Dada uma biblioteca com milhares de vídeos de histeroscopias diagnósticas, sobre a qual deseja-se realizar consultas como "retornar imagens contendo miomas submucosos" ou "recuperar imagens cujo diagnóstico é pólipo endometrial". Este é o contexto deste trabalho. Vídeos de histeroscopias diagnósticas são usados para avaliar a aparência do útero e são importantes não só para propósitos de diagnóstico de doenças mas também em estudos científicos em áreas da medicina, como reprodução humana e estudos sobre fertilidade. Estes vídeos contêm uma grande quantidade de informação, porém somente um número reduzido de quadros são úteis para propósitos de diagnósticos e/ou prognósticos. Esta tese apresenta um método para identificar automaticamente a informação relevante em vídeos de histeroscopias diagnósticas, criando um sumário do vídeo. Propõe-se uma representação hierárquica do conteúdo destes vídeos que é baseada no rastreamento de pontos geometricamente consistentes através da seqüência dos quadros. Demonstra-se que esta representação é uma maneira útil de organizar o conteúdo de vídeos de histeroscopias diagnósticas, permitindo que especialistas possam realizar atividades de browsing de uma forma rápida e sem introduzir informações espúrias no sumário do vídeo. Os experimentos indicam que o método proposto produz sumários compactos (com taxas de redução de dados em torno de 97.5%) sem descartar informações clinicamente relevantes. / Given a library containing thousands of diagnostic hysteroscopy videos, which are only indexed according to a patient ID and the exam date. Usually, users browse through this library in order to obtain answers to queries like retrieve images of submucosal myomas or recover images whose diagnosis is endometrial polyp. This is the context of this work. Specialists have been used diagnostic hysteroscopy videos to inspect the uterus appearance, once the images are important for diagnosis purposes as well as in medical research fields like human reproduction. These videos contain lots of information, but only a reduced number of frames are actually useful for diagnosis/prognosis purposes. This thesis proposes a technique to identify clinically relevant information in diagnostic hysteroscopy videos, creating a rich video summary. We propose a hierarchical representation based on a robust tracking of image points through the frame sequence. We demonstrate this representation is a helpful way to organize the hysteroscopy video content, allowing specialists to perform fast browsing without introducing spurious information in the video summary. The experimental results indicate that the method produces compact video summaries (data-rate reduction around 97.5%) without discarding clinically relevant information.

