1. Visualization
CSE 694L
Roger Crawfis
The Ohio State University

2. R. Crawfis, Ohio State Univ.
Introduction
Getting acquainted - teams
Current studies or major
Hometown
A interesting data problem
2/24/2019
R. Crawfis, Ohio State Univ.

3. R. Crawfis, Ohio State Univ.
Outline
Scientific Visualization
Data Topologies and Data Sources
1D and 2D Visualization Algorithms
Overview of 3D Visualization techniques
Iso-contour surfaces
Volume Rendering
Transfer functions and segmentation (2D)
Flow Visualization Algorithms (2D)
2/24/2019
R. Crawfis, Ohio State Univ.

4. R. Crawfis, Ohio State Univ.
Outline (con’t)
Volume Rendering
Optical Models
Algorithms
Transfer Functions
Global Vector Field Visualization
Virtual Environments
The CAVE
Data or Information Visualization
Overview of perceptual issues
Brushing
Focus + context
Successes
2/24/2019
R. Crawfis, Ohio State Univ.

5. R. Crawfis, Ohio State Univ.
What is Visualization?
Understanding of data
Insight into information
Presentation and sharing of insights.
2/24/2019
R. Crawfis, Ohio State Univ.

6. R. Crawfis, Ohio State Univ.
Data Sources
Computational Science
Data Acquisition / Imaging
Historical Observation
Survey, Census, etc.
2/24/2019
R. Crawfis, Ohio State Univ.

7. Data Topologies - Structured
Cartesian
x j+1 = xi + 
Regular or Uniform
x j+1 = xi + x
Rectilinear or Perimeter
x j+1 = x(i)
2/24/2019
R. Crawfis, Ohio State Univ.

8. Data Topologies - Structured
Curvilinear
x j+1 = x(i,j)
y j+1 = y(i,j)
Curves may intersect in i or j
Curves may not cross in i or j
i=3
i=0
j=1
2/24/2019
R. Crawfis, Ohio State Univ.

9. Data Topologies - Unstructured
Unstructured or cell data or finite-element data
Tetrahedral
2/24/2019
R. Crawfis, Ohio State Univ.

10. Data Topologies - Unstructured
Hexahedral
2/24/2019
R. Crawfis, Ohio State Univ.

11. Data Topologies - Unstructured
Finite-element zoo
2/24/2019
R. Crawfis, Ohio State Univ.

12. R. Crawfis, Ohio State Univ.
New Data Topologies
Improved data topologies offer huge potential for savings in computational science
Hierarchical models are becoming more common
2/24/2019
R. Crawfis, Ohio State Univ.

13. R. Crawfis, Ohio State Univ.
New Data Topologies
Hierarchical
Multi-Block Curvilinear
N-sided Polyhedron where n>6
Multi-Grid or Adaptive Mesh Refinement
2/24/2019
R. Crawfis, Ohio State Univ.

14. R. Crawfis, Ohio State Univ.
Working with data
Reconstruction is critical
Useful Image Processing operations
Histogram
Data mappers
Region of interest selection
Edge detection
Noise removal or blurring
2/24/2019
R. Crawfis, Ohio State Univ.

15. 1D and 2D Visualization Techniques
CSE 694L

16. R. Crawfis, Ohio State Univ.
1D Visualization
y = f(x)
Line Charts
Curve Fitting
Smoothing or Approximation
2/24/2019
R. Crawfis, Ohio State Univ.

17. R. Crawfis, Ohio State Univ.
1D Visualization
Non-cartesion coordinate systems
2/24/2019
R. Crawfis, Ohio State Univ.

18. Basic 2D Visualizations
Scalar Data on a Regular Grid
Surface plots (2D graphics)
2/24/2019
R. Crawfis, Ohio State Univ.

19. R. Crawfis, Ohio State Univ.
2D Visualizations
Contour Lines - f(x,y) = constant
2/24/2019
R. Crawfis, Ohio State Univ.

20. R. Crawfis, Ohio State Univ.
2D Visualizations
Psuedo Coloring
2/24/2019
R. Crawfis, Ohio State Univ.

21. R. Crawfis, Ohio State Univ.
2D Computer Graphics
Image formats and pixel limitations
Color Tables
grey-scale
hot to cold
perceptual
2/24/2019
R. Crawfis, Ohio State Univ.

22. R. Crawfis, Ohio State Univ.
Transfer Functions
Besides the basics, most tools allow you to define your own color mappings.
2/24/2019
R. Crawfis, Ohio State Univ.

23. R. Crawfis, Ohio State Univ.
2D Visualization
Vector Fields
Hedgehogs
Streamlines
2/24/2019
R. Crawfis, Ohio State Univ.

24. R. Crawfis, Ohio State Univ.
1D Visualization
Vector?
2/24/2019
R. Crawfis, Ohio State Univ.

25. R. Crawfis, Ohio State Univ.
2D Contouring
Continuous f(x,y)
Use steepest decent to find zero crossing (root) of the function f(x,y)-c
Follow contour from this seed point until we reach a boundary or loop back.
Direction close to f  z
Problems?
2/24/2019
R. Crawfis, Ohio State Univ.

