2
$\begingroup$

I have an image with some sources in it. I need to get the pixel size and beam size of that image. How can I calculate it?

Improve this question
$\endgroup$
4
  • $\begingroup$ Are you saying you already have a picture of a known part of the sky, and want to know the image size in degrees, or similar? $\endgroup$
    – Andy
    May 4, 2016 at 10:34
  • $\begingroup$ Infact I have a synthetic image with a source of known flux. Inorder to give it to a source extraction software as input, it require that the pixel size should be equal to one third of the beam size. And I don't know how to calculate the beam size in this case $\endgroup$
    – Rian
    May 4, 2016 at 10:47
  • 1
    $\begingroup$ From your description, it's impossible to know. You need to ask whoever prepared the image what the beam size is in pixels; there's no way to know otherwise. $\endgroup$
    – Peter Erwin
    Nov 5, 2018 at 12:43
  • $\begingroup$ Is it a FITS file? The beam size will likely be in the header $\endgroup$
    – usernumber
    Feb 3, 2020 at 8:55

3 Answers 3

Reset to default
2
$\begingroup$

If you know the angular size of your image (e.g., 10", 1' etc), then pixel size (in "/pixel units) is just image size / number of pixels. For beam size, depends a lot on the wavelength-range of your observations (for instance sub-mm/radio observations are conceptually very different from UV/optical), but to have an idea of the beam/psf size, you can fit a 2D gaussian/moffat profile to the emission of a point-like object (i.e., a star) in the field of your image.

Note that most of the time these info are contained in the header.

Improve this answer
$\endgroup$
0
$\begingroup$

Pixel size is usually (beamsize/3). Based on the axes and number of pixels, you can find out pixel size by doing pixel_size = image size(in arcsec) / number_of_pixels.

Usually, the image is done by first finding the beamwidth of the telescope and then putting 3-5 pixels inside a beamwidth in any axis.

Improve this answer
$\endgroup$
1
  • $\begingroup$ > Pixel size is usually (beamsize/3) This is completely arbitrary. $\endgroup$
    – user19333
    May 8, 2019 at 11:23
0
$\begingroup$

You can estimate the pixel size just from the number of pixels if you know the scale of your image or certain object in the image, but really you should already know this and the beamsize (which you could take a guess at but can't be sure). This information should be given with the image by whoever actually made it. If you have a data file it may be in the headers, or if you got the image from a website check the caption for details or an original source.

Improve this answer
$\endgroup$

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .