next up previous contents
Next: Graduate education Up: Undergraduate education Previous: UU courses   Contents

Master theses

  1. Visual Planning and Verification of Deep Brain Stimulation Interventions
    Student: Elhassan M. Abdou
    Supervisor: Timo Ropinski, Dept. of Science and Technology, Linköping University
    Reviewer: Robin Strand
    Publisher: CBA Master Thesis No. 125 / IT nr 11 090
    Abstract: Deep Brain Stimulation (DBS) has resulted in a renaissance as an alternative way for treatment of Parkinson's disease and essential tremor. Deep brain stimulation employ the use of high electric field to stimulate some brain centers. The electric field in the brain is generated from chronic implanted electrodes in the brain. The final position of the electrodes in the brain is specified by the aid of CT and MR scans for the patient's head before and after the operation. A study of electric field distribution in the brain is required to interpret and improve the action of DBS. In this master thesis project, Voreen was extended to visualize a multimodal volume of the CT and MR images. The MR volume was segmented to extract the brain from the skull in T1 weighted images. Some image processing techniques were developed to enhance the contrast of CT and MR images. In order to stimulate electric field in the brain, the neurologists are allowed to design and position the electrodes in the reconstructed volume. The electrodes and some pre-modeled electric fields can be visualized in the reconstructed volume and the slice views. A mesh generator was developed using delaunay tetrahedralization. The generated mesh can be sent to PDE solver to solve Laplace equation describing the distribution of electric field.

  2. Digital Distance Functions Defined by Sequence of Weights
    Student: Alexander Denev
    Supervisor: Robin Strand
    Reviewer: Gunilla Borgefors
    Publisher: CBA Master Thesis No. 126 / IT nr 11 082
    Abstract: In this paper, digital distance functions using sequences of weights are studied and used to approximate the Euclidian distance. Sequences of weights that guarantee a low maximum absolute error for path lengths of up to 10000 are calculated. A necessary condition and a sufficient condition for metricity of this kind of distance function are established.

  3. Efficient Implementation of Polyline Simplification for Large Datasets and Usability Evaluation
    Student: Sadan Ekdemir
    Supervisor: Jörn Letnes
    Reviewer: Stefan Seipel
    Publisher: CBA Master Thesis No. 127 / IT nr 11 069
    Abstract: An in-depth analysis and survey of polyline simplification routines is performed within the project. The research is conducted using different simplification routines and performing evaluative tests on the outputs of each simplification routine. The project lies in between two major fields, namely Computer Graphics and Cartography, combining the needs of both sides and uses the algorithms that are developed for each field. After the implementation of the algorithms, a scientific survey is performed by comparing them according to the evaluation benchmarks, which are performance, reduction rate and visual similarity. Apart from the existing routines, one new simplification routine, triangular routine is developed and recursive Douglas-Peucker routine is converted into non-recursive. As a preprocessing part, Gaussian smoothing kernel is used to reduce noise and complexity of the polyline, and better performances are achieved. The end of research shows that there is no best model instead there are advantages and disadvantages of each simplification routine, depending on the prior need. It is also shown that usage of Gaussian smoothing as a filtering process improves the performance of each simplification routine.

  4. Image Analysis on Wood Fiber Cross-Section Images
    Student: Sitao Feng
    Supervisor: Bettina Selig
    Reviewer: Cris Luengo
    Publisher: CBA Master Thesis No. 128 / IT nr 11 028
    Abstract: Lignification of wood fibers has a significant impact on wood properties. To measure the distribution of lignin in compression wood fiber cross-section images, a crisp segmentation method had been developed. It segments the lumen, the normally lignified cell wall and the highly lignified cell wall of each fiber. In order to refine this given segmentation the following two fuzzy segmentation methods were evaluated in this thesis: Iterative Relative Multi Objects Fuzzy Connectedness and Weighted Distance Transform on Curved Space. The crisp segmentation is used for the multi-seed selection.

