Surface View

Representative data characteristics

• 2-d scalar data arrays, e.g. z = f(x,y), such as elevation map
•  

Wireframe technique

• treat "z" as elevation over 2-d terrain and use projection from 3-d to 2-d
•  
• project mesh of lines parallel to x and y axes
•  

Shaded surface technique

• treat "z" as elevation over 2-d terrain and use projection from 3-d to 2-d
•  
• project each data element
•  
• remove hidden surfaces
•  
• assign grey value / color value
•  

Special note on effectiveness

• source of grey value/color value must be transparent to viewer
•  

Bump Map

Representative data characteristics

• 2-d scalar data arrays, e.g. z = f(x,y), such as elevation map
•  

Technique

• treat "z" as elevation over 2-d terrain
•  
• illuminate terrain to produce shading
•  

Special note on effectiveness

• no distortions present in relief
•  
• may confuse if used on top of surface view
•  

Reference(s) [KEL93]

Image Display

Representative data characteristics

• 2-d scalar data arrays, e.g. z = f(x,y), such as LANDSAT image
•  

Technique

• straightforward: map each 2-d data element to brightness or color of screen pixel
•  

Special note on effectiveness

• color/brightness scale necessary
•  

Color Transformations

Representative data characteristics

• up to three scalar data arrays defined over same two dimensions, e.g. zi = f(x,y), i=1,2,3 such as three TM (Thematic Mapper) channels of same terrain
•  

Technique

• choose same technique (e.g. image display or surface) for each data array
•  
• read zk = f(i,j), k=1,2,3 for each pixel location on screen, resulting in 3 brightness values (z1, z2, z3)
•  
• use (z1, z2, z3) as coordinates to color space, e.g. RGB, HSV =%gt; 'color'
•  
• use 'color' to paint pixel at screen location (i,j)
•  

Special note on effectiveness

• use RGB for data arrays of same data type; use HSV or HLS for different data types
•  
• effective for correlation/association of data elements
•  

Examples of techniques for 3-dimensional (continuous) scalar data

Volume Slices

Representative data characteristics

• 3-d scalar data arrays, e.g. w = f(x,y,z), such as medical scans of human organs
•  

Technique

• intersect 2-d plane(s) with volume
•  
• use image display for visual representation
•  
• project planes to screen
•  

Special note on effectiveness

• use appropriate coordinate system to depict location of plane(s) in volume
•  
• animation (change of view point), hidden surfaces and perspective geometry for 3-d effect
•  

Basket Weave

Representative data characteristics

• 3-d scalar data arrays, e.g. w = f(x,y,z), such as medical scans of human organs
•  

Technique

• calculate (contour) lines at cross-sections parallel to coordinate planes
•  
• project contour lines to screen
•  
• draw thick, opaque bands
•  

Special note on effectiveness

• use appropriate coordinate system to depict location of plane(s) in volume
•  
• hidden surfaces and perspective geometry for 3-d effect
•  

Reference [SEW88]

Surface Rendering

Representative data characteristics

• 3-d scalar data arrays, e.g. samples of w = f(x,y,z), where w (voxel value) might indicate color, opacity, density, material, or time.
•  

Technique

• surface reconstruction (define surfaces in 3-d raster)

  (--e.g. by using marching cubes algorithm or surface detection)

•  
• surface rendering (illumination, shading, projection)
•  

Reference [KAU91]

Volume viewing

Representative data characteristics

• 3-d scalar data arrays
•  

Technique

• project volumetric data elements onto the display space, by either
  • backward projection (object-order): scan voxel space and project to screen
  •  
  • forward projection (image-order): scan screen pixels and determine voxel contributions
  •  
  • combination
  •  
• assign pixel brightness/color
•  

Special note on effectiveness

• transparency/translucency
•  

Reference [KAU91], [KAU94]

Tiny Cubes

Representative data characteristics

• discrete 3-d scalar data array
•  

Technique

• place small objects, such as cubes and spheres, in the volume
•  
• determine brightness/color of pixel by the value at the corresponding location
•  

Special note on effectiveness

• open space between objects allows insight
•  

Reference [NIE90]

Examples of techniques for vector data

Arrows

Representative data characteristics

• vector fields
•  

Technique

• use arrow as glyph, vary following attributes of arrow depending on variables:
  direction/length/width/reflection properties of shaft, type/color of arrow head
•  

