L’objectif de ce travail était de valider une méthode simple permettant de déterminer la sollicitation des métabolismes aérobie et anaérobie en cours d’activité en prenant l’exemple du tennis. Son principe général est de séquencer l’activité globale composite en sous-activités appelées activités fondamentales AF et de déterminer pour chacune d’entre elles la consigne énergétique DEc(i) afin de modéliser l’adaptation des filières énergétiques grâce au modèle bioénergétique Astrabio©. Des étalonnages permettent d’obtenir les fonctions d’astreinte qui relient DEc(i) à un paramètre mécanique M pertinent, vitesse de déplacement ou cadence de coups, pour les 6 AF sélectionnées au tennis. Une méthode d’analyse vidéo simple utilisant une seule caméra est appliquée lors des matchs de tennis pour déterminer les AF et estimer M. Ces paramètres sont introduits dans le modèle Astrabio© qui détermine la DEc(i) de chaque AF, DE aérobie et DE anaérobie par différence. Les quantités d’énergie aérobie prédites à chaque jeu des matchs ne présentent pas de différence statistique avec celles réellement mesurées par la méthode d’analyse directe de DE (K4b2). Elle met en évidence qu’en dépit d’une DE aérobie moyenne assez faible comprise entre 40% et 60% VO2max, le tennis se caractérise en réalité par des efforts supra-maximaux pouvant atteindre 300% VO2max lors des frappes de balle et le métabolisme anaérobie représente environ 30% de l’énergie totale dépensée sur un match. Cette méthode présente l’intérêt majeur d’être facilement applicable et de décrire l’astreinte réelle d’une activité en fournissant une estimation des DE aérobie et anaérobie à partir de l’analyse vidéo de l’activité / The aim of this study was to present a simple method to access the aerobic and anaerobic components of energy expenditure during activities as tennis. The principle consists in dividing the exercise into several fundamental activities (FA) based on a video recording and to estimate the metabolic power (MP) of each FA, the aerobic energy expenditure (EEO2mod) and the anaerobic energy expenditure thanks to the Astrabio© model. Six FA are selected and each FA is described by a “strain function” connecting MP to the selected pertinent mechanical variable M (speed running V and events frequency ω). The video analysis is used to determine FA, V and ω during tennis match. The parameters V and ω are introduced in the Astrabio© model to calculate MP of each FA. Then, this model calculated the aerobic energy expenditure (EEO2mod) in terms of MP and the anaerobic energy expenditures was calculated by substracting this (EEO2mod). There was no significant difference between calculated and measured oxygen consumptions (p<0.05). This method shows that even when the aerobic EE average is quite low and nearing 50% of VO2max, the total energy expenditure of a point can reach up 2 or 3 times the VO2max of the subjects during points and strokes. In tennis anaerobic metabolism can account for around 30% of the total energy expenditure per game, and almost 70% during points. The interest of this method is to provide a good estimation of aerobic and anaerobic energy expenditures thanks to a simple video analysis. This method presents the advantage that it is easy to apply and requires only simple and inexpensive equipment: a camera, a computer, and a tracking software

