About Thredgeholder
Pro v1.2
Thredgeholder Pro is an
advanced edge detection plugin, simple in
operation but extremely effective.
Using Thredgeholder Pro
Thredgeholder Pro selectively
detects edges for a number of different
'channels' in the image, such as the red
component, the degree of saturation, or
the 'a' channel in the LAB colour model.
The image
on the left shows the interface
when the plugin is first run.
At the top of
the controls are four buttons
for selecting which page of
controls you want to change;
HSL, RGB, IUV or LAB filtering.
Each page sets the parameters
for a set of three channels.
All sets of controls
are active at the same time
when filtering; the HSL/RGB/IUV/LAB
selection is only for selecting
the page of controls you want
to change.
For each channel
(HSL/RGB/IUV/LAB) there is an
activation checkbox. If the
checkbox is ticked, that channel
is filtered. If not ticked,
that channel plays no part in
the filter's results. By default
only the Luminance channel is
enabled on filter startup.
Each channel
has two sliders.
'Threshold'
determines the degree of
sensitivity to edges. Low
values will tend to show
background noise. By increasing
the threshold you can select
only the more prominent
edges in the image.
'Brightness'
specifies the lightness/darkness
of the edges. Higher values
give a softer image, whereas
low values create higher
contrast. Setting Brightness
to the same value as Threshold
gives an absolute threshold
of black and white. Brightness
cannot be set to a value
lower than Threshold, and
vice versa.
|
The following illustrates
using Thredgeholder Pro to select the edges
in this image...

. |
|

With only the
Red channel active there's strong
definition of the glass but
the ribbons are less visible |
|
|
|

|
|

Deactivating
the Red channel and activating
filtering on the V channel,
the glass definition is virtually
lost but the ribbons are clear
and strong |
|
|
|

|
|
By
activating both the Red and
V channels, we get the best
of both worlds with clear glass
edging and ribbon definition |
|
|
|
Hints for use
Wherever possible, use high
quality source images. JPEGs have low-level
colour distortion created when the image
is compressed. This introduces a lot of
noise to sensitive edge detection, especially
in the Hue channel, which appear as rectangular
areas of edges.
The preview image shows
a scaled down version of the Source data.
The settings are applied to this downscaled
image during the preview. For an accurate
preview of the filter when applied, set
the preview zoom to 100%.
The 'Luminance', 'i' and
'L' channels produce similar results, but
with different areas of sensitivity. eg.
The 'L' channel is effective at finding
detail in areas of shadow. |