Find FWHM: Pixel Values, Gaussian Curve, Standard Deviation

[bugra]

Hi,
I've been searching on google for minutes but couldn't find out a consistent way so far. I mean, it seems everybody has something to say about it :)

My data consists of pixel values (ADU's). I am writing a script to obtain fwhm of seeing disk.

By definition of the gaussian curve, i should only calculate the standart deviation of these values theni get fwhm=stdev*2.355 .
Am I correct?

Or should i obtain a gaussian fit then i calculate fwhm as its name suggests?

Thanks in advance,
bugra

 

[EnumaElish]

If you do it both ways, and the answers are different, what does that tell you about the fitting algorithm that you used?

 

[bugra]

for multiple sources on the same frame, should fwhm be same?

 

[chemisttree]

I would think the actual value of FWHM is less important than the rate of change near perfect focus.

 

[pmacias]

Hi Bugra,

To find the FWHM (full-width at half-maximum) of a Gaussian curve, you can use either method you mentioned. Both approaches should give you the same result.

Calculating the standard deviation of the pixel values and multiplying it by 2.355 is the simplest and most straightforward method. This is based on the fact that the FWHM is equal to approximately 2.355 times the standard deviation of a Gaussian curve. So yes, you are correct in your approach.

Alternatively, you can also use a Gaussian fit to the data and calculate the FWHM from the fit. This method may be more accurate if the data does not perfectly follow a Gaussian distribution. However, it requires more steps and may be more complex to implement in your script.

In summary, both methods are valid and should give you the same result. It ultimately depends on the accuracy and complexity you require for your script. I hope this helps!