Despite extensive international acceptance of the critical role of mammalian post-dispersal seed predation in many plant communities, in New Zealand we have limited knowledge of these predators’ influence on plant recruitment in our forests. The principle objective of my thesis was to determine the importance of exotic mammals as post-dispersal seed predators in a New Zealand conifer-broadleaf forest remnant. To address this goal, I used a series of field-based experiments where the actions of different post-dispersal seed predators were separated by wire-mesh exclosures. My study was conducted at Mount Peel Forest Park Scenic Reserve, South Canterbury, New Zealand. Being a human modified conifer forest currently dominated by broadleaf species, it is typical of forest remnants in New Zealand. This presented an opportunity to study a wide range of both potential post-dispersal seed predators and broadleaf tree species. My findings indicate that exotic mammals are not only post-dispersal seed predators at Peel Forest, but are responsible for the majority of post-dispersal predation events observed. Ship rats (Rattus rattus) were the dominant post-dispersal seed predators, while brushtail possums (Trichosurus vulpecula), house mice (Mus musculus) and native invertebrates were also important post-dispersal seed predators for several tree species. Through use of time-lapse video and cafeteria experiments I found that exotic mammalian seed predators, when compared to native invertebrate seed predators, preyed upon larger-seeded plant species and were responsible for considerable seed losses of several tree species. However, exotic mammalian seed predators do share several foraging characteristics with native invertebrate seed predators, as predators foraged in similar habitats and responded in a similar way to changes in seed density. In investigating if post-dispersal seed predation by mammals had a flow-on effect to plant recruitment, I observed natural seedling densities at Peel Forest were significantly higher in the absence of mammalian seed predators, but I found no evidence that the presence of mammals significantly altered the overall species richness. At the community level, I did not find an interaction between habitat and exotic mammals, however I present evidence that for individual plant species a significant mammal : habitat interaction occurred. Consequently, even though my cafeteria experiment implied there was no significant difference in the overall amount of seed preyed upon within different habitats, the less favourable microsite conditions for germination under an intact continuous canopy allows mammals to exacerbate habitat-related patterns of seed mortality and have a noticeable effect on seedling establishment. In an effort to validate the use of manipulative experiments to predict the long-term effect of post-dispersal seed predation on plant dynamics, I attempted to link results of my cafeteria experiment with observed seedling abundance at Peel Forest. Seven tree species were used in this comparison and a strong correlation was observed. This result shows that the level of post-dispersal seed predation determined in the cafeteria experiment provided a good predictor of the effect of mammalian post-dispersal seed predation on seedling establishment. To fully gauge the impact of mammalian post-dispersal seed predators on seedling establishment, the relationship between these seed predators and the type of recruitment limitation experienced by a plant species was also investigated. By using a combination of seed addition, plot manipulations and seed predator exclusion I was able to investigate this relationship. I found evidence that seed limitation at Peel Forest is positively correlated with seed size, and that while mammalian post-dispersal seed predators can further reduce plant recruitment of plant species experiencing seed limitation, the influence of mammals in determining plant recruitment was limited for plant species experiencing microsite limitation. My study has proven that exotic mammals are now the dominant post-dispersal seed predators at Peel Forest, the amount of seed preyed upon varies among plant species, and post-dispersal seed predation by mammalian species can lead to differences in seedling richness and abundance. I proved that the influence of exotic mammals on seedling establishment is also linked to habitat structure and recruitment limitations. When combined these observations suggest that exotic mammalian post-dispersal seed predators may play an important role in determining landscape abundance and distribution of plants at Peel Forest.

