Primary channel and masks at the orthomosaic stage

[nprp]

Hi,
I am processing multicamera system and some images are overexposed. There is certain chance such areas are not overexposed in neighbouring images. My idea is to apply masks to overexposed areas so that they won't be used in orthomosaic.
The question arise: does MS use EACH channel masks at orthomosaic stage or it only uses PRIMARY CHANNEL masks to all the channels?

As I can suppose from seamlines, MS only uses primary channel masks.
Is there any option to independently apply masks to separate channels at the orthomosaic stage?  Masks are generated for each image, separate channels are stored in separate images.

I have a backup plan - to generate 5 single-channel orthos (with different primary channels) and then merge outside MS but I think it is an interesting issue to discuss here.

[Alexey Pasumansky]

Hello nprp,

If you are using a multi-camera system layout, then masks should be applied to each image in such system independently.

Or do you mean to import the masks from the external source for all the images at once?

[nprp]

Hi Alexey,
I've looked again at the materials.

It's clear now that MS use independent masks at orthomosaic stage. However, at the aligning stage some areas can be omitted because of masks and that may result in holes at the orthomosaic.

Then, I suppose, in order to hide overexposed areas, masks should be applied after aligning stage.

[Alexey Pasumansky]

Hello nprp,

Masks are optional during image matching stage - you can select Apply Masks To option in Align Photos dialog, if you need to apply the masks to tie points or keypoints.

To avoid holes in the surface, you should enable Extrapolated option in the Interpolation field when building mesh or DEM (depending on the source surface to be used for the orthomosaic reconstruction).