SPATIAL DATA

Spatial data is any type of data that directly or indirectly references a specific
geographical area or location. sometimes called geospatial data or geographic
information, spatial data can also numerically represent a physical object in a
geographic coordinate system.
 In creating a visualization of spatial data, attributes of the data must match attributes(location) on the
screen.
 It may involve many forms of transformations , including scaling, rotation, translation,shearing, and
projection.
 In this the wide range of techniques that have been applied to spatial data according to the
dimensionality are:
 One-Dimensional Data.
 Two-Dimensional Data.
 Multidimensional Data.
 Dynamic Data.
 One – Dimensional Data
 It is often the result of accumulating samples or readings of some phenomenon while moving along a path in
space.
 For Example, A drill-hole sample with mineral & ore content.
 This sort of sampling is called “Probe” when exploring structures of higher dimensions.
 Given a one-dimensional sequence of univariate data(only one value per data item), we can map the spatial
data to one of the screen dimension(to form a line-graph) or to the color of a mark or region along the spatial
axis.
 Data needs to be scaled to fit within the range might be reserved for other aspects of the visualization, such as
axes, labels and key.
 The general structure of the alogorithm to generate such a visualization will use parameters for the bounds of
both data and display spaces.
 datamin,datamax are computed as the minimum and maximum values for the data,
 Count indicated no. of data points to be displayed.
 Section of the display that will hold the visualization is defined by the rectangle(xmin,ymin,xmax,ymax)
 The algorithm to render a line graph from a set of Data

 drawLineGraph(data[],count,xMin,xMax,yMin,yMax)
 dataMin = minimum(data,count)
 dataMax = maximum(data,count)
 xFrom = xMin
 yFrom = yScaling(dataSet[0],dataMin,dataMax,yMin,yMax)
 for i=1 to count //for each data value
 xTo = xScaling(i,count,xMin,xMax)
 yTo = yScaling(dataSet[i],dataMin,dataMax,yMin,yMax)
 drawLine(xFrom,yFrom,xTo,yTo)
 xFrom = xTo
 yFrom = yTo

 xScaling(value,count,min,max)
 return min + index*(max-min)/count

 yScaling(value,min,max,yMin,yMax)
 A bar graph is similar to a line graph,except that each datapoint is replaced by a coloured rectangle whose
height is proportional to the value.
 The rectangle is usually centered on the spatial attribute of the data, and its width is often uniform.
 If the data set is multivariate, e.g., it contains more than one variable or value per data entry, we can expand
on any of the univariate strategies using juxtapositioning or superimpositioning.
 Juxtapositioning:
Juxtaposition is an act or instance of placing two elements close together or side by side. This is
often done in order to compare/contrast the two, to show similarities or differences
 For line graphs, it means that visualization
consist of a stack of non-overlapping graphs
(if variables with different scales) or a graph
containing plots of two or more variables.
 TWO-DIMENSIONAL DATA
 Data with two spatial dimensions get visualized predominantly by mapping the spatial
attributes of the data to the spatial attributes of the screen.
 Result can be of one of the visualizations.
1. An Image results if a single data value at each location is mapped to color and all
intermediate pixels are coloured via Interpolation.
2. A Rubber sheet results if the data, whether regularly or irregularly spaced, is mapped to
the height of a point in three dimensions, with the points trianguled so that surface can be
formed.
3. A Cityscape is formed by drawing three-dimensional objects(generally boxes) at
locations on a plane, where the data can control the attributes of the graphical objects(i.e.,
height & color), such as seen.
4. A Scatterplot results if, at each location on the plot, the data value(s) control the color,
shape, or size of a marker.
5. A Map results if the data contains linear and area features , as well as point objects.
 Linear feature – road or stream represented by a sequence of coordinates, which are plotted
as a series of line segments.
 Area feature – lake or political boundary is represented as a closed contour, a set of
coordinates where first and last points are same.
6. A contour or isovalue map conveys boundary information extracted from an image depicting
a continuous phenomenon such as elevation or temperature.
Isovalue means “single value” contour on such a map indicates boundary between points above
this value and points below the value.
 A contour plot is a graphical technique for representing a 3-dimensional surface by
plotting constant z slices, called contours, on a 2-dimensional format. That is, given a value
for z, lines are drawn for connecting the (x,y) coordinates where that z value occurs.
 A Contour Plot is a graphic representation of the relationships among three numeric
variables in two dimensions. Two variables are for X and Y axes, and a third variable Z is for
contour levels. The contour levels are plotted as curves; the area between curves can be
color coded to indicate interpolated values.
 Rubber Sheet Visualization A Cityscape showing the density of airtraffic at a timeperiod.

Contour lines of image of hydrogen molecule.

