Solved

Axial Point spread function

  • 2 January 2024
  • 3 replies
  • 128 views

Dear all,

I am currently working on extended depth of field and aim to plot the axial (z-stacks) Point Spread Function (PSF) using the Julia programming language. After generating different Huygens PSF data in Zemax for both in-focus and defocused scenarios, I intend to examine their axial sizes.

The Huygens PSF cross-sections have only x or y sections (refer to the attached Figures). My question is whether taking the Huygens PSF data and simply plotting it in Julia would provide the axial PSF.

Your insights on this matter would be greatly appreciated.

 

Thanks

icon

Best answer by MichaelH 9 January 2024, 18:55

View original

3 replies

Please any idea!

 

Userlevel 6
Badge +2

Hi Kibru,

Yes, you can simply use the PSF Cross Section with different thicknesses on the last surface in the LDE and then “stack” them all together to get the Y-Z or X-Z through-focus PSF.

The few things to keep in mind are:

  •  You have to use Huygens PSF Cross Section (FFT PSF is calculated in pupil space and therefore there is no “common” coordinate system between different PSF slices)
  • For off-axis points that have a large image-space angle, you will need to shift each PSF slice so the Center coordinates align
  • You need to make sure the Normalize box is unchecked

To automate this, I would suggest using the ZOS-API to change the defocus, run the Huygens PSF Cross Section, and create the PSF stack.  The 5 supported languages for the ZOS-API are Python, C#, Matlab, Mathematica, and C++ (the first 4 connect via .NET while C++ connects using COM); I have not seen any user on the forum who has dealt with Julia yet so it might take a little effort to connect the ZOS-API to Julia.  

@MichaelH Thank you very much.

Which one is used for depth of field? first surface or last surface thickness?
I used ZPL macro to change the object focus, run Huygens PSF, and create the PSF stacks.  But, I am not sure if  ZPL gives the same result as ZOS-API. 

Reply