    The crisp and the two fuzzy segmentations are then evaluated by comparing with the manual segmentation. It shows that Iterative Relative Multi Objects Fuzzy Connectedness has the best performance on segmenting the lumen, whereas Weighted Distance Transform on Curved Space outperforms the two other methods regarding the normally lignified cell wall and the highly lignified cell wall.

  5. Audio Editing in the Time-Frequency Domain Using the Gabor Wavelet Transform
    Student: Ulf Hammarqvist
    Supervisor: Erik Wernersson
    Reviewer: Anders Brun
    Publisher: CBA Master Thesis No. 129 / UPTEC F nr F 11 022
    Abstract: Visualization, processing and editing of audio, directly on a time-frequency surface, is the scope of this thesis. More precisely the scalogram produced by a Gabor Wavelet transform is used, which is a powerful alternative to traditional techinques where the wave form is the main visual aid and editting is performed by parametric filters. Reconstruction properties, scalogram design and enhancements as well audio manipulation algorithms are investigated for this audio representation. The scalogram is designed to allow a flexible choice of time-frequency ratio, while maintaining high quality reconstruction. For this mean, the Loglet is used, which is observed to be the most suitable filter choice. Re-assignmentare tested, and a novel weighting function using partial derivatives of phase is proposed. An audio interpolation procedure is developed and shown to perform well in listening tests.The feasibility to use the transform coefficients directly for various purposes is investigated. It is concluded that Pitch shifts are hard to describe in the framework while noise thresh holding works well. A downsampling scheme is suggested that saves on operations and memory consumption as well as it speeds up real world implementations significantly. Finally, a Scalogram ``compression'' procedure is developed, allowing the caching of an approximate scalogram.

  6. Enriching Circuit Switched Mobile Phone Calls with Cooperative Web Applications
    Student: Måns Hommerberg
    Supervisor: Johan Kristiansson
    Reviewer: Olle Eriksson
    Publisher: UPTEC F nr 11051
    Abstract: The thesis investigates the possibility to enrich standard mobile phone calls with cooperative web applications. Originating from the research field know as Computer Supported Cooperative Work (CSCW) this thesis report introduces and describes the implementation of several applications which can be used by the calling parties together during a phone call. Additionally, the report describes a proof-of-concept prototype for the Android platform, and discusses the performance of cooperative web application running on mobile devices in terms of network and CPU use.

    The conclusions of the thesis describe a prototype application addressing and implementing the requirements as described by the theory of computer supported collaborated work. The performance of the running application showed to be satisfactory, both regarding to network demand and processor use.

  7. Registration of 3D Volumetric CT Images
    Student: Shuo Li
    Supervisor: Erik Wernersson
    Reviewer: Anders Brun
    Publisher: CBA Master Thesis No. 130 / IT nr 11 080
    Abstract: This master thesis aims to develop a system for analyzing transformation between two volumetric CT images. The volumetric image data we process is taken from a composite material. This composite material combines wood fibre and plastic and can be used to make for instance hockey sticks or furniture. Because of the wood fibre embedded in this composite material, it absorbs water and sometimes deforms. By observing volumetric images generated by micro computed tomography (micro-CT), we know that the organization of fibre embedded in this material is very complicated. This makes it difficult to predict the deformation on beforehand. In our study, we have seen rigid transformations, non-rigid transformations and even discontinuities transformations (cracks). For a pair of very small sub volumes, in dry and wet condition, we have found that the transformation can approximated by a rigid transformation combined with a scaling value. To find this transformation, our system includes two key phases. In the first phase, we extract feature points in dry and wet condition. In the second phase, we register the feature points derived from dry and wet condition. In the feature point extraction phase, we have adapted different methods, for instance the Scale- Invariant Feature Transform (SIFT) method is used to extract features. In the registration phase, we have tested three different registration algorithms. The first algorithm is based on concepts from Random Sample Consensus (RANSAC). The second algorithm is inspired from the Iterative Closest Point (ICP) method. The third method is a novel algorithm that we call Spatial Invariant Registration. In the report, we compare the different methods in the feature extraction phase and in the registration phase. Finally, we discuss how our system can be extended to give better results with better accuracy.

