1,720,978 research outputs found
Techniques and software architectures for medical visualisation and image processing
No full textThis thesis presents a flexible software platform for medical visualisation and image processing, a technique for the segmentation of the shoulder skeleton from CT data and three techniques that make contributions to the field of direct volume rendering. Our primary goal was to investigate the use of visualisation techniques to assist the shoulder replacement process. This motivated the need for a flexible environment within which to test and develop new visualisation and also image processing techniques with a medical focus. The Delft Visualisation and Image processing Development Environment, or DeVIDE, was created to answer this need. DeVIDE is a graphical data-flow application builder that combines visualisation and image processing techniques, supports the rapid creation of new functional components and facilitates a level of interaction with algorithm code and parameters that differentiates it from similar platforms. For visualisation, measurement and pre-operative planning, an accurate segmentation from CT data of the bony structures of the shoulder is required. Due to the complexity of the shoulder joint and the fact that a method was required that could deal with diseased shoulders, existing techniques could not be applied. In this thesis we present a suite of techniques for the segmentation of the skeletal structures from CT data, especially designed to cope with diseased shoulders. Direct volume rendering, or DVR, is a useful visualisation technique that is often applied as part of medical visualisation solutions. A crucial component of an effective DVR visualisation is a suitable transfer function that assigns optical characteristics to the data. Finding a suitable transfer function is a challenging task. We present two highly interactive methods that facilitate this process. We also present a method for interactive direct volume rendering on ubiquitous low-end graphics hardware. This method, called ShellSplatting, is optimised for the rendering of bony structures from CT data and supports the hardware-assisted blending of traditional surface rendering and direct volume rendering. This characteristic is useful in surgical simulation applications. ShellSplatting is based on the object-order splatting of discrete voxels. As such, maintaining a correct back-to-front or front-to-back ordering during rendering is crucial for correct images. All existing real-time perspective projection visibility orderings show artefacts when splatting discrete voxels. We present a new ordering for perspective projection that remedies these artefacts without a noticeable performance penalty.Electrical Engineering, Mathematics and Computer Scienc
Visual Analysis of Multi-Field Data
No full textThis thesis investigates methods for the visualization of multi-field medical data. In the medical field, data complexity has been growing consistently over the past years. Not only the size of the data grows, but also the need to visualize beyond traditional boundaries. We present a number of novel facets that encompass a general approach to the exploration of multi-field data. We strongly believe that human-in-the-loop visual data analysis on large and complex datasets is best aided by multiple linked different representations. The presented techniques demonstrate how complex data from multiple modalities can be visualized and interactively explored. We explore the use of linked selections to aid in reducing the complexity of the visualizations. Using multiple-linked views, we can integrate multiple orthogonal representations of the data simultaneously. We have applied aforementioned techniques in the design and implementation of a number of prototype frameworks, with applications ranging from brain imaging for neurosurgical planning to the study of the behavior of marine animals through the use of sensor data. We also present a conceptual framework for studying complex longitudinal data, by means of aggregation and multi-level visualization. We successfully adapted techniques from information visualization in order to use them on datasets that are orders of magnitude larger than they are originally used for.Computer Science, Computer Graphics groupElectrical Engineering, Mathematics and Computer Scienc
Visualizing Cumulus Clouds in Virtual Reality
No full textThis thesis focuses on interactively visualizing, and ultimately simulating, cumulus clouds both in virtual reality (VR) and with a standard desktop computer. The cumulus clouds in question are found in data sets generated by Large-Eddy Simulations (LES), which are used to simulate a small section of the atmosphere over a period of several hours. These data sets are large, 3D, multi-variate, and time-varying, which pose several difficult visualization challenges. In order to overcome these challenges and gain insight into such complex data, several techniques are developed and employed together. At a high level, the research presented in this thesis can be divided into four categories: analyzing and visualizing interesting features in the data, giving the user sufficient control over the visualization, keeping the visualization interactive, and interactively simulating the cumulus clouds. The first step to understanding the data was to find and track the features, i.e. the clouds, in the data. Through the use of a connected component labeling algorithm, individual clouds in the data could be identified. These