Η παρούσα διατριβή πραγματεύεται το πρόβλημα της αποτελεσματικής αξιολόγησης ευχρηστίας φορητών εφαρμογών. Με τον όρο “φορητές εφαρμογές” περιγράφονται οι ηλεκτρονικές υπηρεσίες που είναι σχεδιασμένες να παρέχονται μέσω φορητών συσκευών και έχουν ως στόχο να βοηθήσουν τον άνθρωπο στις καθημερινές του προσωπικές ή επαγγελματικές δραστηριότητες. Η αλληλεπίδραση ενός χρήστη με μία τέτοια ηλεκτρονική υπηρεσία είναι μια διαδικασία πολύπλοκη που δεν υποστηρίζεται από μία μεμονωμένη εφαρμογή και δεν περιορίζεται πλέον στα στενά όρια ενός γραφείου. Αντίθετα, επιτυγχάνεται με διαμεσολάβηση σύγχρονων και συχνά πολυχρηστικών υπολογιστικών μονάδων (φορητές συσκευές, υπολογιστές) και επιμέρους συσκευών και λαμβάνει χώρα σε διάφορα φυσικά περιβάλλοντα. Σε πολλές περιπτώσεις εμπλέκει περισσότερους από έναν χρήστες οι οποίοι μπορεί να συνεργάζονται είτε εκ του σύνεγγυς είτε εξ αποστάσεως προκειμένου να εκτελέσουν μια εργασία. Η μελέτη αυτής της αλληλεπίδρασης μπορεί να δώσει πολύτιμα συμπεράσματα τόσο για την ευκολία χρήσης των σύγχρονων υπολογιστικών συστημάτων όσο και για τη χρησιμότητά τους και αποτελεί μια σημαντική ερευνητική πρόκληση του επιστημονικού πεδίου Αλληλεπίδρασης Ανθρώπου Υπολογιστή και ιδιαίτερα της Τεχνολογίας Ευχρηστίας. Οι κλασσικές μέθοδοι ευχρηστίας καλούνται να αντιμετωπίσουν την πρόκληση της επιτυχούς αξιολόγησης των φορητών εφαρμογών, δηλαδή της αποτίμησης των χαρακτηριστικών ποιότητάς τους που επηρεάζουν την εμπειρία χρήσης τους. Τόσο τα χαρακτηριστικά των χρησιμοποιούμενων συσκευών όσο και το περιβάλλον χρήσης των εφαρμογών αυτών, τις διαφοροποιούν κατά πολύ από τις εφαρμογές επιτραπέζιου υπολογιστή. Καθώς η διείσδυσή τους στις καθημερινές δραστηριότητες του ανθρώπου είναι ολοένα και αυξανόμενη, η ανάγκη για μεθοδολογίες και εργαλεία που μπορούν να βοηθήσουν την αξιολόγηση της χρήσης τους κρίνεται επιτακτική. Στην παρούσα εργασία καταρχήν περιγράφονται οι κλασσικές μέθοδοι αξιολόγησης υπολογιστικών συστημάτων και στη συνέχεια αξιολογούνται ως προς την επάρκεια των δεδομένων που μπορούν να συλλέξουν και την πληρότητα των αποτελεσμάτων που μπορούν να δώσουν στην περίπτωση χρήσης τους για αξιολόγηση ευχρηστίας φορητών εφαρμογών. Παράλληλα γίνεται επισκόπηση της βιβλιογραφίας που Περίληψη 12 Διδακτορική διατριβή αφορά την αξιολόγηση ευχρηστίας των φορητών εφαρμογών και εξετάζονται τα κυριότερα ζητήματα της μεθόδου αξιολόγησης, περιλαμβάνοντας μεταξύ άλλων την επιλογή του χώρου αξιολόγησης και τα δεδομένα που πρέπει να συλλεγούν και αναλυθούν. Στη συνέχεια προτείνεται η μεθοδολογία αξιολόγησης ευχρηστίας φορητών εφαρμογών MOBELIC (MOBile Evaluation Life Cycle methodology). Η MOBELIC υποστηρίζει την αξιολόγηση ευχρηστίας μιας φορητής εφαρμογής σε όλο τον κύκλο ανάπτυξής της. Αποτελεί συνδυασμό από υπάρχουσες μεθόδους αξιολόγησης λαμβάνοντας όμως υπόψη τη φάση ανάπτυξης στην οποία βρίσκεται η υπό αξιολόγηση φορητή εφαρμογή, την ποιότητα των αποτελεσμάτων που προκύπτουν από κάθε μέθοδο, το χρόνο και τους πόρους που απαιτούνται για τη διεξαγωγή καθεμιάς από αυτές, καθώς και το χρόνο για την ανάλυση των δεδομένων τους και την εξαγωγή πολύτιμων συμπερασμάτων. Η συνεισφορά της μεθοδολογίας MOBELIC έγκειται στο γεγονός ότι βελτιστοποιεί τη διαδικασία αξιολόγησης μιας φορητής εφαρμογής υποστηρίζοντάς την από τις αρχικές φάσεις ανάπτυξης μέχρι τα τελευταία στάδια ολοκλήρωσής της. Προκειμένου να υποστηριχθεί η μεθοδολογία ΜOBELIC, σχεδιάστηκε και αναπτύχθηκε το εργαλείο ActivityLens. Αποτελεί ένα εργαλείο παρατήρησης και ανάλυσης των δεδομένων που συλλέγονται κατά τη διάρκεια μελετών της αλληλεπίδρασης χρηστών με φορητές εφαρμογές. Το εργαλείο παρέχει τη δυνατότητα παρακολούθησης των