RGB and U8 - improve typecasting and typecasting useability in Vision Assistant

[b.ploetzeneder]

This is something that is bugging me regularly. I use Vision Assistant to test out something on image data. That image data can come from all kinds of sources.. is modified by other programs, etc. What happens very often, is that the input data type switches between RGB and U8... even if the data itself may only be grayscale.

And then you pick a grayscale function on an RGB image, and get invalid image type errors.

I completely agree with displaying the error, but I think it should go a step further. That error occurs for two reasons:
(a) I clicked the wrong function - no big deal - thanks for the warning

(b) I want to actually perform an operation on the monochrome version of an RGB image. Which is the case most of the time, and then I use a color plane extraction, and usually the value plane. I think this is a step that should be automatically inserted by VA.

Or thinking it further.. about a way that unifies input data: Could there be an option where I could "Open image as U8" or similar?

.. I am aware that these are not the problems that occur a lot in production, but they occur all the time when doing quick-and-dirty prototyping, and that is what Vision Assistant is all about for me.

Welll... or do offer typecasting to RGB32. That way, I would call this as a first function in any script, and start with RGB32 all the time.

[altenbach]

Is this really a LabVIEW idea? If not, please have the moderator move it to the Vision idea exchange.

[Darren]

Any idea that has received less than 3 kudos within 3 years after posting will be automatically declined.