no 1º, 2º, 3º e 6º dias pós-operatório; realizar associação entre intensidade da dor e as atividades e em repouso, considerando os dias de pós-operatório; realizar associação entre intensidade de dor e variáveis clínicocirúrgicas; caracterizar a dor por meio de descritores de dor. Foi realizado um estudo descritivo, de coorte prospectivo. Foi utilizado um instrumento para coleta de dados sociodemográficos e clínico-cirúrgicos, a Escala Multidimensional para Avaliação da Dor percebida (EMADOR) que consta de uma escala numérica de avaliação da intensidade da dor, um diagrama corporal para avaliar a localização da dor e de um escalonamento de descritores de dor aguda. Participaram 48 sujeitos submetidos à cirurgia cardíaca eletiva por esternotomia. Todos os participantes tiveram queixas de dor ao menos em um dos dias de pósoperatório, em uma das atividades. A dor durante o repouso no pós-operatório de cirurgia cardíaca apresentou-se de intensidade decrescente com o passar dos dias de pós-operatório. No entanto, durante as atividades, a intensidade de dor diminuiu a partir do 3º pós-operatório, com exceção da atividade tossir em que a intensidade de dor diminuiu apenas no 6º pósoperatório. A ordem decrescente das atividades, quando avaliados os índices de intensidade de dor de todos os dias, foram tossir, virar-se de lado, respirar profundamente e em repouso. A incisão cirúrgica na região do esterno foi o local de dor mais referido pelos sujeitos, seguido da região epigástrica. As variáveis sexo, idade, tipo e tempo de cirurgia mostraram fraca associação com a intensidade de dor. Os descritores que mais caracterizaram a dor póscirurgia cardíaca por esternotomia foram forte, intensa, terrível, profunda e violenta. Os níveis elevados de dor podem estar contribuindo para um prolongamento do processo de recuperação. Os sujeitos consideram a multidimensionalidade do fenômeno doloroso ao utilizar de descritores para caracterizar a dor percebida. A investigação permitiu a melhor compreensão de aspectos relacionados à dor no pós-operatório de cirurgia cardíaca.