Map with linear and area feature

 Probing Two-dimensional data
 Besides visualization techniques that display entire data set, we choose one-dimensional subsets, projections,
or summarizations of the data. Some of the projections options are:
1. Frequency Histograms: Done by computing the frequency at which values or subranges of values occur in
the data, and then display this information as a bar graph. Considerations used are,
how to decide the no. of subranges to use
where the breaks between subranges should be.
Features of the data can be lost if it is provided incorrectly, and performed with the best when someone
who knows data characterization would be better.
2. Row & Column Aggregations: Visually depicting summarizations of the rows & columns of an image is a
useful mechanism for describing the boundaries and regions of low & high variability.
Many descriptors are used for this such as sum, average, mean, median, standard deviation,
maximum and minimum values.
 Linear probes:A one-dimensional data set can be likened to drill-hole for mineral exploration,a line is passed
through the data, and the values encountered are visualised with one of the dimensional techniques.
Foer this, we use mathematical tools , parametric equations and bilinear Interpolation. For Ex.,a
parametric equation of the probe using input from the user (either a pair of points or point in a direction), assume
two sets of coordinates,P1 & P2
The parametric equation of the line segment joining the points is
p(t) = P1 + t(P2 – P1)., where 0<=t<=1.0 by choosing value of “t” within the range.
 THREE-DIMENSIONAL DATA
 Like two-dimensional data, three-dimensional data may be either discrete samples of a continuous
phenomenon or a structure best described by vertices, edges and polygons which are many visualizations of
science and engineering data which contains a combination of these data representations.
Basic visualization techniques for this type of data are:
 Visualizing Explicit surfaces.
 Visualizing Volume data
 Visualizing Implicit surfaces.
 Visualizing Explicit Surfaces
 An Explicit Surface is one that has been defined in one of two ways.
1. A list of three-dimensional vertices, a list of edges, i.e., connections between the vertices,
and a list of planar polygons.
2. A set of parametric equations for defining the x-, y- and z-coordinates of points on the
surfaces, along with an interconnection strategy (triangular or rectilinear grid) for computing
the edges and patches.
3. For Example, a unit cube can be represented
as ,
Every edge is shared by exactly two
faces, and that every vertex is a member
of three edges
 The parametric form for a unit cylinder aligned with the y-axis can be defined as follows:
y = 1.0 x = cos o Z = sin o 0<=0<=2pie (top)
y = 0.0 x = cos o Z = sin o 0<=0<=2pie (bottom)
z=h x = cos o Z = sin o 0<=0<=2pie,0.0<=h<=1.0 (sides)
The Visualization of the spatial data defined on an explicit surface depends on whether the values to be conveyed
are associated with the vertices, edges or faces. Examples are:
Temperature or stress at a point (vertex information)
Strength of attraction for a chemical bond (edge information)
ground cover for a map region (face information)
 Visualizing Volume data
 As pixels are two-dimensional visualization, voxels , or volume elements, are to three-dimensional
visualization.
 Volume data is generally a sampling of a continuous phenomenon , and can be acquired via
sensors(Ex.,tomographic data set) or generated via simulations(ex., computational fluid dymanics).
 In each case, we have one or more dimensions with regular or irregular positions, and the goal is to convey to
the viewer the structure, patterns and anomalies within the data.
 Approaches to visualizing volume data fall into categories:

1. Slicing Techniques - Slicing Volume data with cut planes.

2. Isosurface Techniques - Isosurface Extraction using Marching cubes.
3. Direct volume Rendering - Direct volume visualization techniques.
 Slicing Techniques: Using a cut plane, either aligned with an axis or arbitrary oriented,
probe the data to extract a two-dimentional slice of the data, and then use one of the two-
dimensional spatial data visualization.
 Isosurface techniques: Given a user-specified value, generate a surface description and
visualize it using one of the explicit surface visualization techniques.
 Direct volume rendering: Either cast rays into the volume and compare a pixel value based
on the data encountered by the ray, or project each voxel onto the projection plane using
some method of accumulating effects on pixels.