I de svenska skogarna är tallen (Pinus sylvestris L.) det näst vanligaste trädslaget och konkurrerar bäst på torrare och blötare marker. Av den anledningen är det viktigt vid föryngring att ståndortsanpassa. Det har dock blivit allt vanligare att privata skogsägare planterar gran (Picea abies L.) på typiska tallmarker på grund av det höga betestrycket. Detta kan vara förödande för tillväxten och kvaliteten. Ett sätt att klara betestrycket är att ha en högstamtäthet för att få många alternativa stammar kvar till älgsäker höjd. Där kan föryngringsmetoden vara avgörande. Mellan 2013–2016 planterades och självföryngrades 92,2 % av tallföryngringsarealen i Sverige, de resterande 7,8 % såddes. Att så tall på lämpliga marker resulterar ofta i en hög stamtäthet och det finns möjlighet att välja förädlat material. Detta ger en trängseleffekt vilket bidrar till hög virkeskvalitet, samtidigt som det kan ge en tillväxtökning upp till 25 % gentemot beståndsfrön om förädlat material används. Detta är en kombination som bara är möjlig att åstadkomma med sådd. Studiens syfte har varit att undersöka vilken inställning skogsägare har haft till tallsådd och varför så få använt metoden. Dessutom har det kartlagts vilka som använt metoden och varför. Detta har gjorts med en enkätundersökning som skickades ut till 240 skogsägare jämt fördelat över Sverige, varav 103 svarade. Det har även gjorts intervjuer med skogsbolag med eget skogsinnehav från nord- och mellansverige. Resultatet visade att 15 % av respondenterna från enkätundersökningen någon gång sått tall. Det var vanligare att de som ägde en skogsfastighet på 100 hektar eller mer någon gång hade sått tall i jämförelse med de som ägde 49 hektar eller mindre. Av de som sått svarade 92 % att de var nöjda med resultatet och 73 % kunde tänka sig att så igen. Det var ingen skogsägare från Götaland som sått tall. Av de som inte hade sått tall svarade 48 % att anledningen till de inte gjort de var för att de inte tänkt på det och 69 % svarade att de hade en varken positiv eller negativ inställning till metoden. Detta kan jämföras mot de som någon gång hade sått tall där 86 % svarade att de hade en positiv eller ganska positiv inställning. De skogsbolag som låtit sig intervjuas svarade samtliga att de har sått tall, dock i olika omfattning. Det bolag som sått mest var Sveaskog som svarade att de har sått 27 % av deras tallföryngringsareal. Detta kan jämföras med Holmen skog som sådde strax över 20 % och då är contortatall inkluderad. De främsta anledningarna till att de har sått var för att det bedömdes vara kostnadseffektivt och ger många stammar per hektar vilket har varit positivt i områden med högt betestryck. Alla intervjupersoner berättade att det är viktigt att välja rätt mark för ett positivt resultat. Resultatet från enkäten visade att de som svarat att de någonsin sått tall hade en positivare inställning till sådd, vilket tyder på positiva erfarenheter. Nära hälften av de som inte sått tall svarade att de hade varken en positiv eller negativ inställning. En möjlig orsak till det är att de inte hade någon erfarenhet av metoden och därför ingen åsikt. De som någon gång har sått var i regel större skogsägare. En möjlig orsak till det är att det är större sannolikhet att de har en lämplig mark tillgänglig och att de troligen föryngrar oftare än mindre skogsägare. En hög andel av de som inte sått svarade att de inte tänkt på det. Detta skulle kunna bero på ointresse eller okunskap. En stark orsak till detta är sannolikt på grund av tradition. Skulle de privata skogsägarna skaffa sig kunskap om detta och vilka marker som är lämpliga skulle det kunna finnas pengar att tjäna. Detta eftersom det är billigare, lättare, och skapar bättre kvalitet än de andra föryngringsmetoderna. Tallsådd går dessutom bra att kombinera med andra metoder och anpassa föryngringsåtgärden efter förutsättningarna. I områden där betesskador ställer till stora problem kan tallsådd vara ett outnyttjat verktyg i verktygslådan. / The aim of this study was to investigate the attitudes of forest owners to direct sowing of pine (Pinus sylvestris L.) and also why so few have used the method. The study also aimed to identify reasons for forest owners to use or not to use the method. A questionnaire was sent to 240 non-industrial private forest owners (NIPFs) all over Sweden. As a complement, interviews were performed with forest companies from the north and middle parts of Sweden. The results showed that 15 % of the NIPFs respond that they had used the method direct sowing of pine at some occasion in the past. No forest owner origination from Götaland had used the method though. It was more common among those who owned a forest property of 100 hectares or more than among those who owned 49 hectares or less. Most of those who had used direct sowing in the past were satisfied with the outcome of the regeneration, 92 %, and 73 % were willing to use the method again. They also showed a more positive attitude to direct sowing, 86 %, compared to those who had not tried the method, where 23 % were positive and 69 % were neither positive nor negative. The main reason for not have used the method showed to be that the NIPFs answering this questionnaire had not thought about it. Of the companies interviewed in this study, Sveaskog and Holmen were the ones that have used direct sowing the most: Sveaskog sow 27 % of the yearly pine regeneration area and Holmen just over 20 % of the yearly pine regeneration area, including lodge pole pine (Pinus contorta Douglas ex Loudon). The main reason to why the companies have used direct sowing was that they consider the method cost effective. The also appreciate that the method results in a lot of stems per hectare which has been positive in areas where browsing is a problem. The results from the questionnaire showed that those had sown pine in the past tended to have a more positive attitude to the method than the forest owner that had not sown. This indicates that they have had a positive experience of the method. Nearly half of those who had not sown had neither a positive nor a negative attitude to it. One possible reason for it is that they had no experience of the method and therefore no opinion. Direct sowing works well to combine with other methods and can be adapted to different conditions. In a time when damage from browsing can cause big problems to regenerations of pine, sowing could be an important tool in the toolbox.