Special note on effectiveness

• avoid cluttering by reducing amount of data to display
•  
• additional problems in 3-d through directional ambiguity
•  

Reference(s) [POS94]

Particle traces and motion, streamlines, stream ribbons and surfaces

Representative data characteristics

• vector fields
•  

Techniques

• particle traces: polyline tracing particle path
•  
• particle motion: animation of particle movement
•  
• streamlines: polylines tracing lines tangent to vector field
•  
• stream ribbons: surface between two adjacent stream lines
•  
• stream surfaces: surface defined by set of adjacent stream lines
•  

Special note on interaction

• special interaction tools (DataGlove) and methods (gesturing) necessary
•  
• interactive steering and interactive computation on data necessary
•  

Reference(s) [POS94], [HEL94]

Examples of techniques for information visualization

Perspective Wall

Representative data characteristics

• content of relational data base
•  

Technique

• folds a 2-d layout into a 3-d wall
•  
• integrates a central region for viewing details with two perspective regions, one on each side, for viewing context.
•  

Special note on effectiveness

• efficient space utilization
•  
• smooth transitions of views
•  

Special note on interaction

• move along linear direction
•  

Reference(s)

• [ROB93]

Cone Trees

Representative data characteristics

• content of hierarchical data base
•  

Technique

• hierarchies laid out uniformly in three dimensions
•  
• top of hierarchy is apex of a cone
•  
• place children evenly spaced along base of cone
•  
• nodes are drawn as index cards and contain textual information
•  
• body of cone is transparent
•  

Special note on interaction

• cones may be rotated to reveal information
•  

Reference(s) [ROB93]

Examples of techniques for software visualization

Scalar Mapping Technique

Representative data characteristics

• all data types defined in algorithms
•  

Technique

• user-defined variables can be mapped to display components
•  

Special note on interaction

• interaction to understand algorithm behavior
•  

Reference(s): [HIB92] for VIS-AD, [KIM94] for PV

Program Flow Diagrams

Visual Programming

Representative data characteristics

• program modules and relations to solve specific problem
•  

Technique

• iconize each module/function of software system
•  
• connect selected modules by (multi-colored) pipes into network
•  
• activate network to execute program
•  

Special note on interaction

• interactive generation of program network
•  

Reference(s): e.g. AVS, Khoros, SGI Explorer, apE

Ball-and-Stick Technique

Representative data characteristics

• molecular structures
•  

Technique

• represent atoms as balls and bonds a sticks
•  

Special note on effectiveness

• reflections and shadows enhance 3-d effect and reduce concealed surfaces
•  

Special note on interaction

• "conditional boxes", if large amount of atoms/molecules

Reference(s)

[KEL93], [FOL94]

Animation

Representative data characteristics

• sequence of pictures changing over one parameter, usually time, spectral properties, temperature, view points
•  

Technique

• careful interpolation may be used between keyframes
•  
• rapid updating of screen to present sequence of phenomena
•  
• creates illusion of movement
•  

Special note on effectiveness

• update of frames necessary: at least 10 frames/sec
•  
• can be used to create 3-d effect (e.g. fly-over, rotation)
•  

Reference(s) [BRY94], [THA94]; for algorithm visualization see [BRO84]

Examples of techniques for special purposes

• Missing data technique, if interpolation is permitted
  • use interpolation method meaningful to problem space
  •  
  • use color for measured data
  •  
  • brightness alone for missing data
  •  
• Fisheye View
  • to provide focus within larger (continuous) information space
  •  

Glossary

animation
A movie. A sequence of related images viewed in rapid succession to see and experience the apparent movement of objects. [KEL93]

back-to-front
A volume viewing algorithm in which the traversal for viewing is performed from the farthest voxel backward to the closest one. Voxels nearer to the viewer overwrite voxels that are farther to the back. [KAU91]

brightness
The apparent intensity of light. Often a synonym for intensity [KEL93]

contour plots
A technique for plotting scalar data of the form f(x,y) by constructing closed (level) curves of equal values of f. [WOL93]

effectiveness
An effective graph presents all information clearly in view of visualization aims. [MAC86]

expressiveness
An expressive graph encodes all relevant information and only that information. [MAC86]

front-to-back
A volume viewing algorithm in which the traversal for viewing is performed from the voxel closest to the viewer to the farthest one. Voxels are written only to pixels that are not painted yet. [KAU91]

glyph
An object or symbol for representing data values. Glyphs are generally a way of representing many data values and are sometimes called icons. A common glyph is the arrow, often chosen to represent vector fields. The arrow depicts both speed and direction at a point. [KEL93]

