Resize 3d volume

If i want to resize 3d volume 256 x 256 x 26 to 184 x 220 x 26 , is it right to use Resample Sacalar/Vector/DWI Volume module?
The Original volumes spacing is 1.25 x 1.25 x 4

The reason to resize 3d volume is that I’m going to extract the axial slices of volume to set of png file and use it for 2D CNN.

Yes, that should work.

Note that your CNN uses numpy arrays as inputs, so there is no need to force your data into compressed 8-bit (or even worse, 24-bit) png images. Instead, you can get numpy array directly from your volume using slicer.util.arrayFromVolume and save it to a npy file that you can directly use in your CNN.