Chapter 2 describes the concept of canonical correlation. This you have to know about in order to understand the continuing discussion. Chapter 3 introduce you to the problem that was to be solved. Chapter 4, 5 and 6 discusses three different suggestions how to approach the problem. Each chapter begins with a section of experiments as a motivation of the approach. Then follows some theory and mathematical manipulations to structure the thoughts. The last sections contains discussions and suggestions concerning the approach. Finally chapter 7 contains a summary and a comparismental discussion of the approaches.

Segmentation is a process that separates objects in an image. In medical images, particularly image volumes, the field of application is wide. For example 3D visualisations of the anatomy could benefit enormously from segmentation. The aim of this thesis is to construct a segmentation tool. The project consist three main parts. First, a survey of the actual need of segmentation in medical image volumes was carried out. Then a unique three-step model for a segmentation tool was implemented, tested and evaluated. The first step of the segmentation tool is a seed-growing method that uses the intensity and an orientation tensor estimate to decide which voxels that are part of the project. The second step uses an active contour, a deformable “balloon”. The contour is shrunk to fit the segmented border from the first step, yielding a surface suitable for visualisation. The last step consists of letting the contour reshape according to the orientation tensor estimate. The use evaluation establishes the usefulness of the tool. The model is flexible and well adapted to the users’ requests. For unclear objects the segmentation may fail, but the cause is mostly poor image quality. Even though much work remains to be done on the second and third part of the tool, the results are most promising.

This thesis investigates the possibilities of using GIS (Geographic Information System) data with an airborne autonomous vehicle developed in the WITAS project. Available for the thesis are high resolution (0.16 meter sample interval) aerial photographs over Stockholm, and vector data in a common GIS format containing all roads in the Stockholm area. A method for removing cars from aerial photographs is presented, using the filtering method normalized convolution, originally developed for filtering uncertain and incomplete data. By setting the certainty to zero over the cars, this data is disregarded in the filtering process, resulting in an image without cars. This method is further improved by choosing an anisotropic applicability function, resulting in a filtering that preserves structures oriented in certain directions. The available vector data is investigated with regard to its use in a simulator for vehicle movement, and is found to be missing much of the essential information needed in such a simulator. A new data format better suited to these requirements is created, using the extensible markup language (XML) which generates a humanreadable data format and can use existing parsers to make the implementation simpler. The result is a somewhat complex, but highly general data format that can accurately express almost any type of road and intersection. Cars can follow arbitrary paths in the road database and move with a smooth motion suitable for use as input to image processing equipment. The simulator does not allow any dynamic behaviour such as changing speeds, starting or stopping, or interaction between cars, takeovers or intelligent behavior in intersections. In the airborne vehicle, a mapping from pixels in a camera image (like the ones output from the simulator) to locations in the road database is needed. This is an inverse mapping with respect to visualizing as described above. This gives important information to a car tracking system regarding the probable movement of cars and also making it possible to determine if a car breaks traffic regulations. A mapping of this kind is created using a simplified form of ray tracing known as ray casting, together with space partitioning methods used to vastly improve efficiency. All above mentioned tasks are implemented using C++ and object oriented methods, giving maintainable and extendable code suiting a quickly changing research area. The interface to the simulator is designed to be compatible to the existing simulation software used in the WITAS project. Visualization is done through the OpenGL graphics library, providing realistic effects such as lighting and shading.

We explore the use of colour for interpretation of unstructured off-road scenes. The aim is to extract driveable areas for use in an autonomous off-road vehicle in real-time. The terrain is an unstructured tropical jungle area with vegetation, water and red mud roads. We show that hue is both robust to changing lighting conditions and an important feature for correctly interpreting this type of scene. We believe that our method also can be deployed in other types of terrain, with minor changes, as long as the terrain is coloured and well saturated. Only 2D information is processed at the moment, but we aim at extending the method to also treat 3D information, by the use of stereo vision or motion.

This Master's Thesis discusses the different trade-offs a programmer needs to consider when constructing image processing systems. First, an overview of the different alternatives available is given followed by a focus on systems based on general hardware. General, in this case, means mass-market with a low price-performance-ratio. The software environment is focused on UNIX, sometimes restricted to Linux, together with C, C++ and ANSI-standardized APIs.

The aim of this master thesis is to classify the tree class from an image of a leaf with a computer vision classiffication system. We compare different descriptors that will describe the leaves different features. We will also look at different classiffication models and combine them with the descriptors to build a system hat could classify the different tree classes.

