Advances in time-lapse 3D in vivo fluorescence microscopic imaging techniques enables the observation and investigation into the migration of Oligodendrocyte precursor cells (OPCs) and its role in the central nervous system. However, current practice of image-based OPC motility analysis heavily relies on manual labeling and tracking on 2D max projection of the 3D data, which suffers from massive human labor, subjective biases, weak reproducibility and especially information loss and distortion. Besides, due to the lack of OPC specific genetically encoded indicator, OPCs can only be identified from other oligodendrocyte lineage cells by their observed motion patterns. Automated analytical tools are needed for the identification and tracking of OPCs. In this dissertation work, we proposed an analytical framework, MicTracker (Migrating Cell Tracker), for the integrated task of identifying, segmenting and tracking migrating cells (OPCs) from in vivo time-lapse fluorescence imaging data of high-density cell clusters composed of cells with different modes of motions. As a component of the framework, we presented a novel strategy for cell segmentation with global temporal consistency enforced, tackling the challenges caused by highly clustered cell population and temporally inconsistently blurred boundaries between touching cells. We also designed a data association algorithm to address the violation of usual assumption of small displacements. Recognizing that the violation was in the mixed cell population composed of two cell groups while the assumption held within each group, we proposed to solve the seemingly impossible mission by de-mixing the two groups of cell motion modes without known labels. We demonstrated the effectiveness of MicTracker in solving our problem on in vivo real data. / Doctor of Philosophy / Oligodendrocyte precursor cells (OPCs) are a type of motile cells in the central nervous system (CNS). OPCs' migration plays a critical role in the repair and re-distribution of myelin sheaths, a structures that helps to accelerate the transmission of electrical signals from neuron to neuron. But the mechanism behind the motility of OPCs is largely unclear. In recent years, advances in genetic fluorescence indicators and time-lapse optical microscopic imaging techniques, especially 3D in vivo imaging, enables neuroscientists to investigate into the puzzle. However, current practice of OPC motility analysis heavily relies on compressing the 3D data into 2D then manually tracking the OPCs, which suffers from not only massive human labor, subjective biases, weak reproducibility and especially information loss and distortion. Automated analytical tools are needed. Due to the limitation of current techniques in fluorescent labeling of cells in live animals, OPCs cannot be distinctively labeled. Instead, in the field of view there are also other irrelevant cells that cannot migrate but locally vibrate. Therefore, the human analyzer or the analytical software is supposed to detect OPCs from a cluster of touching cells containing multiple types of cells by their motion patterns only. In this dissertation, we presented a fully automatic machine learning based algorithm, MicTracker (Migrating Cell Tracker), to identify and track migrating OPCs. The task cannot be straightforwardly solved by existing generic-purpose cell tracking tools due to quite a few special challenges. To tackle the challenges, we also proposed novel methods for two major modules of MicTracker, segmentation and linking, respectively. We demonstrated the effectiveness of MicTracker and its components on real data and compared it with related existing works. The results of experiments showed notable superiority of MicTracker in solving our problem, compared with existing methods.

Providing computer-aided support for human analysis of videos has been a battle of extremes. Powerful solutions exist, but they tend to be domain-specific and complex. The user-friendly, simple systems provide little analysis support beyond basic media player functionality. We propose a human-centered, domain-independent solution between these two points. Our proposed model and system, VAST, is based on our experience in two diverse video analysis domains: science and athletics. Multiple-perspective location metadata is used to group related video clips together. Users interact with these clip groups through a novel interaction paradigm ? views. Each view provides a different context by which users can judge and evaluate the events that are captured by the video. Easy conversion between views allows the user to quickly switch between contexts. The model is designed to support a variety of user goals and expertise with minimal producer overhead. To evaluate our model, we developed a system prototype and conducted several rounds of user testing requiring the analysis of volleyball practice videos. The user tasks included: foreground analysis, ambiguous identification, background analysis, and planning. Both domain novices and experts participated in the study. User feedback, participant performance, and system logs were used to evaluate the system. VAST successfully supported a variety of problem solving strategies employed by participants during the course of the study. Participants had no difficulty handling multiple views (and resulting multiple video clips) simultaneously opened in the workspace. The capability to view multiple related clips at one time was highly regarded. In all tasks, except the open-ended portion of the background analysis, participants performed well. However, performance was not significantly influenced by domain expertise. Participants had a favorable opinion of the system?s intuitiveness, ease of use, enjoyability, and aesthetics. The majority of participants stated a desire to use VAST outside of the study, given the opportunity.

