in PixelMath, just as a symbol can hold a constant value, it would be nice if it could also hold an image identifier. That would add versatility to the reuse of pixelmath expressions in diverse processing trains.
e.g.
(symbols Fn)
Fn="image21";
med(Fn)*$T+med($T)*Fn
then simply changing the rhs of the definition of Fn, this expression could be rapidly applied elsewhere. Of course this is a very simplified example involving just one image identifier. Ability to handle multiple image identifiers would be great.
AlexW
