Closed Shrediquette closed 8 months ago
More and more people are requesting this feature. The problem from my point if view is that it will be hard to properly display bitmap masks in Matlab. Transparent overlays will not be possible anymore.
I think the best solution would be to allow for two types of masks simultaneously: the polygon mask as it is now, and an additional bitmap mask that has limited display features (it will just be multiplied with the original image, so masked regions appear black).
I will simply add an extra input variable to all relevant functions containing the bitmap mask.
The bitmap masks should be generated outside Matlab. But maybe I can add an extra small GUI that allows to apply several steps of imopen, imerode etc. to the images. And functionality to calculate the average intensity and process that etc. Would be awesome if the user could drag and drop and rearrange processing items to a list... Is that possible with the "new" app designer?
Or even automatic functions from other contributors like this one: https://de.mathworks.com/matlabcentral/fileexchange/59655-pod-based-background-removal-for-particle-image-velocimetry
Displaying two pixel images with transparency seems to be possible: https://de.mathworks.com/company/newsletters/articles/image-overlay-using-transparency.html
Gabor filters for automatic masks: https://de.mathworks.com/content/dam/mathworks/tag-team/Objects/p/88395_93008v00_Texture_Gabor_Filters_2016.pdf
All this would be possible with pixel masks...
https://de.mathworks.com/help/images/Segment-Image-By-Drawing-Regions.html
Maybe use preexisting Matlab Apps?
I should definitely use the built-in ROI tools of Matlab. These may already combine pixel masks and polygon masks if I get it correctly:
https://de.mathworks.com/help/images/subsample-or-simplify-roi.html
Implemented in 3.0
PIVlab uses polygons to store masks. These are converted to BW just before the PIV analysis. Maybe it is feasible to change all the mask handling to pure bitmaps? That would make it much easier and straight forward for users to draw their own masks.