Manual video analysis is apparently a tedious task. An efficient solution is of highly importance to automate the process and to assist operators. A major goal of video analysis is understanding and recognizing human activities captured by surveillance cameras, a very challenging problem; the activities can be either individual or interactional among multiple objects. It involves extraction of relevant spatial and temporal information from visual images. Most video analytics systems are constrained by specific environmental situations. Different domains may require different specific knowledge to express characteristics of interesting events. Spatial-temporal trajectories have been utilized to capture motion characteristics of activities. The focus of this dissertation is on how trajectories are utilized in assist in developing video analytic system in the context of surveillance. The research as reported in this dissertation begins real-time highway traffic monitoring and dynamic traffic pattern analysis and in the end generalize the knowledge to event and activity analysis in a broader context. The main contributions are: the use of the graph-theoretic dominant set approach to the classification of traffic trajectories; the ability to first partition the trajectory clusters using entry and exit point awareness to significantly improve the clustering effectiveness and to reduce the computational time and complexity in the on-line processing of new trajectories; A novel tracking method that uses the extended 3-D Hungarian algorithm with a Kalman filter to preserve the smoothness of motion; a novel camera calibration method to determine the second vanishing point with no operator assistance; and a logic reasoning framework together with a new set of context free LLEs which could be utilized across different domains. Additional efforts have been made for three comprehensive surveillance systems together with main contributions mentioned above.

Title: Athletic physical literacy of secondary school pupils Objectives: The aim of this thesis was to compare athletic literacy of seventh and ninth-graders and to test if these pupils are able to perform athletic excercises and can be therefore called athletically literate. Methods: Cross-sectional study, video analysis. Results: The final evaluation showed that the results of both tested grades in the nodal points are similar. Both groups have comparable athletic skills, however, the ninth-graders reached slightly better results. These pupils have more experience and they were able to coordiate their movements better and their performance was also smoother. This "extra" ability is what can be called athletic literacy. The pupils have basic athletic skills and they also know how and when to use these skills properly. Keywords: Athletic literacy, video analysis, kinogram. pupils

Submitted by Larissa Vitoria Cardoso Cusielo (larissavitoria@id.uff.br) on 2017-09-05T17:18:04Z No. of bitstreams: 3 license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) Rafael Pinheiro Santos - Dissertação Final.pdf: 3826907 bytes, checksum: 9fc225c7d1e17bc76239ae3a282bfd50 (MD5) Rafael Pinheiro Santos - Produto Final.pdf: 2418114 bytes, checksum: e33fdbf6658370c3c052fc82d477414e (MD5) / Rejected by Biblioteca do Aterrado BAVR (bavr@ndc.uff.br), reason: Titulo em caixa alto on 2017-09-05T19:17:58Z (GMT) / Submitted by Larissa Vitoria Cardoso Cusielo (larissavitoria@id.uff.br) on 2017-09-05T19:44:52Z No. of bitstreams: 3 license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) Rafael Pinheiro Santos - Dissertação Final.pdf: 3826907 bytes, checksum: 9fc225c7d1e17bc76239ae3a282bfd50 (MD5) Rafael Pinheiro Santos - Produto Final.pdf: 2418114 bytes, checksum: e33fdbf6658370c3c052fc82d477414e (MD5) / Approved for entry into archive by Biblioteca do Aterrado BAVR (bavr@ndc.uff.br) on 2017-09-27T21:48:05Z (GMT) No. of bitstreams: 3 license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) Rafael Pinheiro Santos - Dissertação Final.pdf: 3826907 bytes, checksum: 9fc225c7d1e17bc76239ae3a282bfd50 (MD5) Rafael Pinheiro Santos - Produto Final.pdf: 2418114 bytes, checksum: e33fdbf6658370c3c052fc82d477414e (MD5) / Made available in DSpace on 2017-09-27T21:48:05Z (GMT). No. of bitstreams: 3 license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) Rafael Pinheiro Santos - Dissertação Final.pdf: 3826907 bytes, checksum: 9fc225c7d1e17bc76239ae3a282bfd50 (MD5) Rafael Pinheiro Santos - Produto Final.pdf: 2418114 bytes, checksum: e33fdbf6658370c3c052fc82d477414e (MD5) / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior / Universidade Federal Fluminense. Instituto de Ciências Exatas. Programa de Pós-Graduação em Ensino de Física. Volta Redonda, RJ / A proposta desta dissertação é apresentar o desenvolvimento de um produto educacional para ensinar Cinemática através de Vídeo análise utilizando o software Tracker para turmas do ensino médio. O produto educacional consiste em uma sequencia didática, que será conduzida por uma série de perguntas com o intuito de construir com os alunos, a partir da observação dos fenômenos e da experimentação, um modelo matemático dos movimentos tradicionalmente estudados no ensino médio. Para alicerçar o desenvolvimento deste produto, apresentamos uma breve fundamentação teórica sobre Cinemática, Aprendizagem Significativa e Tecnologia da Informação e Comunicação (TICs). Essa proposta de trabalho foi levada à sala de aula e os resultados da aplicação são avaliados pela técnica de análise de conteúdo aplicado à sequencia didática. Acreditamos que o produto apresentado pode participar diretamente da formação das Competências e Habilidades dos alunos. Além disso, a realização da vídeo análise motivou os estudantes e foi possível perceber um maior envolvimento nas atividades, o que é prerrogativa fundamental para atingir-se a aprendizagem significativa. / The purpose of this dissertation is to present the development of an educational product to teach kinematics through analysis video using the Tracker software for high school classes. The educational product consists on a didactic sequence, which will be conducted through a series of questions in order to build with students from the observation of phenomena and experimentation, a mathematical model of movements traditionally studied at high school. In support of the development of this product there is a brief theoretical framework of kinematics, Meaningful Learning and Information and Communication Technology (ICT). This work proposal was brought to the classroom and the application of the results are evaluated by content analysis technique applied to the teaching sequence. We believe that the product presented may participate directly in the formation of skills and abilities of the students. Furthermore, the realization of video analysis motivated the students and it was possible to observe a greater involvement in the activities, which is a fundamental prerogative to achieve meaningful learning.