clouds were then visualized in a VR visualization environment, which allowed atmospheric scientists to visually identify interesting clouds for further study. The next step along the way was to improve the visualization experience. One aspect of this was to develop data “reprocessing”. This allowed the atmospheric scientists to generate derived data from the raw simulation data for inclusion in the visualization environment. Another aspect of this was to improve the user interface with a more intuitive interaction technique and through the use of familiar 2D widgets. The next challenge to address was the data access bottleneck. In order to move more data from disk to main memory and from main memory to the GPU, a lossy vector compression method based on quantization is developed. By analyzing and bounding the angular error introduced by quantization, unit vectors can be quantized using 16 bits with less than 0.4 degrees of angular error. This can reduce the size of mesh geometry, and, when also quantizing vector length, can be used to compress vector fields. In order to help understand the relationship between the clouds and the air around them, interactive particle tracing was the next research area. Using the power of the GPU, millions of particles could be advected interactively. These particles could be visualized as particles in different visual styles or as flow curves such as stream-lines or streak-lines. By combining vector-field compression with a multi-resolution advection scheme, users could interactively seed and advect particles around selected clouds of interested through time. Most of the techniques developed in this thesis have been integrated into Cloud Explorer, which is an experimental VR visualization platform for interactively visualizing and studying the cumulus cloud data. Several techniques have also been integrated into stand-alone applications. The final research area addressed in this thesis was interactive simulation of the cumulus clouds. GALES, a GPU-based, atmospheric, Large-Eddy Simulation, was the result of this effort. GALES runs 16x faster than the Dutch Atmospheric Large-Eddy Simulation, which is a Fortran-based LES, while maintaining comparable numerical accuracy. This speedup enables GALES to interactively run and visualize simulations that fit into GPU memory. This opens new and exciting possibilities for future computational steering research.MediamaticsElectrical Engineering, Mathematics and Computer Scienc
Techniques and architectures for 3D interaction
No full textSpatial scientific datasets are all around us, and 3D visualization is a powerful tool to explore details and structures within them. When dealing with complex spatial structures, interactive Virtual Reality (VR) systems can potentially improve exploration over desktop-based systems. However, from previous experience it was found that traditional interaction tools as used in desktop visualization systems do not function well within a Virtual Environment (VE), especially not for complex spatial data exploration tasks. Also the application of basic 3D interaction techniques is not straightforward, as they require customization for specific scenarios, applications and VR systems. This process of design, development and customization of 3D interaction techniques is not well supported by existing software tools for VR and visualization. In this thesis we focus on enhancing 3D interaction techniques and interfaces for data exploration in VR. Based on observations from human actions and perception in tracked, projection-based VR systems, we extend some fundamental 3D interaction metaphors. We propose an enhanced technique for 3D selection of objects in a VE, and explore 3D selection and manipulation in collaborative situations. From our work on these techniques we have come to consider 3D interaction as a complex, first-class component of 3D systems. In the integration of these advanced techniques in our framework and applications, we experienced many iterations of redesign and tuning. To better support this emergent development style, we developed a prototyping architecture for constructing and adapting VR applications. Because of the complexity of 3D interaction techniques, further support for their specific design and development was required. We extended the prototyping architecture with a modelling approach, specific to the description, execution and iterative improvement of interaction techniques. To truly enable the potential of an interactive VR system, expertise from many areas such as perception, usability design and software engineering needs to be integrated in its design. With the contributions presented in this thesis, we hope to improve this view and facilitate prototypical design and development cycles for improved 3D environments for interactive exploration.MediamaticsElectrical Engineering, Mathematics and Computer Scienc
Presentation and exploration of flow data
Get PDFElectrical Engineering, Mathematics and Computer Scienc
Visualization on a Budget for Massive LiDAR Point Clouds