διακριτών φάσεων ανάπτυξης μιας φορητής εφαρμογής και επιτρέπει τη σύγκριση των προβλημάτων ευχρηστίας που παρατηρούνται σε αυτές, χρησιμοποιώντας κοινό σχήμα κωδικοποίησης των προβλημάτων. Παράλληλα, ενσωματώνοντας κατάλληλα σχεδιασμένες λειτουργίες, βοηθάει τον αναλυτή να απομονώσει συγκεκριμένους τύπους δράσεων που υποδεικνύουν πρόβλημα στη χρήση των εξεταζόμενων εφαρμογών. Τέλος στα πλαίσια της παρούσας διατριβής, εξετάζεται η εφαρμοσιμότητα της προτεινόμενης μεθοδολογίας καθώς και η αποτελεσματικότητα του εργαλείου ανάλυσης, μέσα από μία μελέτη περίπτωσης που περιλαμβάνει τη διενέργεια 3 διακριτών πειραμάτων. Η αξιολόγηση αφορά μία συνεργατική εκπαιδευτική φορητή εφαρμογή προορισμένη να πληροφορεί μαθητές για τα εκθέματα ενός Μουσείου. / This thesis deals with the problem of effective usability evaluation of mobile applications. The term “mobile applications” describes the electronic services that are designed to operate under mobile devices in order to meet the everyday needs of people. There are special characteristics of mobile applications, compared to established desktop applications. Thus, the dynamic context of use and the small size of mediating mobile devices differentiate their use and increase the complexity of their usability evaluation. There are various established methods that have been extensively used in traditional human-computer interaction research in order to evaluate the use of desktop applications. This thesis, firstly examines the suitability of such methods in usability evaluation of mobile applications. The survey indicated that all these methods can be effective. However their application produces different types of data and requires different amount and kind of resources. In the frame of this thesis, MOBELIC, a new methodology for the usability evaluation of mobile applications, is proposed. The goal of MOBELIC is the optimization of evaluation process due to the short life-cycle of mobile applications. It is an effective combination of selected established methods supporting the redesign of a mobile application during its development life-cycle. The main characteristic of this methodology is that it proposes the iterative application of each evaluation method, depending on certain factors, such as the mobile application implementation phase, the quality of the expected method results and the required resources. Subsequently, the design and the implementation of a usability evaluation tool, the ActivityLens, that support the proposed methodology, are described in this thesis. ActivityLens integrates and synchronizes multiple heterogeneous data that describe the interaction of real users with mobile applications. Furthermore, the tool incorporates especially designed mechanisms for annotation and analysis of collected data in order to facilitate usability experts to extract meaningful information. The applicability of the proposed methodology and the effectiveness of the ActivityLens are examined through an extended case study in which the usability of an educational mobile application is evaluated. The case study consists of three discrete experiments including the use of different evaluation methods according to the proposed methodology.

Mémoire numérisé par la Division de la gestion de documents et des archives de l'Université de Montréal

The Thesis "Videosequences and their use in teaching physics at the elementary school" provides a basic understanding of the issue of the use of audiovisual technology in an environment at the elementary school, focusing on the integration of video in teaching physics at the second stage. It describes the possibilities of video sources on the Internet and instructions for their preparation and to their own creation. The Thesis is also based on practical experiences that means the inclusion of specific movies in two selected topics in teaching physics.