Dada uma biblioteca com milhares de vídeos de histeroscopias diagnósticas, sobre a qual deseja-se realizar consultas como "retornar imagens contendo miomas submucosos" ou "recuperar imagens cujo diagnóstico é pólipo endometrial". Este é o contexto deste trabalho. Vídeos de histeroscopias diagnósticas são usados para avaliar a aparência do útero e são importantes não só para propósitos de diagnóstico de doenças mas também em estudos científicos em áreas da medicina, como reprodução humana e estudos sobre fertilidade. Estes vídeos contêm uma grande quantidade de informação, porém somente um número reduzido de quadros são úteis para propósitos de diagnósticos e/ou prognósticos. Esta tese apresenta um método para identificar automaticamente a informação relevante em vídeos de histeroscopias diagnósticas, criando um sumário do vídeo. Propõe-se uma representação hierárquica do conteúdo destes vídeos que é baseada no rastreamento de pontos geometricamente consistentes através da seqüência dos quadros. Demonstra-se que esta representação é uma maneira útil de organizar o conteúdo de vídeos de histeroscopias diagnósticas, permitindo que especialistas possam realizar atividades de browsing de uma forma rápida e sem introduzir informações espúrias no sumário do vídeo. Os experimentos indicam que o método proposto produz sumários compactos (com taxas de redução de dados em torno de 97.5%) sem descartar informações clinicamente relevantes. / Given a library containing thousands of diagnostic hysteroscopy videos, which are only indexed according to a patient ID and the exam date. Usually, users browse through this library in order to obtain answers to queries like retrieve images of submucosal myomas or recover images whose diagnosis is endometrial polyp. This is the context of this work. Specialists have been used diagnostic hysteroscopy videos to inspect the uterus appearance, once the images are important for diagnosis purposes as well as in medical research fields like human reproduction. These videos contain lots of information, but only a reduced number of frames are actually useful for diagnosis/prognosis purposes. This thesis proposes a technique to identify clinically relevant information in diagnostic hysteroscopy videos, creating a rich video summary. We propose a hierarchical representation based on a robust tracking of image points through the frame sequence. We demonstrate this representation is a helpful way to organize the hysteroscopy video content, allowing specialists to perform fast browsing without introducing spurious information in the video summary. The experimental results indicate that the method produces compact video summaries (data-rate reduction around 97.5%) without discarding clinically relevant information.

The purpose of this paper is to explore web use and web navigation behavior. This paper presents findings from qualitative research based on interviews and diarie studies. The basis of this study is Eli Parisers work about “the filter bubble”. As personalization systems shape our experience of the web, based on who we are, our bubble of knowledge and information becomes smaller. But it’s not only how technology works to show us a narrow part of reality. Our own choices about websites, content and the activities on the web, lead us into another type of bubble. A behavior bubble. A bubble where all the content we consume is interesting. The conclusion of this paper is that narrow web navigation behavior can be observed among all subjects studied irrespective of age and devices used, whereas monotonous web navigation behavior only can be identified among individuals using a smartphone for web browsing.

With the evolution of Internet, users exploit it increasingly intensively for achieving their goals: sending emails, playing games, watching videos, reading news. They do it through Web browsers, that may vary in exact implementation, but have similar core functionalities . One of these core functionalities is the access to a browsing history. However, as browsing patterns are getting more complex, the traditional history tools become insufficient. Visualization of browsing history might be helpful in that case. In this study we propose a novel approach for browsing history visualization, named HistoryLane, which ts the parallel browsing paradigm, common for modern browsers. The main goal of HistoryLane is enabling the user to gain insight into his own or into other users' parallel browsing patterns over time. Principles of HistoryLane visualization approach are formulated based on recommendations, found during structured literature review. These principles constitute the base for a prototype, which was implemented as a Fire- fox extension. To evaluate the e ffectiveness of HistoryLane we conducted a survey and a quantitative experiment. The results of the evaluation show that HistoryLane is perceived by users as effective and intuitive method for browsing history visualization.

Web browsing has been one of the most popular activities on the internet. The increasing importance of the Web in everyday life calls for device-independent access to existing web sites. Although, there is full Web access available on mobile phones the user experience is often poor when compared to the Web user experience on Personal Computer (PC). Therfore, it is important for the Internet service provider to find the indications of user dissatisfaction from the network. While using the mobile Web, user can normally abort the transfer by pressing the stop or reset buttons in the browsers, leaving the page being downloaded by following a bookmark or can close the connection. Such events can be observed through the TCP reset (RST) flag from traffic on network level. In this thesis we have investigated the possible causes of TCP RST flags mobile Web session is interrupted. We further analyze up to what extent we can rely on TCP RST flags for being an indication of user dissatisfaction in mobile web browsing. Therefore, an experiment testbed is developed to capture the TCP packet traces during controlled active tests. Results are gathered using mobile devices with four popular mobile operating systems (OS). The trace files captured are analyzed using perl script to dicern the flow and focusing on the TCP RST flag during the flow. Further, TSTAT tool is used to validate our trace files. / 0760-882089