IT systems rely on correct authentication of their users in order to provide confidentiality and integrity of data. When accessing systems remotely, for instance over the Internet, no assumptions can be made regarding the level of security on the computer used. Such computers may be exposed to malware, keyloggers and other threats and must therefore generally be considered as untrusted. To increase security when users connect remotely from untrusted clients various authentication mechanisms can be used. Usability must however be considered when deploying new mechanisms. Protection must also be balanced to the load put on users. This thesis gives a presentation of common authentication mechanisms available and enumerates the main attack vectors threatening correct authentication and credentials. Furthermore a ranking method is proposed in order to evaluate authentication mechanisms in relation to each other. Using the outcome of the ranking of existing methods an authentication system called Smokey (Synchronizable Mobile Key) is proposed and implemented. Smokey uses Java capable cell phones as hardware tokens generating short time valid one time passwords. Whereas traditional tokens may cease to work under certain circumstances Smokey provides users the ability to synchronize with the authentication server aiming for high usability.

Around the world, vehicle emission regulations become stricter, increasing exhaust emission demands. To manage these rules and regulations, vehicle manufacturers put a lot of effort into minimizing the exhaust emissions. The three-way catalytic converter was developed, and today it is the most commonly used device to control the exhaust emissions. To work properly the catalytic converter needs to control the air-fuel mixture with great precision. This then increases the demands on the engine management systems, causing them to become more complex. With increased complexity, the time effort of optimizing parameters has grown drastically, hence increasing development costs. In addition to this, operating conditions change due to vehicles age, requiring further optimization of the parameters while running. To minimize development cost and to control the air-fuel mixture with great precision during an engines full life span, this master thesis proposes a self-optimized system, i.e. an adaptive system, to control the air-fuel mixture. In the suggested method, the fuel injection to the engine is controlled with help of a linear lambda sensor, which measures the air-fuel mixture. The mapping from injection to measured air-fuel mixture forms a nonlinear system. It can be approximated as a linear function at static engine operating points, allowing the system at each static point to be modelled as a first order system with long time delay. To enable utilization over full operating area, and not only in static point, the controller uses large maps, so called gain-scheduling maps, to change control parameters. The tested controller is model based. It uses an Otto-Smith Predictor and a feed forward connection of target air-fuel. The model parameters in the controller are updated while driving and the adaptation method used is based on a least squares algorithm. The performance of the adapted controller and the adaptation method is tested in both simulation environment and in vehicle, showing good potential.

Analytical and finite element solutions to the problem of a vibrating beam on an elastic foundation are presented. An application example is a concrete railway sleeper embedded in an elastic medium (the ballast). The sleeper is also elastically connected to the rails. Eigenfrequencies are calculated and vibration modes are discussed. The beam (sleeper) is divided into sections where each section may or may not be supported by the elastic foundation. The elastic connections to the rails are situated at the two joinings of the three sleeper sections. Some conclusions are that Euler-Bernoulli beam theory can be used to calculate two, or maximum three, eigenfrequencies of the sleeper. The foundation stiffness influences the lowest bending-mode eigenfrequency the most; higher eigenfrequencies are practically unaffected by the foundation stiffness. The influence of railpad (and rail) stiffness on the sleeper eigenfrequencies is negligible.

With growing economies and better living standards in many parts of the world today, there is a need of expanding the capacity of generating electricity. The alternatives for renewable energy technologies are not fully developed and have still some problems to conquer. The company Minesto has developed a technology that is called Deep Green. It is designed to gain electricity from tidal currents. Deep Green works in a similar way as a kite that is attached to the seafloor with a moving wire. It contains of a wing that moves due to the water current. The path of the movement is circular or a shape of an “8”. The turbine under the wing is powered by the water flow which in turn leads to generation of electricity by the generator. The purpose of this thesis is to investigate the potential of using the Deep Green technology in ocean currents with respect to construction and economy. The site for the investigation is off the east coast of Florida where the Florida current streams with a speed of about 1, 5 m/s close to the surface. The major difference compared to the use in tidal currents is the deep waters of about 300 m and a oneway current direction. A mooring construction was developed and optimized and with the use of carbon fiber ropes, buoys and foundations lets 9 Deep Greens operate between 20 and 79 m depth. The idea is to install 7 such clusters with a total of 63 Deep Greens on the site in Florida. This configuration generates a power of 24 MW and gives a production of 185 GWh/year. Economic accounts were made with support of earlier made calculations regarding a park of 60 Deep Greens in a tidal current application. The results for the clustered configuration in ocean currents indicated a cost of 0,67 SEK/kWh with a discount rate of 8%. The required capital for the installation is about 780 mSEK (CAPEX). The operating costs (OPEX) are 43,3 mSEK. The design seems reasonable in many respects and it operates in a continuous ocean current with good electricity generation. The use of Deep Green in ocean currents speaks for being a profitable application. Though, it will be a large‐scale economic project, mainly because installations in a small scale will not be profitable due to costs such as grid connection. Critical issues to look at in a further development was considered to be surveys of the installation site, the displacement and movements of the mooring, the buoyancy‐system and the installation procedure.