![background image](http://html1.mh-extra.com/html/zeiss/elyra-7/elyra-7_operating-manual_3332235253.webp)
ELYRA 7
Left Tool Area and Hardware Control Tools
ZEISS
03/2019 V_02
000000-2262-999
249
5.3.21.2.4
Fusion
After Registration of the Views, the
Fusion
process
can be performed (Fig. 423). This will lead to one
Z-Stack or Z-Stack over time.
You can choose the following methods using the
drop-down menu:
None
: no Fusion is performed after Registration
and Dual Side Fusion.
Mean Fusion
: The pixel in the fused image is
determined by averaging the intensity level of the
pixels for the involved views.
Mean Multiview DCV
: This option is only available if the dataset has not undergone
deconvolution before. The pixel in the fused image is determined by averaging the intensity level of the
pixels of the involved views. After fusion, a deconvolution is performed. The
Deconvolution
interface
becomes available when chosen (see above).
When all Z-stacks (hence all views) are fused into one Z-stack, the intensity values at the Z-stack borders
can lead to visible edges. To prevent this, the borders of all Z-stacks can be faded during the Fusion
process. With the
Blending
slider and input box with arrows, the amount of blending (smoothing) can
be controlled. The value for
Blending
should be as low as possible, with typical values from 20 to 80; but
this needs to be tested for every dataset.
It is possible to define which portion of the Z-stacks of the contributing views of the experiment shall be
used for the fusion process. To do so the checkbox
Fusion Subset in Z
must be activated. The view can
be chosen with the according dimension slider of the
image control area
, and is displayed as
Current
View
. For each of the views the z range can be chosen with the spin boxes or the sliders from each side.
Alternatively a click on the
Mouse Click
button and afterwards onto the image (e.g. in the
Gallery
View
) can be used to select the images to define the Z-range.
Whiles the whole Z-stacks are still used for registration (e.g. fiducials for Landmark based registration
remain accessible), only the image information of the selected Z-range will be used for
Fusion
.
Portions of the Z-stacks, highly scattered and/or blurry content, can be discarded from the resulting fused
image, to overall enhance the image quality. It must be considered for later deconvolution, that the
calculated point spread function (PSF) of a multiview experiment, is based on the assumption that all
views equally contribute to all areas of the image. This is not the case when using
Subset Z
and might
result in artefacts of the deconvolution result.
If you want to keep the registered image as an interim result, even if the registration and fusion are done
in one routine, activate the
Save Unfused Data
checkbox.
Fig. 423
Processing – Lightsheet Processing.
Multiview Processing, Fusion