Cette dissertation traite du problème du suivi d'objets visuels, dont le but est de localiser un objet et de déterminer sa trajectoire au cours du temps. En particulier, nous nous concentrons sur les scénarios difficiles, dans lesquels les objets subissent d'importantes déformations et occlusions, ou quittent le champs de vision. A cette fin, nous proposons deux méthodes robustes qui apprennent un modèle pour l'objet d'intérêt et le mettent à jour, afin de refléter ses changements au cours du temps.Notre première méthode traite du problème du suivi dans le cas où les objets subissent d'importantes transformations géométriques comme une rotation ou un changement d'échelle. Nous présentons un nouvel algorithme de sélection de propositions, qui étend l'approche traditionnelle de ``suivi par détection''. Cette méthode procède en deux étapes: proposition puis sélection. Dans l'étape de proposition, nous construisons un ensemble de candidats qui représente les localisations potentielles de l'objet en estimant de manière robuste les transformations géométriques. La meilleure proposition est ensuite sélectionnée parmi cet ensemble de candidats pour précisément localiser l'objet en utilisant des indices d'apparence et de mouvement.Dans un second temps, nous traitons du problème de la mise à jour de modèles dans le suivi visuel, c'est-à-dire de déterminer quand il est besoin de mettre à jour le modèle de la cible, lequel peut subir une occlusion, ou quitter le champs de vision. Pour résoudre cela, nous utilisons des indices de mouvement pour identifier l'état d'un objet de manière automatique et nous mettons à jour le modèle uniquement lorsque l'objet est entièrement visible. En particulier, nous utilisons des trajectoires à long terme ainsi qu'une technique basée sur la coup de graphes pour estimer les parties de l'objet qui sont visibles.Nous avons évalué nos deux approches de manière étendue sur différents bancs d'essai de suivi, en particulier sur le récent banc d'essai de suivi en ligne et le jeu de donnée du concours de suivi visuel. Nos deux approches se comparent favorablement à l'état de l'art et font montre d'améliorations significatives par rapport à plusieurs autres récents suiveurs. Notre soumission au concours de suivi d'objets visuels de 2015 a par ailleurs remporté l'une de ces compétitions. / In this dissertation we address the problem of visual object tracking, whereinthe goal is to localize an object and determine its trajectory over time. Inparticular, we focus on challenging scenarios where the object undergoessignificant transformations, becomes occluded or leaves the field of view. Tothis end, we propose two robust methods which learn a model for the object ofinterest and update it, to reflect its changes over time.Our first method addresses the tracking problem in the context of objectsundergoing severe geometric transformations, such as rotation, change in scale.We present a novel proposal-selection algorithm, which extends the traditionaldiscriminative tracking-by-detection approach. This method proceeds in twostages -- proposal followed by selection. In the proposal stage, we compute acandidate pool that represents the potential locations of the object byrobustly estimating the geometric transformations. The best proposal is thenselected from this candidate set to localize the object precisely usingmultiple appearance and motion cues.Second, we consider the problem of model update in visual tracking, i.e.,determining when to update the model of the target, which may become occludedor leave the field of view. To address this, we use motion cues to identify thestate of the object in a principled way, and update the model only when theobject is fully visible. In particular, we utilize long-term trajectories incombination with a graph-cut based technique to estimate parts of the objectsthat are visible.We have evaluated both our approaches extensively on several trackingbenchmarks, notably, recent online tracking benchmark and the visual objecttracking challenge datasets. Both our approaches compare favorably to thestate of the art and show significant improvement over several other recenttrackers. Specifically, our submission to the visual object tracking challengeorganized in 2015 was the winner in one of the competitions.

