Ticket #1944 (new feature)

Example: Thresholding or normalization for example: If a user has a 1000x1000x1000x1000 dimensional image: how would you easily provide him a way to run thresholding plane wise, cube-wise or hyper-cube-wise. Same holds for e.g. normalization and all other algorithms which a) can be used in several dimensions or b) are restricted to some dimensions.

Requirements:

• ImageJ2 Plugins should be able to be annotated with a preprocessor/viewprocessor where you have to specify the number of dimensions an img minimal/maximal must have, s.t. an algorithm can be used on it.

• If an ImageJ2 Plugin is annotated with such a preprocessor, a gui should appear in the configure dialog, where one can select the dimensions which shall be processed (e.g. if X,Y are chosen, all other dimensions are used to iterate over the X-Y planes).

• The X,Y etc. axis names need to be matched to the actual indices. Intervals need to be generated (  https://github.com/imagej/imglib/blob/knip/ops/src/main/java/net/imglib2/ops/operation/interval/binary/IntervalsFromDimSelection.java)

• The processing of the planes/cubes/hypercubes could be parallelized (see  https://github.com/imagej/imglib/blob/knip/ops/src/main/java/net/imglib2/ops/operation/SubsetOperations.java).

• In general: The framework should work an for abitrary algorithms, maybe it's a good a idea to use Operations here (see SubsetOperations again).