Browsing Mastergradsoppgaver i teknologi - informatikk by Author
DSpace/Manakin Repository
- Home
- Fakultet for naturvitenskap og teknologi
- Institutt for informatikk
- Mastergradsoppgaver i teknologi - informatikk
- Browsing Mastergradsoppgaver i teknologi - informatikk by Author
Browsing Mastergradsoppgaver i teknologi - informatikk by Author
Previous Page
Now showing items 23-30 of 30
-
Larsen, Webjørn Ivar (Master thesis; Mastergradsoppgave, 08-Jun-2010)[+][-]
Abstract: The duration of video playlists, might be a problem when watching on mobile devices, while on the go. We are exploring the problem of how to reduce the size of a video playlist, while maintain the important parts of the videoclips in the playlist. We have designed algorithms that automatically can cut down a playlist by cutting the duration of clips or sometimes even removing clips. We have shown that our algorithm is better than a naive solution that just cut at the end. Furthermore we have looked at playlist cutting for Davvi. This means playlist cutting for the soccer domain. We have created a hint based cutting algorithm that uses domain knowledge to achieve good results compared to the other algorithms available. URI: http://hdl.handle.net/10037/2636 File(s) in this item: 1
thesis.pdf (965.9Kb) -
Leer, Erik Bræck (Master thesis; Mastergradsoppgave, Jun-2011)[+][-]
Abstract: Recently, television broadcasters such as the NRK and TV2 channels, have begun offering live internet television and movie archive along with their regular schedule, much like the known video archives such as Youtube and Vimeo. The amount of all television offered reduces the ability of the user to get an overview of the programs that are available at any given time, making the user will probably miss important events. Regular indexing engines for recommending does not generally work on media since it is hard to index media data. Tags and keywords describing a media le does only describe the whole le, making it difficult to use them for indexing and recommendation of specific scenes within the media. This thesis presents a text event detection system for discovering interesting events based on video subtitles. By performing textual analytics, our system is able to discover events that are not discoverable through regular syntactic search. We have, based on related work, extended algorithms used for discovering semantic relationships between different words. Also, we have experimented with several algorithm for capturing the essence of each sentence, relating the prominent sense of each sense towards the events. Our experiments illustrates how we can increase the accuracy of the algorithm used by performing a context exploration based on event keywords. The results shows that our system improved the base algorithm of the prototype by including more relevance methods like consecutive sentence similarity and similarity based on sentence internals. URI: http://hdl.handle.net/10037/3521 File(s) in this item: 1
thesis.pdf (1.736Mb) -
Mortensen, Mats (Master thesis; Mastergradsoppgave, 15-Jun-2007)[+][-]
Abstract: Applications that adapt to environmental and situational changes are difficult to build because computers cannot capture, represent or process context information as easily as human beings. Nevertheless, context information is very valuable because it allows applications to be made more user-friendly, flexible, and adaptable. This realization has spawned a multitude of research efforts to simplify development of context-sensitive applications. A result of one of these research efforts is the Argos middleware platform, which is an application server created specifically for personal applications that can adapt to changes in their environment. Applications that rely on context information must often collect this information from external measurement devices, commonly known as sensors. These devices respond directly to physical stimulus to produce meaningful information about their surroundings. Typical examples are sensors that produce location, temperature or motion information, but they can also, for instance, be devices that monitors the physical condition of a person. The goal of this thesis has been to design, develop and evaluate functionality for the Argos middleware platform that makes it easier for Argos applications to collect and use sensor measurements. The functionality has been developed in collaboration with the National Center for Telemedicine (NST) who wants to use Argos for monitoring patients. They intend to develop a system that can give patients semi-automatic feedback and advice on how to maintain and improve their lifestyle. To do this they want to use personal sensors that monitor attributes relevant to a patient's condition. The functionality developed in this thesis has provided a starting point for NST to develop their system and contributed lots of technical information that will prove useful for their project. URI: http://hdl.handle.net/10037/1137 File(s) in this item: 2
cd.zip (100.9Mb)thesis.pdf (1.734Mb) -
Nilsen, Bjørn-Hugo (Master thesis; Mastergradsoppgave, 01-Oct-2007)[+][-]
Abstract: Current GPUs have many times the memory bandwidth and computing power compared to CPUs. The difference in performance is getting bigger as the evolution speed of the GPUs is higher than of the CPUs. This make it interesting to use the GPU for general purpose computing (GPGPU). I begin by looking at the architecture of the GPU, and some different techniques for programming on a GPU, including some of the available high-level languages. I have implemented the Mandelbrot computation on a cluster of GPUs (the HPDC display wall), and compared it against two different CPU implementations on the cluster. I have also implemented the Mandelbrot computation in both Cg and Brook, and compared the performance of the two languages. My experimental study shows that the GPU implementation of the Mandelbrot application is up to twice as fast as the load-balanced CPU implementation on the cluster of 28 computers, and up to 6 times faster on one computer. URI: http://hdl.handle.net/10037/1565 File(s) in this item: 1
thesis.pdf (920.3Kb) -
Nilsen, Karl Magnus (Master thesis; Mastergradsoppgave, 15-Dec-2003)[+][-]
Abstract: Software systems today are becoming larger and more complex, resulting in a growing need for good and efficient testing routines. An approach used by several software developers is to automate the test process. Test automation has the benefits that it reduces the time of the testing process and that automated tests are more accurate and precise than manual tests.