Ces travaux de recherche concernent la modélisation de l'information dynamique fonctionnelle fournie par les champs de déplacements apparents à l'aide de base de polynômes orthogonaux. Leur objectif est de modéliser le mouvement et la texture extraites afin de l'exploiter dans les domaines de l'analyse et de la reconnaissance automatique d'images et de vidéos. Nous nous intéressons aussi bien aux mouvements humains qu'aux textures dynamiques. Les bases de polynômes orthogonales ont été étudiées. Cette approche est particulièrement intéressante car elle offre une décomposition en multi-résolution et aussi en multi-échelle. La première contribution de cette thèse est la définition d'une méthode spatiale de décomposition d'image : l'image est projetée et reconstruite partiellement avec un choix approprié du degré d'anisotropie associé à l'équation de décomposition basée sur des transformations polynomiales. Cette approche spatiale est étendue en trois dimensions afin d'extraire la texture dynamique dans des vidéos. Notre deuxième contribution consiste à utiliser les séquences d'images qui représentent les parties géométriques comme images initiales pour extraire les flots optiques couleurs. Deux descripteurs d'action, spatial et spatio-temporel, fondés sur la combinaison des informations du mouvement/texture sont alors extraits. Il est ainsi possible de définir un système permettant de reconnaître une action complexe (composée d'une suite de champs de déplacement et de textures polynomiales) dans une vidéo. / The research relies on modeling the dynamic functional information from the fields of apparent movement using basic orthogonal polynomials. The goal is to model the movement and texture extracted for automatic analysis and recognition of images and videos. We are interested both in human movements as dynamic textures. Orthogonal polynomials bases were studied. This approach is particularly interesting because it offers a multi-resolution and a multi-scale decomposition. The first contribution of this thesis is the definition of method of image spatial decomposition: the image is projected and partially rebuilt with an appropriate choice of the degree of anisotropy associated with the decomposition equation based on polynomial transformations. This spatial approach is extended into three dimensions to retrieve the dynamic texture in videos. Our second contribution is to use image sequences that represent the geometric parts as initial images to extract color optical flow. Two descriptors of action, spatial and space-time, based on the combination of information of motion / texture are extracted. It is thus possible to define a system to recognize a complex action (composed of a series of fields of motion and polynomial texture) in a video.

Submitted by CAMILO CARRANZA (carranzacamilo@gmail.com) on 2017-07-25T19:58:22Z No. of bitstreams: 1 Camilo_PhD_Thesis.pdf: 11035322 bytes, checksum: efe400c07b13cabff41e927078789c59 (MD5) / Approved for entry into archive by Luiz Galeffi (luizgaleffi@gmail.com) on 2017-07-26T18:31:30Z (GMT) No. of bitstreams: 1 carranzalopez_jc_dr_ilha.pdf: 11035322 bytes, checksum: efe400c07b13cabff41e927078789c59 (MD5) / Made available in DSpace on 2017-07-26T18:31:30Z (GMT). No. of bitstreams: 1 carranzalopez_jc_dr_ilha.pdf: 11035322 bytes, checksum: efe400c07b13cabff41e927078789c59 (MD5) Previous issue date: 2017-06-05 / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) / Nesta tese são investigadas, teórica e experimentalmente, a sincronização em fase e a sincronização em anti-fase de dois metrônomos oscilando sobre uma base móvel, a partir de um modelo aqui proposto. Uma descrição do funcionamento do mecanismo de escapamento dos metrônomos é feita, junto a um estudo da relação entre este e o oscilador de van der Pol. Também uma aproximação experimental do valor do amortecimento do metrônomo é fornecida. A frequência instantânea das respostas, numérica e experimental, do sistema é usada na analise. A diferença de outros trabalhos prévios, os dados experimentais têm sido adquiridos usando vídeos dos experimentos e extraídos com ajuda do software Tracker. Para investigar a relação entre as condições iniciais do sistema e seu estado final de sincronização, foram usados mapas bidimensionais chamados ‘basins of attraction’. A relação entre o modelo proposto e um modelo prévio também é mostrada. Encontrou-se que os parâmetros relevantes em relação a ambos os tipos de sincronização são a razão entre a massa do metrônomo e a massa da base, e o amortecimento do sistema. Tem-se encontrado, tanto experimental quanto teoricamente, que a frequência de oscilação dos metrônomos aumenta quando o sistema sincroniza-se em fase, e se mantém a mesma de um metrônomo isolado