Gitta Domik

An introduction to high-performance scientific computing

An overview of scientific computing: introduction, large-scale scientific problems, the scientific computing environment, workstations, supercomputers, further reading. Part 1 Background: a review of selected topics from numerical analysis - notation, error, floating-point numbers, Taylors series, linear algebra, differential equations, fourier series IEEE arithmetic short reference - single precision, double precision, rounding, infinity, NaN, and zero, of things not said, further reading UNIX, vi, and ftp - a quick review - UNIX short reference, vi short reference, ftp short reference elements of UNIX make - introduction, an example of using make, some advantages of make, the makefile, further examples, dynamic macros, user-defined macros, additional features, other examples, a makefile for C, creating your own makefile, futher information, a makefile for fortran modules, a makefile for C modules elements of fortran - introduction, overview, definitions and basic rules, description of statements, reading and writing, examples. Part 2 Tools: elements of matlab - what is MATLAB?, getting started, some examples, short outline of the language, built-in functions, MATLAB scripts and user-defined functions, input/output, graphics, thats it! elements of IDL - getting started, exploring the basic concepts, plotting, programming in IDL, input/output, using IDL efficiently, summary elements of AVS - basic concepts, AVS graphical programming - the Network editor, the geometry viewer, AVS applications, further reading. Part 3 Scientific visualization: scientific visualization - definitions and goals of scientific visualization, history of scientific visualization, example of scientific visualization, concepts of scientific visualization, visual cues, characterization of scientific data, visualization techniques, annotations, interactivity, interpretation goals to pursue with visualization, quantitative versus qualitative data interpretation. Part 4 Architectures: computer performance - introduction and background, computer performance, benchmarks, the effect of optimizing compilers, other architectural factors, vector and parallel computers, summary. (Part contents).

Multiple Incidence Angle SIR-B Expenment Over Argentina: Stereo-Radargrammetrc Analysis

Four overlapping SIR-B radar images were obtained across southern Argentina; these form a total of six stereo models with intersection angles ranging from 5° to 23°. This data set is uniquely suited for experimental evaluation of some basic assumptions on stereo -radargrammetry. Each stereo model was measured on a specially programmed photogrammetric analytical plotter; the resulting coordinates of ground points were compared with those from maps. It is concluded that accuracies are lower than expected at the larger stereointersection angles, amounting to about ±60 m in each coordinate direction. This might be explained by limitations of the quality of stereofusion caused by look angle differences and specular point migration, backscatter differences due to different incidence angles, differences in azimuth directions, and image noise and speckle.

User modeling for adaptive visualization systems

Meaningful scientific visualizations benefit the interpretation of scientific data, concepts and processes. To ensure meaningful visualizations, the visualization system needs to adapt to desires, disabilities and abilities of the user, interpretation aim, resources (hardware, software) available, and the form and content of the data to be visualized. We suggest describing these characteristics with four models: user model, problem domain/task model, resource model and data model. The paper makes suggestions for the generation of a user model as a basis for an adaptive visualization system. We propose to extract information about the user by involving the user in interactive computer tests and games. Relevant abilities tested are color perception, color memory, color ranking, mental rotation, and fine motor coordination.<<ETX>>

Real-Time Marching Cubes on the Vertex Shader

In this paper we propose a new approach for visualizing volumetric datasets by their isosurfaces. For an interactive isosurface reconstruction an optimized version of the well-known marching cubes algorithm is used. We extend the original algorithm by an additional vertex shader program. Contrary to other hardware-accelerated solutions our program is not based on a tetrahedral algorithm and thus the implementation for structured grids is more effective. Furthermore, surfaces of time-varying datasets at distinguished threshold values can be extracted in real-time.

Dependence of image grey values on topography in SIR-B images

This paper focuses on the use of a high resolution digital elevation model (DEM) to aid in rectifying and enhancing synthetic aperture radar images. Using a synthetic backscatter image, the SIR-B images are manually rectified and resampled to remove geometric distortions caused by topography. In a second step, an improved reflectance function of incidence angle is derived from the DEM and the rectified image and this function is used to reduce radiometric effects of topography yielding an albedo image which clearly shows the thematic, as opposed to topographic content of the image. The procedure is tested on four SIR-B images of a scene in Argentina (crossover point) that is imaged under different azimuth and incidence angles. The similarity of the resulting images indicates that the procedure effectively reduces artefacts from the images that are dependent on topography.

Multiple Incidence Angle SIR-B Experiment Over Argentina: Generation of Secondary Image Products

Original radar images may be geometrically and radiometrically distorted. This may be a particular problem when multiple angle imagery is analyzed and there is topographic relief in the area of interest. This paper describes a set of techniques designed to combine a multiple angle radar data set with a digital terrain elevation model, to generate a set of new images called secondary image products. These new images are geometrically rectified radar ortho-images, radiometrically rectified images, and stereo ortho-images. These secondary images can then reliably be used for thematic interpretation.

Do we need formal education in visualization

The following three reasons are sufficient to answer this question in the affirmative: careless mapping from data to pictures may lead to erroneous interpretation; a substantial amount of knowledge is necessary to generate images depicting complex information in a way that prevents erroneous interpretation; and decision making is increasingly based on visual representations. The organization and content of the core topics of visualization were finalized at a 1997 workshop at the Colorado School of Mines. These comprise the following eight themes: introduction to visualization; the data; the user and the task; mapping process; the representations; interaction issues; concepts of the visualization process; and systems and tools. The needs of educators and students vary too widely to bring one curriculum into focus. Instead, the eight themes recommended can be expanded into a curriculum, or compressed into several subtopics of a high-performance computing or computer graphics course. The themes can stand as separate modules, taught in a different order from that suggested. While my collaborators and I strongly recommend covering each of the core topics, we also encourage educators to expand individual themes to encompass the particular objectives of their students.

A survey of transfer functions suitable for volume rendering

There are many transfer functions (TF) that emphasize or hide special features of volume data. Their potential to cleverly generate a color and opacity value for direct volume rendering is primarily determined by the used metrics besides the input data value. Despite this variety of TFs, for the most part simple one dimensional data value only based TFs are used in practice. This survey will therefore examine the differences in representative TF types (defined by their used metrics) to provide a basis for selecting the right TF type for the boundary conditions that describe an individual field of application and task. Besides fundamental properties like metrics or memory consumption, we will also give an assessment about user interaction and quality of feature emphasis.

How effective are 3D display modes

The increasingavailability of 3D input and output devices demands a better understanding and comparison of their quality. This report describes an empirical experiment for comparing 3D display modes: traditional perspective viewing, anaglyph stereo and shutter glass stereo. We followed two hypotheses 1. shutter glass stereo viewing allows a faster and more accurate recognition than the anaglyph and the perspective viewing, and 2. subjects experienced with particular 3D representationsare faster and more accurate than subjects without experience. The experiment is based on a true research scenario in organicchemistry.Organic molecules were used as 3D objects.Mean response error and mean response time were calculated for a series of six tasks and 81 subjects.

A Breadth-First Approach for Teaching Computer Graphics.

Two themes of computer graphics, namely computer-generated color and computer-generated visualization, are introduced in a teaching/learning tool using a breadth-first approach. The breadth-first approach provides a holistic view of the topics to teach, aids in interdisciplinary teaching and is equally motivating to male and female students. The breadth-first approach has now been used for several years and we can also report on evaluations of the teaching/learning tool.