Grand Tour
The grand tour is a method for viewing multivariate statistical data via orthogonal projections onto a sequence of two-dimensional subspaces. The sequence of subspaces is chosen so that it is dense in the set of all two-dimensional subspaces. Desirable properties of such sequences of subspaces are considered, and several specific types of sequences are tested for rapidity of becoming dense. Tabulations are provided of the minimum length of a grand tour sequence necessary to achieve various degrees of denseness in dimensions up to 20. [ASI85]

HLS (hue,lightness,saturation)
This color model is defined in the double-cone subset of a cylindrical space. Hue (H) is measured by the angle around the vertical axis, with red at 0 degree, green at 120 degree and so on. The height of the cone represents lightness (L) in a range from 0 (black) at the apex of the first cone to 1.0 (white) at the apex of the second one. Saturation (S) is a ratio ranging from 0 on the center line (V axis) to 1 on the triangular sides of the cones.

HSV (hue,saturation,value)
This colour model is user-oriented, being based on the intuitive appeal of the artist's tint, shade and tone. The coordinate system is cylindrical and the HSV model is defined as a cone within this cylinder. Hue (H) is measured by the angle around the vertical axis, with red at 0 degree, green at 120 degree and so on. The height of the cone represents value (V) in a range from 0 (black) at the apex to 1.0 (white) at the base of the cone. Saturation (S) is a ratio ranging from 0 on the center line (V axis) to 1 on the triangular sides of the cone.

icons
See glyphs.

intensity (of color)
The amount of measured light energy. Often a synonym for brightness. [KEL93]

interactive
Describes behavior of the computer and program designed to respond to the user's request in a timely manner, generally a few seconds or milliseconds. [KEL93]

interpolation
The process of computing new intermediate data values between existing data values. [KAU91]

interpretation aims
User's goals when interpreting picture, such as identifying objects, comparing values of objects, distinguishing objects, focusing on certain details in text

isosurface
Surfaces within a volume that have the same parameter value. [WOL93]

Marching Cubes
A method of visualizing 3-D data structures by looking for level surfaces in a 3-space comprised of a lattice of points. In contrast to volume rendering, where one can see the entire structure, marching cubes only allows a single surface to be rendered. [WOL93]

nominal data types
are unordered collections of symbolic names without units. For instance, the names of the orbiters, such as Hubble, Magellan, Mariner, Viking and Voyager form a nominal data set.

opacity
A material property that prevents light from passing through the object. [KAU91]

ordinal data types
are rank-ordered only, where the ordering does not reflect the magnitudes of the differences. A typical example of an ordinal data set is the sequence of names of the calendar months, January through December.

pixel
Equivalently, picture element, a pixel is the smallest unit of a computer image and is assigned a unique color after rendering. [WOL93]

quantitative data types
are usually expressed as REAL values in the data set. The precise numerical value has a certain importance in the semantics of the data. have concrete values like reals . A typically quantitative data set is the length of objects.

radiosity
In Computer Graphics, the rate at which light energy leaves a surface, which includes transmission and reflection. Rendering techniques which compute the radiosity of all surfaces in a scene have been termed radiosity methods. [WOL93]

ray-casting
A volume viewing algorithm in which sight rays are cast from the viewing plane through the volume. The tracing of the ray stops when the visible voxels are determined by accumulating or encountering an opaque value. [KAU91]

ray-tracing
The general technique of computing an image by projecting rays into a scene and using their interactions with the contents of that scene to determine pixel colors. In surface- rendering methods, rays are intersected and possibly reflected or refracted by objects in the scene to determine visible colors. Ray-tracing is also used in volume visualization and is a type of DVR. [WOL93]

render
The process of converting the polygonal or data specification of an image to the image itself, including color and opacity information. [WOL93]

renderer
A software algorithm which renders an image, calculating a color at each pixel based on object visibility and lighting and shading models. [WOL93]

