Finding and using the Point Spread Function via FFT in Labview

[Prodigalvanic]

Hello!

My partners and I are building an NSOM microscope and it is mostly finished. Now we need to find the Point Spread Function.

My questions is how do I find the PSF in labview and how do I use it to deblur in labview?

What follows is some background on what I have attempted so far.

I built a VI and tested it with the image, object and PSF provided at the top of the wikipedia page (see below). It made FFTs of the image and object then divided the image FFT with the object FFT. In other words it deconvolved the image by the object. I figure the result should look like the PSF provided. It does not it is totally black.
Neither can I deconvolve the provided image by the provided PSF to get the provided object.

I am new to all of this. I understand another approach is to do a gaussian fit to a surface of the image. Then I can deconvolve that with the image. I cannot understand how to do this as the resources assume too much prior knowledge.

How do I deconvolve an FFT by a gaussian?! The websites discusses use of a "kernel" but I can find no explanation of what a kernel is. (I assume it has nothing to do with my operating system).


A webs resource I consulted
http://zeiss-campus.magnet.fsu.edu/articles/basics/psf.html

The source of my test images
http://en.wikipedia.org/wiki/Point_spread_function

 

[Vailanor]

If you know of a resource that explains this in lay terms please let me know. Thanks for your help!Unfortunately, deconvolution using Labview is not an easy task. The best way to approach this would be to use a mathematical approach, such as a gaussian fit. You can find some resources online that explain how to do this in Labview, although they might be a bit technical. It's also important to realize that the PSF is something that needs to be empirically determined for your particular system, so it's not something that can easily be calculated in Labview. You can also try using a library like scipy, which provides a deconvolution function. This can be used with Labview as well, but requires some basic knowledge of programming in python. If you are still having trouble, you could try reaching out to someone in your local area who has experience with this type of thing. There may be a professor or lab technician who could provide more specific advice on how to get started. Good luck!