In all these approach resampling is the essential component and coordinate

system transformations, are essential tools in the visualization of spatial data
 Slicing volume data into cut planes
 One strategy we can follow is to probe the three-dimensional data to create a subset of data with a lower
dimension.
 A popular technique for volume data is the use of cut planes, where the data block is sliced by a plane with a
given orientation and position, and the data that the plane intersects are mapped to the plane for display.
 Many variations on this technique exist, and each can provide significant details about the interior of the
volume includes.,
 Non-planar slices.
 Consecutive slices in varying orientation to remove blocks of the data.
 Stacked slices displayed simultaneously.
 Orthogonal slices displayed simultaneously.
 Isosurface Extraction using Marching cubes
 Marching cubes is an algorithm developed for rendering isosurfaces in volumetric data.
 Basic notion is that we can define a voxel(cube) by the values at the eight corners of the
cube.
 If one or more corners of a cube have values less than the user-specified isovalue , and one
or more values greater than this value, then voxel must contribute some component of the
isosurface.
 By determining which edges of the cube are intersected by the isosurface, we can create
triangular patches, that divide the cube between regions within the isosurface and regions
outside.
 By connecting the patches from all cubes on the isosurface boundary, we get a surface
representation.
15 Voxel configurations for Marching Cubes
 Marching Cube Algorithm
Marchingcubes (Isovalue)
For EachCell
do if Isosurface passes through Cell
Then
Classify each vertex as inside or outside.
Determine index of the 15 cell types.
Get edge list from the table[index]
Interpolate the edge location
Create small polygons for surface within Cell.
 Problems and Alternatives: One main problem is amount of memory needed to store the resulting surface, as
each boundary cube can generate upto 4 sub-pixel facets., the result can be quite large.
We can reduce this by sharing vertices & edges are even merging coplanar patches into large
facets., another solution is try to fit parametric surfaces to groups of boundary points
 Direct Volume Visualization techniques
 It means that no three-dimensional polygons are created for use with traditional graphics rendering techniques,
rather pixels in the resulting image are computed in an individual basis,, either by casting rays through the
pixel through the volume, or by projecting voxels onto the plane of projection.
 Initially process starts transforming the position of the voxels into the viewing coordinate system, following
the same procedure ued in the traditional graphics pipeline.
 The Viewer must represent the,
1. View Reference point( the origin of plane of projection)
2. View Direction( the normal to the plane of projection)
3. The height & width of the image to be projected on the plane of projection
4. The distance from the camera to the plane of projection.
 Once the Voxels have been positioned, we have option of
 Forward mapping – project each voxel onto the plane of projection & determine which pixels will be affected
and in what way.
 Inverse mapping – also called as ray casting. Send a ray from each pixel from the plane of projection through
the volume, sampling values along the ray, and determining the resulting value for each pixel.
 For each mapping, we have a number of problems like,
1. How to deal with pixels that are influenced by multiple voxels.
2. How to handle pixels to which no voxel directly maps.
3. How to deal with the fact that voxels usually project to positions between pixels.

 For inverse mapping, we have similar problems,

1. How to select the number of points to be sampled along the ray.
2. How to compute the value at these points, which usually fall between voxels.
3. How to combine the points encountered along each ray
 Transfer functions :
An important issue in rendering the volume in determining the opacity and color associated with
particular data values, this process is called classification, and results in set of functions defining how the
opacity,red, green, and blue channels( or alternatively ,hue,saturation,, and value) will be set for a given voxel
value.
 Initially they may be set by analysing the data and determining the voxel values where significant transfers
occur. These make logical points for changes in color , opacity or both.
 Another important problem in direct volume rendering is computing the effects of lighting and shading. An
approximation is made , based on the calculation of gradients
(rate of change) in each direction.
 For a given (vx,vy,vz), we can estimate how rapidly the voxel is changing direction, namely (vx-1,vy,vz) and
(vx+1,vy,vz).
 Simplest gradient is difference between voxel and one of its immediate neighbours. Thus gx, the x-component
of the gradient , would just be vx-vx-1, referred as intermediate difference operator.
 The central difference gradient estimator uses the points on bothr sides of the voxel, rather than the voxel itself.
 Implicit surfaces
 Typical method of modelling surfaces is through parametric equations to define points on the
surface which can then be connected to form polygonal meshes.
 This representation is very useful for performing transformations and computing normal
surfaces.
 An alternative method is the use of implicit representations, where the surface is defined as
the contour of a function of two or more variables.
 Implicit representations have strengths in operations such as blending and metamorphosis,
and have been becoming increasingly popular.
 Metaballs, also known as blobby objects, are a type of implicit modelling technique, it
is like a particle surrounded by a density field, where the density attributed to the
particle decreases with distance from the particle location.
 Key to use metaballs is the definition of the equation for specifying the influence on an
arbitrary point from an arbitrary particle.
 Given a set of data points, we define a grid snd compute the density t each point based on the data in the
neighbourhood.
 Once the field is generated , any scalr field visualization technique can be used to render it, such as marching
cubes algorithm or direct volume rendering.
 Example of types of renderings that can be performed with implicit surfaces
 DYNAMIC DATA
 Flow visualization is the study of methods to display the dynamic behaviour in liquids & gases.
 Several images of fine particles of sand and wood cravings into flowing liquids where careful control of size
and distribution taken place.
 More recently computational fluid dynamics(CFD) has extended the abilities of scientists to study flow by
creating simulations of dynamic behaviour of fluids under a wide range of conditions.
 Result of these analysis is usually a 2D or 3D grid of velocity vectors, which may be uniformly or nonuniformly
spaced.
 The structure of these field dynamic data can take two forms:
1. Static field – there is one single, unchanging velocity field.
2. Time-varying field – may either have a fixed positions with changing vector values or both changing positions
& changing vectors.
 Definitions of Mathematics of particle advection
1. Pathline: Trajectory of a particle released into a flow field, assumes multiple time instances, this can be
achieved by long-term film exposure.
2. Streakline: Simultaneous of a set of particles continuously released from one or more locations.
3. Timeline: Position at an instant of time of a batch of particles that have been released simultaneously.
4. Streamline: A line through the velocity field that is tangent to the velocity field at every point.
5. Steady Flow: A flow field that does not change with time. For steadyflow, straklines,pathlines, and streamlines
coincide.
6. Particle advection: Computing the motion of partiocles through a flow field.
7. Vorticity: The curl of the velocity field, giving the magnitude and direction of angular velocity for each
particle in the velocity field