RGB
Red-green-blue, the color standard employed by the most computer manufacturers and which roughly corresponds, in frequency, to the three bands of colors sensitivity of the human eye. [WOL93]

scalar data types
possess a magnitude, but not directional information other than a sign; they are simply defined as single numbers. Same as quantitative data types in this text.

surface modeling
Techniques and tools for building up computer representations of objects by modeling their surfaces, usually as a collection of polygonal facets. [WOL93]

surface rendering
an indirect technique used for visualizing volume primitives by first converting them into an intermediate surface representation (see surface reconstruction) and then using conventional computer graphics techniques to render them. [KAU91]

surface reconstruction
A procedure that converts a set of data points or cross sections into a surface representation by identifying the surface and representing it with geometric surface primitives. The reconstruction procedure may use one of several techniques, such as contouring, tiling, marching cubes, surface detection, and surface tracking. [KAU91]

thresholding
A technique used primarily with surface rendering, in which a density value of the interface between two materials in the dataset is selected so that the interface surface can be identified for rendering. [KAU91]

translucency
describes the property that allows light to partially pass through and partially reflect. Translucency has the effect of making the translucent area appear smoky or cloudlike, thus revealing objects behind. [KEL93]

transparency
A material property that allows light to pass through the object. [KAU91]

vectors
have direction and magnitude. Quantitatively, their mathematical presentation requires a number of scalar components equal to the dimensionality of the coordinate system. In general, a vector is a unified entity, which implies the problem of displaying independent, multivariate scalar fields.

visualization idiom
is a specific sequence of data enrichment and enhancement transformations, visualization mappings and rendering transformations that produce an abstract display of a scientific data set [HAB90]

volume rendering
Volume rendering is a direct technique for visualizing volume primitives without any intermediate conversion of the volumetric dataset to surface representation. [KAU91]

volume viewing
The process of projecting the volumetric dataset onto the image space by determining which voxels are visible and what their contribution to the final image is. [KAU91]

volume visualization
Volume visualization is a visualization method concerned with the representation, manipulation, and rendering of volumetric data. [KAU91]

volumetric dataset
A volumetric dataset is represented as a 3D discrete regular grid of volume elements (voxels) and is commonly stored in a volume buffer (or cubic frame buffer, like frame buffer in 2D), which is a large 3D array of voxels. [KAU91]

volumetric graphics
Volume graphics is the subfield of computer graphics concerned with volume synthesis, volume modeling and volume visualization, typically using a cubic frame buffer to store the volumetric dataset. Volumetric graphics is the 3D conceptual counterpart of raster graphics. [KAU91]

voxel
An abbreviation for "volume element" or "volume cell." It is the 3D conceptual counterpart of the 2D pixel. Each voxel is a quantum unit of volume and has a numeric value (or values) associated with it that represents some measurable properties or independent variables of the real objects or phenomena. [KAU91]

References

[ARE94], H. Aref, R. D. Charles and T. T. Elvins, Scientific Visualization of Fluid Flow, in C.A. Pickover and S.K. Tewksbury (eds), Frontiers of Scientific Visualization, 1994, Wiley Interscience.

[ASI85] D. Asimov, The Grand Tour: A Tool for Viewing Multidimensional Data. SIAM J. Sci. Statistical Computing, Vol. 6, No.1, 1985, pp. 128-143.

[BAE86] R. M. Baecker, 1986, An Application Overview of Program Visualization. Computer Graphics: SIGGRAPH '86, 20 (4):325, July 1986

[BED90] J. Beddow, 1990, Shape Coding for Multidimensional Data on a Microcomputer display. Proc. of IEEE Conf. VISUALIZATION '90, pp. 238-246.

[BER67] J. Bertin (1967), Semiologie Graphique, Gauthier-Villars, Paris.

[BES93] C. Beshers and Feiner, S., 1993, AutoVisual: Rule-based design of interactive multivariat visualizations. IEEE Computer Graphics and Applications. 13 (4), pp. 41-49.

[BES94] C. Beshers and S. Feiner, Automated Design of Data Visualization, in Scientific Visualization, 1994, Advances and Challenges, Ed: L. Rosenblum, R.A. Earnshaw, J. Encarnacao, H. Hagen, A. Kaufman, S. Klimenko, G. Nielson, F. Post, D. Thalmann , Academic Press.