  8. Evaluation of a Model-free Approach to Object Manipulation in Robotics
    Student: Guoliang Luo
    Supervisors: Danica Kragic, Carl-Henric Ek, Royal Institute of Technology
    Reviewer: Anders Brun
    Publisher: CBA Master Thesis No. 131 / IT nr 11 035
    Abstract: Action Recognition is crucial for object manipulation in robotics. In recent years, Programming by Demonstration has been proposed as a way for a robot learning tasks from human demonstrations. Based on this concept, a model-free approach for object manipulation has recently been proposed in [1]. In this thesis, this model-free approach is evaluated for Action Recognition. In specific, the approach classifies actions by observing object-interaction changes from video.

    Image segmentation to videos presents various difficulties, such as motion blur, complex environment, Over- and Under- segmentation. This thesis investigates and simulates these image segmentation errors in a controllable manner. Based on the simulation, two different similarity measure methods are evaluated: The Substring Match (SSM) and Bhattacharyya Distance (B-Distance) method. The results show that the B-Distance method is more consistent and capable to classify actions with higher noise level compare to the SSM method.

    Further, we propose an action representation using kernel method. The evaluation shows that the novel representation improves Action Recognition rate significantly.

  9. Rendering Software for Multiple Projectors
    Student: Fredrik Nysjö
    Supervisor: Robin Strand
    Reviewer: Ingrid Carlbom
    Publisher: CBA Master Thesis No. 132 / IT nr 11 081
    Abstract: CBA is currently developing a haptic glove that will be integrated with a new type of 3D holographic display. A prototype of this display has been developed at the Royal Institute of Technology and consists of standard off-the-shelf projectors and a special holographic optical element. The latter acts as a projection screen that creates a narrow viewing slit for each projector's image; this allows for autostereoscopic viewing from more than one viewing angle. The rendering software for the display prototype at the Centre for Image Analysis can render a fixed number of static perspective views of virtual 3D scenes. But the software's rendering pipeline was not properly calibrated for the intended application of the display: co-located haptics. Thus, the views are rendered without proper off-axis projection matrices, and they also exhibit keystone distortion from oblique projector angles when they are projected on the holographic optical element. In this master's thesis work, we develop a software library that extends the existing rendering software with support for keystone correction and arbitrary off-axis projections. We use this library to calibrate the rendering pipeline and the display. We also develop software tools that facilitate the calibration task. Furthermore, when views are rendered with static perspective, a viewer perceives a discrete transition between two distinct perspectives whenever he or she moves an eye from one viewing slit to an adjacent slit. To make these view transitions smooth and reduce other types of perspective errors, we couple the display with an optical tracking system for head-tracking and experiment with adding dynamic perspective to the display. We conclude that while the addition of dynamic perspective helps reduce perspective errors, the display would need narrower viewing slits in order to allow smooth view transitions.

  10. Orbit Segmentation for Cranio-Maxillofacial Surgery Planning
    Student: Johan Nysjö
    Supervisor: Ingela Nyström
    Reviewer: Ewert Bengtsson
    Publisher: CBA Master Thesis No. 124 / IT nr 11 010
    Abstract: A central problem in cranio-maxillofacial (CMF) surgery is to restore the normal anatomy of the facial skeleton after defects, e.g., malformations, tumours, and trauma to the face. There is ample evidence that careful pre-operative surgery planning can significantly improve the precision and predictability of CMF surgery as well as reduce the post-operative morbidity. In addition, the time in the operating room can be reduced and thereby also costs. Of particular interest in CMF surgery planning is to measure the shape and volume of the orbit (eye socket), comparing an intact side with an injured side. These properties can be measured in 3D CT images of the skull, but in order to do this, we first need to separate the orbit from the rest of the image - a process called segmentation.

    Today, orbit segmentation is usually performed by experts in CMF surgery planning who manually trace the orbit boundaries in a large number of CT image slices. This manual segmentation method is accurate but time-consuming, tedious, and sensitive to operator errors. Fully automatic orbit segmentation, on the other hand, is unreliable and difficult to achieve, mainly because of the high shape variability of the orbit, the thin nature of the orbital walls, the lack of an exact definition of the orbital opening, and the presence of CT imaging artifacts such as noise and the partial volume effect.

