Thread: Galaxy imaging
View Single Post
  #14  
Old 24-12-2014, 10:58 AM
Amaranthus's Avatar
Amaranthus (Barry)
Thylacinus stargazoculus

Amaranthus is offline
 
Join Date: Jan 2014
Location: Judbury, Tasmania
Posts: 1,203
As kind of noted above, if you are pushing to 1 arcsec per pixel or lower, which is required for detail on smaller galaxies, then you typically need to 'beat the seeing'. A useful way of doing this is deconvolution, but for this to be most effective, it needs really low-noise data (so lots of subs and/or long ones - ideally subs that get you well above the noise floor, but not too long so as to introduce tracking problems etc.). There is no bootstrap here!

Drizzling can also help, especially if you have a larger image scale and yet your target object is relatively small. You can do this by cropping the frame down to focus just on your target, and then drizzling by 2-3 times. This requires lots of subframes (16+, more the better) and dithering. But it can be effective.

This is an example of NGC253 that I imaged at an ultra-short 384mm FL, using a CCD with 8.3 micron pixels (so the image scale was >4 arcsec/pix). Yet when 3x drizzled, it looked pretty decent!
http://www.astrobin.com/118924/
Attached Thumbnails
Click for full-size image (NGC253_384mmFL.jpg)
197.1 KB46 views
Reply With Quote