quando o sistema sincroniza-se em anti-fase. A partir de simulações numéricas encontrou-se que, em geral, incrementos no amortecimento do sistema levam ao sistema se sincronizar mais em fase do que em anti-fase. Adicionalmente se encontrou que, para dado valor de amortecimento, diminuir a massa da base leva a uma situação em que a sincronização em anti-fase é mais comum do que a sincronização em fase. / This thesis concerns a theoretical and experimental investigation into the synchronization of two coupled metronomes. A simplified model is proposed to study in-phase and anti-phase synchronization of two metronomes oscillating on a mobile base. A description of the escapement mechanism driving metronomes is given and its relationship with the van der Pol oscillator is discussed. Also an experimental value for the damping in the metronome is determined. The instantaneous frequency of the responses from both numerical and experimental data is used in the analysis. Unlike previous studies, measurements are made using videos and the time domain responses of the metronomes extracted by means of tracker software. Basins of attraction are used to investigate the relationship between initial conditions, parameters and both final synchronization states. The relationship between the model and a previous pendulum model is also shown. The key parameters concerning both kind of synchronization have been found to be the mass ratio between the metronome mass and the base mass, and the damping in the system. It has been shown, both theoretically and experimentally, that the frequency of oscillation of the metronomes increases when the system reaches in-phase synchronization, and is the same as an isolated metronome when the system synchronizes in anti-phase. From numerical simulations, it has been found that, in general, increasing damping leads the system to synchronize more in-phase than in anti-phase. It has also been found that, for a given damping value, decreasing the mass of the base results in the situation where anti-phase synchronization is more common than in-phase synchronization.

Motion is an important cue in videos that captures the dynamics of moving objects. It helps in effective analysis of various event related tasks such as human action recognition, anomaly detection, tracking, crowd behavior analysis, traffic monitoring, etc. Generally, accurate motion information is computed using various optical flow estimation techniques. On the other hand, coarse motion information is readily available in the form of motion vectors in compressed videos. Utilizing these encoded motion vectors reduces the computational burden involved in flow estimation and enables rapid analysis of video streams. In this work, the focus is on analyzing motion patterns, retrieved from either motion vectors or optical flow, in order to do various event analysis tasks such as video classification, anomaly detection and crowd flow segmentation. In the first section, we utilize the motion vectors from H.264 compressed videos, a compression standard widely used due to its high compression ratio, to address the following problems. i) Video classification: This work proposes an approach to classify videos based on human action by capturing spatio-temporal motion pattern of the actions using Histogram of Oriented Motion Vector (HOMV) ii) Crowd flow segmentation: In this work, we have addressed the problem of flow segmentation of the dominant motion patterns of the crowds. The proposed approach combines multi-scale super-pixel segmentation of the motion vectors to obtain the final flow segmentation. iii) Anomaly detection: This problem is addressed by local modeling of usual behavior by capturing features such as magnitude and orientation of each moving object. In all the above approaches, the focus was to reduce computations while retaining comparable accuracy to pixel domain processing. In second section, we propose two approaches for anomaly detection using optical flow. The first approach uses spatio-temporal low level motion features and detects anomalies based on the reconstruction error of the sparse representation of the candidate feature over a dictionary of usual behavior features. The main contribution is in enhancing each local dictionary by applying an appropriate transformation on dictionaries of the neighboring regions. The other algorithm aims to improve the accuracy of anomaly localization through short local trajectories of super pixels belonging to moving objects. These trajectories capture both spatial as well as temporal information effectively. In contrast to compressed domain analysis, these pixel level approaches focus on improving the accuracy of detection with reasonable detection speed.

