03-12-2018 10:51 AM
Hello,
I have a beam intensity profile data (Gaussian Beam) . I want to calculate 1/e^2 value. From that value want to calculate the beam diameter.
Please need a help. Looking for quick solution.
Thanks
03-12-2018 12:37 PM - edited 03-12-2018 12:40 PM
1/e^2 is a constant and independent of any data.
There are old discussions showing how to fit 2D gaussian data to get parameters. Can you attach your code and some typical data?
03-12-2018 12:52 PM
Here, is a VI. Which is produced Beam profile using XY graph and image averaging.
I saw that post but i din't get it.
You can load the image and will generate XY graph which is located all the way down.
03-12-2018 01:53 PM - edited 03-12-2018 01:56 PM
I am sure you could have cropped that image to the interesting part (a few percent of the pixels), right? Fitting millions of background pixels is not useful and very expensive.
@parthpanchal07 wrote:I saw that post but i din't get it.
Which post (link!) and what part didn't you "get"?
@parthpanchal07 wrote:
You can load the image and will generate XY graph which is located all the way down.
An image can be represented by a 2D array in an intensity graph. A XY graph is not suitable for image data. I don't have IMAQ and I don't need it for this.
03-12-2018 01:56 PM
Hi,
Yes, you are right. Do you have any idea how can I remove the background?
and what about the calculation of 1/e^2 measurement?
03-12-2018 01:59 PM
I didn't get it about gussian fit model and how can I used it over here.
I
03-12-2018 02:00 PM
@parthpanchal07 wrote:
and what about the calculation of 1/e^2 measurement?
You still have not explained what that means.
You can calculate 1/e^2 with any calculator (= 0.13533) and that's not very interesting.
I am sure you have something else in mind. Please be specific.
03-12-2018 02:04 PM
@parthpanchal07 wrote:
Do you have any idea how can I remove the background?
You can use any image editing software to crop the image around the interesting part (position is irrelevant because you seem to only be interested in the size) and attach it. Should be a very small file.
Is the background zero or is there an offset? If there is an offset, leave sufficient background (e.g. twice the size of the spot) to be able to reliably fit for the offset.
03-12-2018 02:08 PM - edited 03-12-2018 02:18 PM
@parthpanchal07 wrote:
I didn't get it about gussian fit model and how can I used it over here.
What part is not clear to you?
A zip file with the code, the model, and a sample image is here. Should run out-of-the-box. All you need to do is get the width in your desired units.
03-12-2018 02:33 PM
Here,I have attached the image. That's what I want mean.