[BRO84] M.H. Brown and R. Sedgewick, A System for Algorithm Animation, Computer Graphics, Vol. 18, No. 3, July 1984.

[BRO92] K.W. Brodlie, L.A. Carpenter, R.A. Earnshaw, J.R. Gallop, R.J. Hubbard, A.M. Mumford, C.D. Osland, P. Quarendon (eds), Scientific Visualization, Techniques and Applications, 1992, Springer-Verlag.

[BRY94] S. Bryson, Real-Time Exploratory Scientific Visualization and Virtual Reality, in Scientific Visualization, 1994, Advances and Challenges, Ed: L. Rosenblum, R.A. Earnshaw, J. Encarnacao, H. Hagen, A. Kaufman, S. Klimenko, G. Nielson, F. Post, D. Thalmann , Academic Press.

[CAS91] Stephen M. Casner, A Task-Analytic Approach to the Automated Design of Graphic Presentations, ACM Trans Graphics, Vol. 10, No. 2, April 1991, Pages 111-151.

[CRA90] Stuart L. Crawford and Thomas C. Fall, 'Projection Pursuit Techniques for Visualizing High-Dimensional Data Sets', in Visualization in Scientific Computing, G. M. Nielson, B. Shriver and L.J. Rosenblum (eds), IEEE Computer Society Press.

[DOM93] Domik, G., 1993, A Paradigm for Visual Representations, University of Colorado, Department of Computer Science, Boulder, CO. 80309-0430, USA.

[DOM94] G. Domik and B. Gutkauf, 1994, User Modeling for Adaptive Visualization Systems, Proceedings of IEEE Visualization '94, IEEE Computer Society Press, pp. 217-223.

[FOL94] J. Foley and B. Ribarsky, Next-generation Data Visualization Tools, in Scientific Visualization, 1994, Advances and Challenges, Ed: L. Rosenblum, R.A. Earnshaw, J. Encarnacao, H. Hagen, A. Kaufman, S. Klimenko, G. Nielson, F. Post, D. Thalmann , Academic Press.

[FOS95] L.D. Fosdick, E.R. Jessup, C. J.C. Schauble, and G. Domik, An Introduction to High-Performance Scientific Computing, to appear by MIT Press in December 1995.

[GAL94]J. Gallop, Underlying Data Models and Structures for Visualization, in Scientific Visualization, 1994, Advances and Challenges, Ed: L. Rosenblum, R.A. Earnshaw, J. Encarnacao, H. Hagen, A. Kaufman, S. Klimenko, G. Nielson, F. Post, D. Thalmann , Academic Press.

[GER94] N. Gershon, From Perception to Visualization, in Scientific Visualization, 1994, Advances and Challenges, Ed: L. Rosenblum, R.A. Earnshaw, J. Encarnacao, H. Hagen, A. Kaufman, S. Klimenko, G. Nielson, F. Post, D. Thalmann , Academic Press.

[GOR89] I. E. Gordon, Theories of Visual Perception, 1989, John Wiley & Sons.

[GRI90] G. Grinstein and S. Smith (1990), The Perceptualization of Scientific Data in Proc. SPIE (Int. Soc. Opt. Eng.) Extracting Meaning from Complex Data: Processing, Display, Interaction, Santa Clara, CA, 1259: 190-0

[HAB90] R.B. Haber, and D. A. McNabb, Visualization Idioms: A Conceptual Model for Scientific Visualization Systems, in Visualization in Scientific Computing, G. M. Nielson, B. Shriver and L.J. Rosenblum (eds), IEEE Computer Society Press.

[HEL94] James Helman and Lambertus Hesselink, Representation and Display of Vector Field Topology in Fluid Flow Data Sets, in Scientific Visualization, 1994, Advances and Challenges, Ed: L. Rosenblum, R.A. Earnshaw, J. Encarnacao, H. Hagen, A. Kaufman, S. Klimenko, G. Nielson, F. Post, D. Thalmann, Academic Press.

[HIB92] William Hibbard, Charles R. Dyer and Brian Paul,'Display of Scientific Data Structures for Algorithm Visualization', IEEE Proceedings in Visualization 92, pp. 139-146

[INS94] A. Inselberg (panel chair), G. Grinstein, T. Mihalisin, H. Hinterberger, A. Inselberg (panelists), 1994, Visualizing Multidimensional (Multivariate) Data and Relations, Proc. of IEEE Conf. VISUALIZATION '94, pp. 404-409.