No full textThe recent emergence of LiDAR scanning technology has resulted in the availability of very large three dimensional point cloud data sets. An example of such a data set is the AHN2 data set, which contains a high-density point cloud for the Netherlands. The direct visualization of these point cloud data sets is important because it provides the first step in understanding them. However, the size of these data sets prevents them from simply being displayed, let alone from being explored interactively. As such, they present a challenge in visualization. In this thesis, we describe the methods used by earlier works to beat this challenge. These methods involve the use of space driven spatial data structures to provide efficient out-of-core access to the point cloud data set, and the alleviation of one or more of the bottlenecks in the point cloud visualization pipeline to improve interactiveness. Choosing to use a different approach, we present a method to interactively visualize these point cloud data sets using the concept of visualization on a budget. We reduce the complexity of a point cloud visualization by using continuous level-of-detail, and we minimize its impact on the effectiveness of the visualization by using a point analogue of visual importance, for which we present a metric. The complexity of the visualization is automatically adapted to maintain interactiveness. As a result, the visualization always consists of those points that, for a given metric, best represent the point cloud data set for the required interactiveness. We present a prototype toolchain that includes the implementation of our method and demonstrate its performance on a variety of point cloud data sets. We describe its performance characteristics and present a preliminary user study for the method. We evaluate initial results and conclude that our method works, and propose the development of a better metric that does not require user intervention. We end by showing that the prototype toolchain can be used to continue research on this promising method.Computer Graphics and CAD/CAMMediamaticsElectrical Engineering, Mathematics and Computer Scienc
Spatial Navigation for Context-aware Video Surveillance
No full textAn increasing number of cameras is being used to monitor a growing range of environments. Consequently, surveillance systems consist of an increasing number of screens to display all incoming video streams, making it difficult for observers to maintain a mental model of spatial relations between videos. A number of systems is developed aiming at improving the observer’s spatial awareness by integrating videos with their spatial context, a 3D model of the monitored environment. In these systems, video content can be viewed from virtual cameras that correspond with the cameras in the real world. To switch between views on certain videos, the observer has to make a transition between the corresponding virtual cameras by some means of navigation. While studying state-of-the-art 3D surveillance systems, we have observed that not much attention has been paid on exploring more sophisticated viewpoint transitions. In this thesis, different classes of camera relations are identified. For each class, a viewpoint transition mechanism is developed with the focus on reducing distortion of the video information during transitions. In order to navigate through the virtual environment, viewpoint transitions have to be initiated to switch between videos. As part of this thesis, a number of concepts have been developed that provide controls to the viewpoint transitions in the form of 3D elements that are added to the virtual environment. We have implemented our navigation concepts in a prototype, which was used for a user study. This thesis concludes with the results from this user study and our vision on future work in the field of context-aware surveillance.MediamaticsElectrical Engineering, Mathematics and Computer Scienc
Going Beyond Counting First Authors in Author Co-citation Analysis
Get PDFThe present study examines one of the fundamental aspects of author co-citation analysis (ACA) - the way co-citation
counts are defined. Co-citation counting provides the data on which all subsequent statistical analyses and mappings
are based, and we compare ACA results based on two different types of co-citation counting - the traditional type that
only counts the first one among a cited work's authors on the one hand and a non-traditional type that takes into
account the first 5 authors of a cited work on the other hand. Results indicate that the picture produced through this non-traditional author co-citation counting contains more coherent author groups and is therefore considerably clearer. However, this picture represents fewer specialties in the research field being studied than that produced through the traditional first-author co-citation counting when the same number of top-ranked authors is selected and analyzed. Reasons for these effects are discussed
Variations on the Author
Get PDF“Variations on the Author” discusses two of Eduardo Coutinho’s recent films (Um Dia na Vida, from 2010, and Últimas Conversas, posthumously released in 2015) and their contribution to the general question of documentary authorship. The director’s filmography is characterized by a consistent yet self-effacing form of authorial self-inscription: Coutinho often features as an interviewer that rather than express opinions propels discourses; an interviewer that is good at listening. This mode of self-inscription characterizes him as an author who is not expressive but who is nonetheless markedly present on the screen. In Um Dia na Vida, however, Coutinho is completely absent form the image, while Últimas Conversas, on the contrary, includes a confessional prologue that moves the director from the margins to the center of his films. This article examines the ways in which these works stand out in the filmography of a director who offers new insights into the notion of cinematic authorship
- …