Manufacturers who wish to develop products using the Bluetooth technology, the Bluetooth logo and trademark has to go through a qualification program. This program is expensive, thus the manufacturer has incentives to make sure that the product is well tested before sending it to qualification. A Bluetooth stack is an example of a product that must be qualified. An automated tool for testing of Bluetooth stacks is therefore desired.
FAT is a framework that provides functionality to write and execute tests on a Bluetooth stack. The framework makes use of the ability to stitch generic test layers in-between the layers of the stack. These test layers can operate on messages passing through the stack. Our test layers provide an API to insert, modify, copy and delete messages. FAT introduces a test system client (TSC) where a tester can write tests and choose tests for execution. The tests are written in Java, where each test is a single method. The tester uses the test layer’s API to interface with the stack. The communication mechanism between the TSC and the test layer is XML-RPC. The TSC may therefore be executed on a different node than the stack itself. This thesis motivates FAT, and describes how the framework is designed and implemented.URI: http://hdl.handle.net/10037/233 File(s) in this item: 1
thesis.pdf (854.9Kb) -
Simonsson, Carl Erik Joakim (Master thesis; Mastergradsoppgave, 01-Jun-2010)[+][-]
Abstract: This thesis investigates how head tracking can be implemented by using inexpensive off-the-shelf hardware for a 6 x 3 meter high-resolution display wall. The tracking system has been integrated into to an existing event system, Shout, that allows for inter-program communication. An application called htsim has been developed that is used for testing different head tracking configurations in a virtual environment. Developing in a virtual environment does not require access to head tracking hardware. Tracking algorithms developed in the virtual environment can directly be used for head tracking in a physical environment. The tracking system is able to track a user's head with cameras that are placed behind the user. htsim is also used for configuring the head tracking system used in the physical environment. Experiments detail out the overall latency in the system and sources of jitter. URI: http://hdl.handle.net/10037/2634 File(s) in this item: 1
thesis.pdf (7.226Mb) -
Sundby, David (Master thesis; Mastergradsoppgave, 01-Jun-2011)[+][-]
Abstract: The widespread use of digital cameras and mobile phones, along with a rapidly growth of image sharing, challenges current image retrieval techniques. It is difficult for image retrieval system to find the semantic meaning of images based on human’s subjectivity and the size of current image databases makes it difficult to organize and search images. This thesis shows that information retrieval techniques can be used to reduce the search space of existing image collections, by creating collection summaries that holds only the most representative metadata from existing image collections. The representative metadata are metadata that are most distinguishing for the specific image collection. The system take advantage of the metadata available for images, which includes user provided tags, date/time, GPS coordinates, and metadata augmented by the system using already available metadata. Higher level representations of user provided terms are located and ensures that the system captures the most descriptive properties of the image collections. The system designed is able to produce collection summaries that capture properties of an image collection that support human’s natural perception as long as enough metadata are available. Also, the system increase the contextual understanding of images as long as date/time (of capture) and GPS coordinates is available for all or some of the images in the collection. The evaluation of the system indicates that grouping user provided terms into higher level representations is very useful for capturing the most important properties of an image collection. Also the evaluation expresses the usefulness of augmenting images with additional metadata and converting numeric metadata into readable terms. The comparisons made in the evaluation indicates that not only are the collection summaries similar to individual test users perception of image collections, but the summaries also includes more descriptive information of relevance. URI: http://hdl.handle.net/10037/3657 File(s) in this item: 2
source.zip (94.54Mb)thesis.pdf (3.098Mb) -
Thorvaldsen, Idar (Master thesis; Mastergradsoppgave, 06-Feb-2009)[+][-]
Abstract: There is a continuing desire and need for improving the processes of describing and searching for digital images. While good progress has been made adapting traditional information retrieval techniques to perform these tasks, processing images still presents a number of challenges not encountered when working with just text. This project implements a system allowing for the indexing and searching of collections of images where images are not individually described. Such collections are typically largely unsuitable data material for traditional text-based image search systems. The basic idea underlying this system is to not search for images directly, but instead first search for the collections they are part of. When a set of relevant collections have been found, one can then apply the collection-level semantics to the images belonging to these collections. An important concept in this project is that one image can be a part of several collections. Such images can be identified when collection information is added to the system, allowing one to link the same image to several collections before searches are run. Semantic information from several different collections can then be applied to the same image, potentially giving a better idea of what the image actually depicts. URI: http://hdl.handle.net/10037/1881 File(s) in this item: 2
CD.zip (372.7Mb)thesis.pdf (1.826Mb)
Previous Page
Now showing items 23-30 of 30
Search Munin
Browse
-
All of Munin
-
This Collection
Help
Munin is powered by DSpace 1.8.2
The University Library of Tromsø, N-9037 Tromsø
Tel: +47 77 64 40 00, E-mail: munin@ub.uit.no