MultiModal Big Image Data Sharing and Exploration
The “Select Box” tools allows you to select a 3-D box, using the “TransformedBox” code of big. Use cases include documenting a region of biological interest or selecting a region to be further processed in a correlative microscopy setting.
The size of the box can be interactively changed using the graphical user interface of the tool. The rotation however is fixed to be aligned with the current viewing axes. Thus, to set the box rotation you need to configure the viewing axes before starting the tool, using the usual BigDataViewer functionality. In particular, it may sometimes be useful to align the box with the axes of one specific image. To achieve this, press P
to show the BigDataViewer control panel and check the image of interest as the “current” one. Then press Shift + Z
(or X
or Y
) to align the viewer with one of the image axes.
Currently, the only output of this tool is the location of the box, written in the log window.
The box location is given in the following way:
Note that the center is just given for convenience, but could also be computed using the box size and the affine transformation, via center = affine( size[0]/2, size[1]/2, size[2]/2 )
. To locate any voxel within the box in the global coordinate system, you may affine transform any voxel within the interval [ 0, 0, 0 ]-[ size[0], size[1], size[2] ]
.
[ 0, 0, 0 ]-[ 1, 1, 1 ]
. That means the box could be fully specified by just one affine transformation.