[KAU91] Arie Kaufman, 1991, Volume Visualization, IEEE Computer Society Press Tutorial.

[KAU94] A. Kaufmann, Trends in Volume Visualization and Volume Graphics, in Scientific Visualization, 1994, Advances and Challenges, Ed: L. Rosenblum, R.A. Earnshaw, J. Encarnacao, H. Hagen, A. Kaufman, S. Klimenko, G. Nielson, F. Post, D. Thalmann , Academic Press.

[KEL93] P. Keller and M. Keller, Visual Cues, 1994, IEEE Computer Society Press.

[KIM94] D. Kimelman, B. Rosenburg, T. Roth, Strata-Various: Multi-Layer Visualization of Dynamics in Software System Behavior, in IEEE Proceedings Visualization 94.

[MAC86] J. Mackinlay, 1986, Automating the Design of Graphical Presentations of Relational Information, ACM Trans. on Graphics, Vol. 5, No. 2, April 1986, pp 110-141.

[MCC87] McCormick, B.H., T.A. DeFanti, M.D. Brown (ed), Visualization in Scientific Computing, Computer Graphics Vol. 21, No. 6, November 1987

[NIE90] Nielson Gregory M, Hamann B.,'Techniques for the Interactive Visualization of Volumetric Data', IEEE Proceedings in Visualization 90, pp. 45-49.

[POS94] Frits H. Post and Jarke J. van Wijk, 1994, Visual representation of vector fields: recent developments and research directions, in Scientific Visualization, 1994, Advances and Challenges, Ed: L. Rosenblum, R.A. Earnshaw, J. Encarnacao, H. Hagen, A. Kaufman, S. Klimenko, G. Nielson, F. Post, D. Thalmann, Academic Press.

[ROB91] Robertson, P.K. , 1991, A Methodology for Choosing Data Representations, IEEE Computer Graphics and Applications, Vol. 11, No. 3, May 1991, pp. 56-68.

[ROB93] George G. Robertson, Stuart K. Card, and Jock D. Mackinlay,'Information Visualization Using 3D Interactive Animation', CACM, Vol. 36, No. 4, April 1993.

[ROB94]P. Robertson and L. De Ferrari, Systematic Approaches to Visualization: Is a Reference Model Needed? in Scientific Visualization, 1994, Advances and Challenges, Ed: L. Rosenblum, R.A. Earnshaw, J. Encarnacao, H. Hagen, A. Kaufman, S. Klimenko, G. Nielson, F. Post, D. Thalmann , Academic Press.

[ROT90] Roth, S. and Mattis, J., 1990, Data characteristization for intelligent graphics presentation. In Proceedings CHI '90, April, pp. 193-200. ACM Press.

[SEN94] H. Senay and E. Ignatius, A Knowledge-Based System for Visualization Design, IEEE Computer Graphics and Applications, November 1994, pp. 36-47.

[SEW88] Sewell, 1988,'Plotting Contour Surfaces', ACM Transactions on Mathematical Software, Vol 14, (1), Mar 1988.

[STA92] J. T. Stasko and Ch. Patterson, 1992, Understanding and Characterizing Software Visualization Systems, Proc. of 1992 IEEE Workshop on Visual Languages, IEEE Computer Society Press.

[THA94] Nadia M Thalmann and Daniel Thalmann, Computer animation: a key issue for time visualization, in Scientific Visualization, 1994, Advances and Challenges, Ed: L. Rosenblum, R.A. Earnshaw, J. Encarnacao, H. Hagen, A. Kaufman, S. Klimenko, G. Nielson, F. Post, D. Thalmann , Academic Press.

[TUF83] E.R. Tufte (1983), The Visual Display of Quantitative Information, Graphic Press, Cheshire, Conn.

[WEH90] S. Wehrend and C. Lewis, A problem-oriented classification of visualization techniques. In Proceedings IEEE Visualization '90, pp. 139-143.

[WOL93] R.S. Wolff and L. Yeager, Visualization of Natural Phenomena, 1993, Springer Verlag (TELOS).

[WWW1] http://web/msi.umn.edu/WWW/SciVis/whatisviz.html

HyperVis Table of Contents