AstroBin saves small pieces of text information (cookies) on your device in order to deliver better content and for statistical purposes. You can disable the usage of cookies by changing the settings of your browser. By browsing AstroBin without changing the browser settings, you grant us permission to store that information on your device.I agree
An earlier tutorial (here) detailed the techniques for isolating untrailed stars and a sharp defined comet from the collection of image frames taken over many hours of a comet as it moves slowly over the starry background.
This tutorial recaps those steps and then discusses the best way to combine the "stars only" and "comet only" images to create the final image.
Images A to B is the process of isolating stars from the collection of "light" frames
Images E to D is the process of isolating the comet from the same collection of "light" frames
Image C is the final combined image containing sharp stars and comet. The full final image is shown here
All images should be calibrated with darks, flats and dark flats before stacking.
Image B is a "median" combined stack aligned on the stars. The median method ensures that:
a) Transient features such as satellite trails are removed
b) The moving comet is reduced to a smooth blurred artefact
Image C ("stars only") is generated from image B using a background extraction process that eliminates extended objects (in this case the blurred comet). The basic method is described by Hubl here. The basic steps are:
a) Duplicate the image - the second copy will be used to isolate the extended object and this will then be subtracted from the original.
b) In the second image copy use a Photoshop (or equivalent) filter to remove small point like objects (stars). Usually this is the "dust and scratches" filter.
c) Use Photoshops "clone stamp" tool to carefully eliminate residual blurred bright stars that the dust and scratches filter failed to eliminate.
d) Apply a Gaussian blur filter to smooth the dark areas. I modify the Hubl description by using a feathered mask to select only the darkest regions (i.e. not the blurred comet region) and only apply the Gaussian filter there. This ensures that the blurred comet region is not too disturbed and that the subtraction step is more accurate in the comet region.
e) Subtract the treated image containing only the blurred comet from the original. The modification I apply here over Hubl's description is to select an offset of 15 to the "apply image -> subtract" function in Photoshop. This ensure that the background is not black (RGB = 0,0,0) but roughly (15,15,15). This is an essential step to prepare the "stars only" image for combination with the "comet only" image and give the final background of around (20,20,20) - a very dark grey.
f) If residual star streaks are still evident after the background subtraction then use the "back subtraction" destreaking technique described here
g) Finally use the Photoshop "Add noise" filter with a value of 0.1 to 0.2 to treat the unnaturally smooth background in the dark areas and roughen it slightly. You now have image B
Image E shows a "average" combined stack aligned on the comet. It is shown here as a reference and guide for preparing image D.
Image D ("comet only") is a "sigma clip" or "sigma-kappa" combined stack aligned on the comet. The sigma rejection method ensures that:
a) Trailed stars are removed as much as possible
b) A sharply defined comet is isolated
The details of this process - including an optional image "sieving" procedure to treat cases where bright stars interfere with parts of the comet coma or tail is described in the tutorial here.
a) A key step not described in the comet isolation tutorial is a final prepare step to set the mean background in the "comet only" image. Use the "levels" tool in Photoshop to set the mean dark background to a RGB value of (5,5,5) to (10,10,10). Again this is a preparation step for the final combination of images B and D to create the final image C.
Combining the "comet only" and "stars only" image.
a) Its is essential that the star aligned stack (image B) and the comet aligned stack (image D) have used the same reference frame for alignment. If this is the case then the final combination can be done by just placing the images B and D on top of each other to get the comet into a true position matching the comets position in that reference frame. If this is not the case then you will have to use a selected original (non stacked) frame to use to align both image B and D into the correct orientation and position.
b) Combine the two images using pixel RGB addition (if your image processing app has pixel math functions - its image B + image D). If you use Photoshop align the two images B and D as layers and set the upper layer to "screen" blending mode (screen blending is pure RGB additive). Don't use "lighten" blending in Photoshop (as often recommended for comet processing) as this does not fully replicate the coloured transparency of a comet coma. Stars behind a green glowing gaseous coma should get a green tint (go and check a pre-stacked single frame - you should easily see the tinting of background stars). The preparation of the "stars only" background (image B) to have a mean dark background of (15,15,15) and the "comet only" image to have a mean dark background of (5,5,5) to (10,10,10) results in the final image C having a mean dark background of (20,20,20) to (25,25,25)
Have fun trying this and let me know in the comments section below how you experienced the procedures. Remember to shoot has many images as possible for best results. Its the total time interval and the distance the comet travelled in that time compared to the size of the comet that determines the best results.
|You have no new notifications.|
This page or operation is not available at the moment, because AstroBin is in READ ONLY mode. For more information, please check out our Twitter feed: https://twitter.com/AstroBin_com
If this image is the result of your processing of a public data pool, you can send it the pool so it's displayed there.
Use this form to select an existing public data pool.
If this image is the result of your processing of a private shared folder, you can send it the folder so it's displayed there.
Such limitation improves the website as a whole by discouraging people from creating fake accounts to like their own images. Thank you for understanding!
Currently, your AstroBin index is 0.00.