Abstract The road freight transport industry as a labour-intensive sector is dependent on the work ability and well-being at work of employees. The majority of the occupational accidents are related to work phases outside the cab. These work phases, which are performed in various different work environments, contain several kinds of ergonomic discomforts. This poses complex challenges for the employers from a safety and productivity point of view. The framework of this thesis is based on the foundations of ergonomics and design science. The main objective was to provide knowledge that can be implemented into the design and management of work systems for local and short haul delivery operations. Material was obtained from two sources. A meta-synthesis was performed to frame holistic management in a human perspective. Furthermore, additional in-depth design knowledge was obtained through participatory ergonomics video analyses on drivers’ work outside the cab. Video analyses resulted in 262 identifications of demanding work situations where ergonomic discomforts and risks of accidents occurred. Sudden over-exertions and strains, falls and slips as well as losing control of work equipment were the most common deviations related to drivers’ work outside the cab and mainly related to physical activities of movement and carrying by hand. The majority of the work situations identified were performed in cargo spaces or elsewhere within the truck structure or at premises and yards that are administered by the customers or other stakeholders. In these environments, drivers tend to perform their work manually or using different types of work equipment. This thesis provides new in-depth knowledge on drivers’ work outside the cab. The results show that different stakeholders can contribute to drivers’ work systems. The knowledge provided by drivers and other stakeholders can be applied to holistic design and management processes at company level. Moreover, the knowledge can also be applied to broader value chain design and management processes. / Tiivistelmä Tieliikenteen tavarankuljetus työvoimavaltaisena toimialana on riippuvainen henkilöstön työkyvystä ja -hyvinvoinnista. Suurin osa tapaturmista liittyy työtehtäviin ohjaamon ulkopuolella. Näitä töitä tehdään hyvin vaihtelevissa työympäristöissä ja niihin työtehtäviin liittyy monenlaisia ergonomisia haittakuormitustekijöitä. Tämä asettaa haasteita niin työsuojelun kuin tuottavuuden näkökulmasta. Väitöskirjan viitekehys pohjautuu ergonomiaan sekä suunnittelutieteeseen. Tavoitteena on tuottaa tietoa, jota voidaan hyödyntää työjärjestelmien suunnittelussa ja johtamisessa erityisesti maaliikenteen jakelukuljetuksissa. Materiaali koostui kahdesta osiosta. Metasynteesillä muodostettiin näkemys kokonaisvaltaisesta johtamisesta ihmisnäkökulmasta. Lisäksi kuljettajat ja sidosryhmien edustajat analysoivat osallistuvan ergonomian keinoin videoaineistoa jakelukuljettajien työstä ohjaamon ulkopuolella. Videoanalyyseissa tunnistettiin yhteensä 262 työtilannetta, jossa esiintyy erilaisia ergonomisia haittakuormitustekijöitä sekä mahdollisia tapaturmariskejä. Äkilliset fyysiset kuormitukset, putoamiset, liukastumiset ja kaatumiset sekä työvälineiden hallinnan menettäminen olivat yleisimpiä tunnistettuja poikkeamia kuljettajan työssä. Pääasiassa nämä liittyivät kuljettajan liikkumiseen sekä erilaisten taakkojen kantamiseen. Valtaosassa (85&#160;%) havainnoista kuljettaja työskenteli ajoneuvon kuormatilassa tai päällirakenteissa tai asiakkaiden tai muiden sidosryhmien hallinnoimissa työympäristöissä. Näissä työympäristöissä kuljettaja työskenteli sekä manuaalisesti käsin että hyödyntäen erilaisia apuvälineitä. Väitöskirja tarjoaa uudenlaista syvällistä tietoa kuljettajan työstä ohjaamon ulkopuolella. Eri sidosryhmät voivat osaltaan vaikuttaa kuljettajan työjärjestelmiin. Kuljettajien ja sidosryhmien tuottamaa tietoa voidaan soveltaa työjärjestelmien kokonaisvaltaisessa suunnittelussa ja johtamisessa niin yritystasolla kuin myös suunniteltaessa ja johdettaessa laajempia arvoketjuja.