Hi there!

I have a confusion on the idea of what a 3D image really is from the hologram reconstruction.

Sometimes the paper talks about a 3D volumetric image from a stack of amplitude intensity images. My understanding is that you reconstruct a holograph looping at different distances with some objects in focus and other objects out of focus. The focused objects are found in different distance plane .This is done automatically with algorithm to find the highest intensity change and thus the focus plane of each objects, this gets you the Z position of each object in the field and thus you have an volumetric representation of the images from one single hologram.

however the idea of 3D always seems to be combined with the phase contrast images. What i understand is that the phase images somehow give you a depth information. I know that arctan(real/imag) gives you the phase image pixel intensity but i don't know how the depth information is obtained. 

How exactly does phase image, giving the depth information of the overall images, combined with stacks of amplitude imaages, give you an 3d volumetric image?

Do you really need to have a phase images to have a volumetric 3d stack? Can't you just find the volumetric stack through the first strategy I mentioned where only amplitude image stack was used?

Many thanks for all your help!