Målet med detta examensarbete är att ta fram ett gränsnitt för navigering i stora datamängder samt undersöka eventuella styrkor och svagheter hos detta gränsnitt. Bakgrunden till projektet är att företaget Analog Software utvecklar ett system som automatiskt skall klassificera objekt i en datamängd och sedan presentera denna klassificerade datamängd. Gränsnittet skall stödja en informationssökningsstrategi som benämns ”browsing”. En teknik för visualisering av information som varit framgångsrik är den så kallade ”Fisheye-tekniken” som baseras på en analogi med en vidvinkellins. Tekniken visar lokal information med hög detaljrikedom samtidigt som den visar global information med låg detaljrikedom. Ett flertal gränssnitt har använt sig av denna teknik för visualisering av information. I detta arbete har ett textbaserat Fisheye-gränssnitt tagits fram och testat med avseende på två olika funktioner. De två funktionerna är ”anomalier” och ”stöd för bokmärken”. Hypoteserna var: 1. Om anomalier fanns i datastrukturen så skulle en Fisheye-teknik förstärka dessa. 2. En bokmärkesfunktion skall ej behövas då Fisheye-tekniken erbjuder global kontext och byte av fokus går snabbt. Två experiment utfördes med 28 försökspersoner. I experiment 1 använde hälften av försökspersonerna ett standargränssnitt och hälften ett Fisheye-gränssnitt. De fick navigera i en datamängd sorterad på två olika sätt och ange om det upplevde sorteringen som naturlig eller konstig. I experimet 2 ingick samma försökspersoner som i experiment 1. I detta experiment fick hälften navigera med ett Fisheye-gränsnitt med bokmärkesfunktion och hälften med ett Fisheye-gränssnitt utan bokmärkesfunktion. De skulle navigera i en datamängd och svara på ett antal frågor. Resultaten visade att Hypotes 1 kunde förkastas. Försökspersonerna som använde sig av Fisheye-gränsittet upplevde datamängden att vara mer naturlig än den grupp som använt sig av ett standardgränssnitt. För hypotes 2 kunde inga säkra slutsatser dra på grund av att experimentet inte lyckades mäta det som avsågs att mätas.

Abstract Growth in storage capacity has led to large digital video repositories and complicated the discovery of specific information without the laborious manual annotation of data. The research focuses on creating a retrieval system that is ultimately independent of manual work. To retrieve relevant content, the semantic gap between the searcher's information need and the content data has to be overcome using content-based technology. Semantic gap constitutes of two distinct elements: the ambiguity of the true information need and the equivocalness of digital video data. The research problem of this thesis is: what computational content-based models for retrieval increase the effectiveness of the semantic retrieval of digital video? The hypothesis is that semantic search performance can be improved using pattern recognition, data abstraction and clustering techniques jointly with human interaction through manually created queries and visual browsing. The results of this thesis are composed of: an evaluation of two perceptually oriented colour spaces with details on the applicability of the HSV and CIE Lab spaces for low-level feature extraction; the development and evaluation of low-level visual features in example-based retrieval for image and video databases; the development and evaluation of a generic model for simple and efficient concept detection from video sequences with good detection performance on large video corpuses; the development of combination techniques for multi-modal visual, concept and lexical retrieval; the development of a cluster-temporal browsing model as a data navigation tool and its evaluation in several large and heterogeneous collections containing an assortment of video from educational and historical recordings to contemporary broadcast news, commercials and a multilingual television broadcast. The methods introduced here have been found to facilitate semantic queries for novice users without laborious manual annotation. Cluster-temporal browsing was found to outperform the conventional approach, which constitutes of sequential queries and relevance feedback, in semantic video retrieval by a statistically significant proportion.

Although much research exists on human behavior in online environments, research on users with disabilities is still rare. To draw more attention to this population, this dissertation explored browsing patterns and adaptive behaviors of people with visual disability across different online environments common in daily activities: social network, e-commerce, online information, and search engines’ websites. The main objective of this study is to propose a conceptual framework of how blind and visually impaired users browse and adapt to different web environments. We achieve this objective using a qualitative approach through three studies. In the first study, the researchers collect data by means of in-depth, semi-structured interviews with six users with different levels of visual impairment. In the second study, we use survey questionnaires with open-ended questions to reach a larger sample of study participants. Finally, we conduct a follow-up observational study as means to confirm our results. Open, axial, and selective coding are used for data reduction and analysis as part of the grounded theory method.

The purpose of the study was to find alternatives to fences by comparing the frequency of damages on different configurations of chemical treated, mechanically protected and untreated hybrid aspen seedlings, planted on agricultural land. The study also explored whether anthropogenic disturbance (proximity to roads and buildings) had any effect on the frequency of seedlings damages. The study was conducted as a quantitative study with field measurements over the course of a year, in two sample plots with hybrid aspen seedlings planted in May 2016 in Vimmerby kommun, Kalmar län. The field data was compiled into Excel and compared for differences between treatments and proximity to anthropogenic disturbance. The result showed that seedlings closer to anthropogenic disturbance showed a lower frequency of damage. There was no significant difference between the mechanical (Taimitassu) and chemical (Arbinol B) seedling protection. The highest frequency of damages occurred in the summer, peaking in July. In conclusion, when choosing a place to plant hybrid aspen, there should be anthropogenic disturbance to minimize the frequency of browsing.