Hockey is a popular sport, and at its highest levels, it is a complex contact game characterized by physical strength, speed, and skill. The interaction of these characteristics contributes to the inherent risk of injury athletes must face while playing. Among hockey injuries, concussions are one of the most commonly sustained by athletes across all levels of play and age groups. Significant public attention, combined with poorly understood long-term effects, indicates the importance of tangible preventive strategies. The main goal of this thesis was to understand, through video analysis, how playing characteristics and mechanism of injury contribute to concussions in the National Hockey League (NHL). In the first study, the development and validation of an observational recording tool used to code and analyze NHL concussions observed via video analysis was described. The second study attempted to synthesize the description of players’ characteristics, antecedent events, and contextual variables associated with events leading to concussion at the NHL level. Several specific risk factors for concussion in NHL players were identified, including position, body size, specific locations on the ice, and particular situations based on a player's position. The final study systematically analyzed how concussions occur to identify potential pattern(s) of concussions. A common specific injury mechanism characterized by player-to-player contact and resulting in contact to the head by the shoulder, elbow, or gloves, was also identified. When the principal mechanism was refined further, several important characteristics were discernable: (i) contact was often to the lateral aspect of the head; (ii) the player who suffered a concussion was often not in possession of the puck; and (iii) no penalty was called on the play. Collectively, these studies served to address gaps in the literature; the implications for informing prevention and management strategies are also discussed.

Hockey is a popular sport, and at its highest levels, it is a complex contact game characterized by physical strength, speed, and skill. The interaction of these characteristics contributes to the inherent risk of injury athletes must face while playing. Among hockey injuries, concussions are one of the most commonly sustained by athletes across all levels of play and age groups. Significant public attention, combined with poorly understood long-term effects, indicates the importance of tangible preventive strategies. The main goal of this thesis was to understand, through video analysis, how playing characteristics and mechanism of injury contribute to concussions in the National Hockey League (NHL). In the first study, the development and validation of an observational recording tool used to code and analyze NHL concussions observed via video analysis was described. The second study attempted to synthesize the description of players’ characteristics, antecedent events, and contextual variables associated with events leading to concussion at the NHL level. Several specific risk factors for concussion in NHL players were identified, including position, body size, specific locations on the ice, and particular situations based on a player's position. The final study systematically analyzed how concussions occur to identify potential pattern(s) of concussions. A common specific injury mechanism characterized by player-to-player contact and resulting in contact to the head by the shoulder, elbow, or gloves, was also identified. When the principal mechanism was refined further, several important characteristics were discernable: (i) contact was often to the lateral aspect of the head; (ii) the player who suffered a concussion was often not in possession of the puck; and (iii) no penalty was called on the play. Collectively, these studies served to address gaps in the literature; the implications for informing prevention and management strategies are also discussed.