    The outcome of this master's thesis project is a prototype of a semi-automatic system for segmenting orbits in CT images. The system first extracts the boundaries of the orbital bone structures and then segments the orbit by fitting an interactive deformable simplex mesh to the extracted boundaries. A graphical user interface with volume visualization tools and haptic feedback allows the user to explore the input CT image, define anatomical landmarks, and guide the deformable simplex mesh through the segmentation.

    To evaluate the performance of our segmentation system, we let three test users independently segment 14 orbits twice (in a set of seven CT images) with the segmentation tools provided by the system. In order to assess segmentation accuracy, we construct crisp and fuzzy ground truth segmentations from manual orbit segmentations performed by the three test users. The results of this case study indicate that our segmentation system can be used to obtain fast and accurate orbit segmentations, with high intra-operator and inter-operator precision.

  11. Scalable Web Application using Node.JS and CouchDB
    Student: Umesh Paudyal
    Supervisor: Claudijo Borovic
    Reviewer: Olle Eriksson
    Publisher: IT nr 11 066
    Abstract: This report presents design and implementation of a protoype application using server side javascript programming language, node.JS and couchDB as backend database. It scales and evaluates the developed prototype application and the couchDB for their scalability and performance.

    The report concludes that node.JS is a suitable framework for development of scalable web servers and couchDB as a backend database, though natively not distributed and scalable, can be scaled and distributed across multiple nodes using clustering and replication mechanism.

  12. Mobile Application Development for Android - Solving Complex Debt Situations
    Students: Alexander Sjöberg and Emil Larsson
    Supervisor: Tom Smedsaas
    Reviewer: Olle Eriksson
    Publisher: TVE nr 11 007
    Abstract: The goal of the project has been to develop an Android application whose function is to calculate the necessary transactions, minimized in number, to resolve a complex debt situation within a group of individuals. These types of situations frequently occur in everyday life, for example when a group of friends cook dinner together and different people pay for various expenses such as food and beverages. The work resulted in the application SplitIt, a stylish and easy-to-use application that meets the desired specifications. Uncertainties exist however, whether the algorithm thatcalculates the transactions is optimized regarding the minimum number oftransactions required. Some measures should be taken before the product is launched on the Android Market. The development of icons, for example, has been put to the side with the intention to spend more time on other parts of the user interface and algorithm development. Splitit has been developed by studying similar applications on the Android Market and by carefully considering usability. Before starting the implementation of the application, a user study was conducted in which sketches of the proposed user interface was designed and a test panel had the opportunity to navigate through the application. The study clarified unclear as well as appealing parts of the user interface.

  13. Image Analysis for Grain Quality Assessment
    Student: Fraz Ali
    Supervisor: Jaan Luup, Maxx automation AB
    Revierer: Cris Luengo
    Publisher: IT nr 11 004 (confidential report)
    Abstract: Assessing grain quality is a critical task to ensure that the grain-based products meet the food industry standards. Due to the complex texture-based symptoms, this process is carried out manually by quality assurance staff. To overcome the expense and inconsistencies of the process, an automated solution for grain quality is desirable.

    For decades, researchers have been trying to improve the automated analysis of grains, and many image-based solutions for grain sorting have been proposed. However, none of these solutions is reliable and fast enough. Hence, the grain quality assessment is still performed manually. Recent advancement in computer vision systems and rapid progress in computer hardware industry demands that new efforts should be made to automate the complex task of grain quality assessment.

    To develop an image-based solution for grain quality requires deep understanding of the symptoms as well as efficient image analysis techniques, to meet both the accuracy and performance requirements.

    The purpose of this thesis is to investigate and develop efficient image analysis algorithms for grain quality assessment. An existing grain sorting system is used to acquire images. A set of image-based solutions is developed. In most of the algorithms high accuracy is achieved. A machine based on these solutions will be developed in future.

next up previous contents
Next: Graduate education Up: Undergraduate education Previous: UU courses   Contents