treatment correlation box near mask

[Michele Trevisson]

Hello William,

I'm a postdoctoral researcher from UWA. I have been using your software for a while and was wondering whether you could explain what happens when a correlation box overlaps with a mask.

Is the velocity vector calculated based only on the part of the correlation box not covered by the mask? And if so, do you move the velocity vector to the centroid of the unmased area of the correlation box?

Thanks,

Michele

[William]

Dear Michele, there is a discussion here:

https://groups.google.com/g/pivlab/c/6Ijm-McAihw/m/SY170t2mBAAJ

Especially my last comment might help.