26. R. Crawfis, Ohio State Univ.
2D Contouring
Discrete Data
Assume the Mean Value Thereom
Assume monoticity?
1D Analogy
5 Points
2/24/2019
R. Crawfis, Ohio State Univ.

27. R. Crawfis, Ohio State Univ.
2D Contouring
Given a quadrilateral
f(x,y) = 0.5
0.0
1.0
0.0
1.0
<0.5
>0.5
2/24/2019
R. Crawfis, Ohio State Univ.

28. Three-dimensional Visualization Techniques
CSE 694L

29. Visualization Algorithms
2D Slice planes in 3D
2/24/2019
R. Crawfis, Ohio State Univ.

30. R. Crawfis, Ohio State Univ.
2D Slice Planes
Orthogonal to a coordinate axis
Uniform Grids => image
Arbitrary
Specify the normal and a point
Produces a 2D unstructured grid
2/24/2019
R. Crawfis, Ohio State Univ.

31. R. Crawfis, Ohio State Univ.
2D Slice Planes
Mesh with colors at vertices
128x128x128 volume can produce over 50,000 triangles.
Mesh with 2D texture coordinates
Very slow if no hardware support
More precise transitions
Mesh with 3D texture coordinates
2/24/2019
R. Crawfis, Ohio State Univ.

32. R. Crawfis, Ohio State Univ.
2D Flip book of slices
Rather than view the 2D slice in 3D, rapidly play a sequence of orthogonal slice planes in a movie loop.
Sometimes difficult to build a complete mental model.
2/24/2019
R. Crawfis, Ohio State Univ.

33. R. Crawfis, Ohio State Univ.
3D Visualizations
Point plots
Animation can bring out the surface or pattern (MacSpin)
Depth Cueing aids in the depth perception.
2/24/2019
R. Crawfis, Ohio State Univ.

34. R. Crawfis, Ohio State Univ.
3D Visualizations
Spheres or cubes dispersed throughout the volume
color-coded
optional shape-controlled.
2/24/2019
R. Crawfis, Ohio State Univ.

35. R. Crawfis, Ohio State Univ.
3D Visualization
Iso-contour surfaces
2/24/2019
R. Crawfis, Ohio State Univ.

36. R. Crawfis, Ohio State Univ.
3D Visualization
Can add information about an additional variable
Here, two additional variables control the color.
2/24/2019
R. Crawfis, Ohio State Univ.

37. R. Crawfis, Ohio State Univ.
3D Visualization
Volume Rendering
2/24/2019
R. Crawfis, Ohio State Univ.

38. R. Crawfis, Ohio State Univ.
3D Visualization
2/24/2019
R. Crawfis, Ohio State Univ.

39. Material Classification
Drebin, Carpenter and Hanrahan
Material Probabilities
Levoy
Contour surfaces
MRI data
Skin versus Brain
Using Texture mapping
2/24/2019
R. Crawfis, Ohio State Univ.

40. R. Crawfis, Ohio State Univ.
Transfer Functions
Maps raw voxel value into presentable entities: color, intensity, opacity...
Raw-data  material (R, G, B, a, Ka, Kd, Ks,...).
Material  shaded material.
High gradient in the data values detects a surface and is used as a measure of its orientation.
2/24/2019
R. Crawfis, Ohio State Univ.

41. R. Crawfis, Ohio State Univ.
3D Visualization
Volume Rendering can mimic contouring.
Use a transfer function with an impulse at the contour level.
2/24/2019
R. Crawfis, Ohio State Univ.

42. What makes a good visualization?
Why are some things or fields portrayed the way they are?
Take map making for instance.
Which of the following is “better”?
2/24/2019
R. Crawfis, Ohio State Univ.

43. Different map projections
2/24/2019
R. Crawfis, Ohio State Univ.

44. R. Crawfis, Ohio State Univ.
2/24/2019
R. Crawfis, Ohio State Univ.

45. Different map projections
2/24/2019
R. Crawfis, Ohio State Univ.

46. Different map projections
2/24/2019
R. Crawfis, Ohio State Univ.

47. Different map projections
2/24/2019
R. Crawfis, Ohio State Univ.

48. Different map projections
2/24/2019
R. Crawfis, Ohio State Univ.

49. Different map projections
2/24/2019
R. Crawfis, Ohio State Univ.

50. R. Crawfis, Ohio State Univ.
2/24/2019
R. Crawfis, Ohio State Univ.

51. Different map projections
2/24/2019
R. Crawfis, Ohio State Univ.

52. Different map projections
2/24/2019
R. Crawfis, Ohio State Univ.

53. R. Crawfis, Ohio State Univ.
2/24/2019
R. Crawfis, Ohio State Univ.

54. Different map projections
2/24/2019
R. Crawfis, Ohio State Univ.

55. Different map projections
2/24/2019
R. Crawfis, Ohio State Univ.

56. Different map projections
2/24/2019
R. Crawfis, Ohio State Univ.

57. Different map projections
2/24/2019
R. Crawfis, Ohio State Univ.

58. Different map projections
2/24/2019
R. Crawfis, Ohio State Univ.

59. Different map projections
2/24/2019
R. Crawfis, Ohio State Univ.