Image Processing Reference
In-Depth Information
data mapping
glyph instantiation
rendering
glyph properties
windowing
exponentiation
mapping
1
1
1
w
left
w
right
0
0
0
ouput'
ouput”
min
variate
max
1
1
0
0
i
Fig. 13.2
Each data variate is subject to three stages of data mapping: windowing, exponentiation
and mapping. The values are mapped to different glyph properties and used to instantiate the
individual glyphs. Finally, the glyphs are rendered in their spatial context
Similar to Ward [
24
], Lie et al. consider it useful that the glyphs expect normalized
input from the depicted data variates such as values in the range
. During data
mapping, the authors identify three consecutive steps. First, the data values within
a user-selected range
[
0
,
1
]
are mapped to the unit interval. Values outside
this range are clamped to 0 or 1, respectively. This allows to enhance the contrast of
the visualization with respect to a range of interest (sometimes called windowing).
A natural default choice for this step would be a linear map between
[
w
lef t
,
w
right
]
[
w
lef t
,
w
right
]
and
, but also other forms of mapping could be considered (for example, a
ranking-based or discontinuous mapping). After the windowing, an optional expo-
nential mapping
e
[
0
,
1
]
x
γ
can be applied in order to further enhance the contrast
on the one or the other end of the spectrum. Finally, a third mapping step enables
the user to restrict or transform the output range that should be depicted by a glyph
property. Here, also semantics of the data variates can be considered (compare to
the usage guidelines of Ropinski and Preim [
19
]). Using a reverse mapping, for
instance, smaller data values that are possibly more important can be represented in
an enhanced style while larger values are deemphasized.
Several considerations are important for the instantiation of individual glyphs.
When using a 3D glyph shape, one has to account for possible distortions introduced
when viewing the glyph from a different point of view [
9
]. In order to avoid this
problem, Lie et al. suggest to use 2D billboard glyphs instead.
1
In certain scenarios,
however, it makes sense to use 3D glyphs, for example, when depicting a flow field
via arrow glyphs. Another challenge in glyph design is the
orthogonality
of the
different glyph components, meaning that it should be possible to perceive each
visual cue individually (or to mentally reconstruct them as suggested by Preim and
Ropinski [
19
]). When representing a data variate by glyph shape, for example, this
affects the area (size) of the glyph as well. Accordingly, such effects should be
normalized
against each other, for instance, by altering the overall glyph size in
order to compensate for implicit changes of the glyph shape.
(
x
)
=
1
A billboard is a planar structure placed in a 3D scene, which automatically adjusts its orientation
such that it always faces the observer.
Search WWH ::
Custom Search