Emhart Glass AB är en världsledande internationell tillverkare av automatiska glasformningsmaskiner för framställning av olika glasprodukter bl.a glasflaskor och glasburkar. Emhart Glass AB har i dagsläget 14 kontor i 10 olika länder och är sedan 1998 ägt av Bucher Industries. Den svenska divisionen är uppdelade i två enheter. I Örebro tillverkas maskindelar, reservdelar och strategiska komponenter. Komponenterna skickas sedan vidare till Sundsvall för del- och slutmontage. På Emhart Glass Sweden i Örebro har det uppmärksammats att långa ställtider förekommer vid vissa maskingrupper. Uppdraget består av att med hjälp av videoinspelningar och mjukvaran AviX analysera omställningsarbetet och komma med förbättringsförslag. Målet är att med hjälp av AviX reducera ställtiden och utvärdera huruvida AviX är ett lämpligt verktyg för det här sortens undersökning. Utvalda förbättringar ska sedan genomföras och utvärderas. Arbetet har utförts enligt SMED-metoden som är en välbeprövad metod för ställtidsreducering. Med hjälp av videoinspelningar kunde en grundlig nulägesanalys genomföras där alla ingående aktiviteter kunde identifieras. AviX användes sedan för att separera aktiviteterna och arrangera dem i den följd som gav den effektivaste omställningen. Den framtagna metoden standardiserades sedan genom rutinframtagning. Att standardisera utförandet ansågs vara ett krav för det kommande förbättringsarbetet då det är viktigt att samtliga operatörer utför omställningsarbetet på samma sätt. Förbättringsarbetet resulterade i ett antal förslag att förenkla och effektivisera omställningsarbetet för operatörerna. Utvalda förbättringar har genomförts och implementerats i omställningsarbetet. Förbättringarna tillsammans med de framtagna rutinerna uppskattas ge en ställtidsreducering mellan 20-30 %. Förbättringarna har gett operatörerna bättre förutsättningar att genomföra omställningarna snabbt och effektivt. Genom att genomföra fler förbättringsförslag kan ställtiden reduceras ytterligare. Emhart Glass rekommenderas att följa upp och föra statistik över omställningsarbetet och ställtiderna för att se hur dessa påverkats och arbeta med ständiga förbättringar för att effektivisera arbetet ytterligare. Vi anser att AviX är ett effektivt verktyg för att arbeta med ställtidsreducering och kan även användas i effektiviseringsarbetet för att leta fram och eliminera slöserier. Nyckelord: SMED, AviX, omställningsarbete, videoanalys, ställtidsreducering. / Emhart Glass AB is a leading international manufacturer of automatic glass forming machines for the manufacture of various glass products, among other things, glass bottles and jars. Emhart Glass AB has currently 14 offices in 10 countries and is since 1998 owned by Bucher Industries. The Swedish division is divided into two units. In Örebro manufactured machine parts, spare parts and strategic components. The components are then sent to Sundsvall for interim and final assembly. At Emhart Glass Sweden in Örebro has been discovered that long setup times occur at certain machine groups. The mission consists of using video recordings and the software AviX to analyze the changeover process and suggest improvements. The goal is that by using AviX reduce set-up time and evaluate whether AviX is a suitable tool for this sort of investigation. Featured improvements are then implemented and evaluated. The work has been performed according to the SMED-method which is a proven method for set-up time reduction. With the help of video recordings was a thorough situation analysis conducted in which all activities could be identified. AviX was then used to separate activities and arrange them in the order that gave the most efficient changeover. The developed method was standardized with new routines. To standardize the performance was considered to be a requirement for future improvement efforts because it is important that all operators perform changeovers in the same way. Improvement work resulted in a number of proposals to simplify and streamline the changeover process for operators. Featured improvements are in place and implemented in the changeover. These improvements along with the resulting procedures are estimated to provide a set-up time reduction between 20-30%. The improvements have given operators a better basis to carry out reorganization quickly and efficiently. By conducting more improvement suggestions set-up time is reduced even further. Emhart Glass is recommended to conduct a follow up and keep statistics on changeovers and set-up times to see how these are affected and making continuous improvements to improve efficiency further. We believe that AviX is an effective tool for working with set-up time reduction and can also be used to improve efficiency to seek out and eliminate waste.

No description available.

In recent years, due to the increasing usage of videos, manual information extraction is becoming insufficient to users. Therefore, extracting semantic information automatically turns out to be a serious requirement. Today, there exists some systems that extract semantic information automatically by using visual, auditory and textual data separately but the number of studies that uses more than one data source is very limited. As some studies on this topic have already shown, using multimodal video data for automatic information extraction ensures getting better results by guaranteeing increase in the accuracy of semantic information that is retrieved from visual, auditory and textual sources. In this thesis, a complete system which fuses the semantic information that is obtained from visual, auditory and textual video data is introduced. The fusion system carries out the following procedures / analyzing and uniting the semantic information that is extracted from multimodal data by utilizing concept interactions and consequently generating a semantic dataset which is ready to be stored in a database. Besides, experiments are conducted to compare results obtained from the proposed multimodal fusion operation with results obtained as an outcome of semantic information extraction from just one modality and other fusion methods. The results indicate that fusing all available information along with concept relations yields better results than any unimodal approaches and other traditional fusion methods in overall.