This image was captured and post-processed following the Tutorial you are about to read.
M51 - Whirlpool Galaxy
April 4, 2006 21:59 - 22:41 UT and May 2, 2006 21:32 - 22:43 UT.
IMPORTANT: for LRGB click .... HERE !!!!
Deep Sky Objects (DSO's) usually are faint objects and not so easy to image and post-process.
This Guideline hopefully will help you in getting started.
- 1. Preparation and Prequisites
- 2. Getting the DSO on your screen
- 3. The actual imaging
- 4. Post processing [K3CCDTools]
- 5. Post processing [Photoshop]
- 6. Improved layering method
- 7. The result
1. Preparation and Prequisites
- A well tracking Equatorial mount with a low Periodic Error.
Mounts with Periodic Error Control are preferred, but not mandatory.
Ideally your mount should have autoguiding, which surely is not impossible: it took me quite some efforts but I succesfully implemented autoguiding for my TAL-2M mount which has only the RA motorized.
Note: in November 2005 I sold my TAL-2M which has been replaced by an LX200 10" f/10 telescope.
In the meantime I have written the LX200 Autoguiding Tutorial
- The mount must be perfectly Polar Aligned. Here is how I polar aligned my TAL-2M and here is the tutorial for the LX200
- If you have a GOTO mount you must know how to operate it; if you have a 'push-to' mount you must know how to use your Setting Circles.
- You need detailed star maps, like Cartes du Ciel or Megastar (see my Software page)
- You need a Long Exposure camera, which can be a modified webcam or a commercially available camera, such as Atik and others.
- You need imaging software that will work with your Long Exposure camera. I use the latest version of K3CCDTools.
- You need to know how to work with the Post Processing software.
- You need clear skies!
2. Getting the DSO on your screen
My TAL-2M did not have a GOTO, so I describe here how I (tried to) locate and aim at a DSO.
3. Now it is time to do the actual imaging
- I started with using a star map to find out the RA and DEC coordinates of the DSO I wanted to image
- I mounted my camera and use the 'factory settings' on my RAW Vesta SC3a black and white (b/w) camera and start the Long Exposure feature in K3CCDTools.
- Activated the K3CCDTools Long Exposure Panel.
Note: I made sure that the frame rate was set at 5 frames per second:
- Make sure you have activated the live Preview button on the K3CCDTools Long Exposure Panel.
- I focus on a rather bright star; with my Newtonian TAL-2M telescope I prefer my Diffraction Spikes method.
- Next I follow the procedure described on my Setting Circles page to aim at my object, with the camera mounted.
- When I am lucky the object is on the K3CCDTools Preview screen, but in most cases I just see stars and NOT the object I was after.
Now I enter a tricky phase: I have to reposition my scope, but in which direction?
I use my sky map to compare the star pattern on the Preview screen with the stars on the map and gently adjust the aim of the scope.
When still unsuccessful I dismount the camera and insert my eyepiece with Parfocal adapter and try to find the DSO visually. Sometimes this is very hard too, for example: from my light polluted environment I have great difficulties to see M51.
When the DSO is in the centre of my eyepiece I very carefully replace the eyepiece with my camera: the slightest push against my scope will disturb my precious aim. Sometimes I have to reposition/remount several times before the DSO is on my screen.
This whole procedure might take almost an hour :-(
4. Post processing [K3CCDTools]
- It is recommended to activate the Big Status Display and the K3CCDTools Brightness meter.
- Use the K3CCDTools settings dialogue to give your AVI a meaningful name, so that you can properly archive it later.
- Play with the camera settings (gamma, gain [using 90% instead of 100% dramatically reduces the AMP glow!], exposure time) till you are happy with what you see on the Preview screen.
Move the cursor into the image and read the background level values on the status bar: values of 100-120 are recommended to obtain a sufficiently dynamic DSO image.
- Finally: hit the Record button and your AVI will be filled with frames.
- While the actual imaging is going on record your camera settings: I used to use QCamLog, but this has been replaced by WcCtrl (see my Software page).
Here you can see what I mean:
This is a K3CCDTools in night vision full screen print made during the actual imaging.
Note the Brightness Meter and the background level values on the Status Bar.
My Philips Vesta Pro SC3a b/w camera settings: page 1.
My Philips Vesta Pro SC3a b/w camera settings: page 2.
- When you have enough frames (take a LOT: better safe than sorry!) cover the sky end of your scope with a non-transparent black cap while the imaging continues: now you are capturing the Dark Frames (capture at least 10, best is even more).
Dark frames only contain the noise and hot pixels from your camera itself; during the post processing they will be subtracted later.
There is a fine K3CCDTools tutorial Deep Sky processing tutorial written by Carsten Arnholm.
Here follows my current practice:
Creating the Dark Frame
5. Post processing [Photoshop]
- 1. I load the AVI into K3CCDTools.
(note for advanced users: Jon Grove's iPrep is a nice tool to generate a synthetic flat frame; it also handles dark frames and can save the output as an AVI which then can be used as input for K3CCDTools.
- 2. Tab: Sequence processing
- 3. I do a 'Full Rectangles', so that now all frames are checked.
- 4. I uncheck all frames.
- 5. I browse down to the dark frames and hit the Space Bar to checkmark each dark frame I want to include in the final dark frame.
- 6. When done: 'Calculate Result Image (Sum)'
- 7. Save the dark frame: File|Save as a bmp.
As filename I use the name of the AVI, with the prefix df_ for Dark Frame.
Selecting, Aligning and Stacking
- 8. I reload the AVI into K3CCDTools, so that I have a fresh start.
- 9. I activate the dark frame that I created in the previous steps: DF button etc.
A simple subtraction of the hot pixels will leave a 'hole' in the image, which is not nice.
From version2 onwards K3CCDTools has a nice solution: the Average Surround feature, which can be invoked via the Settings Menu, tab Seq. Processing:
- 10. I browse through the frames and disable (with the minus key) those frames I want to exclude from further processing, because a) they do not have nice round stars or b) have a passing satellite on it or c) simply are the deliberately captured 'dark frames'.
For better judgement: use the 2x button: now you can see on which frames the stars are nicely round.
During this selection phase you will be pleased that you took so many frames, so that you can afford to weed very selectively.
Note: you might want to skip step 11 through 16 and leapfrog straight to what is written in the Note after step 16.
- 11. I do an automatic 'Select Center Point (FFT)', which will automatically go through the whole AVI, set the center points and checkmark the frames.
- 12. Next: right-click in the column with the frames and select a 'best' frame as Reference Frame followed by a 'Calculate Quality' of all checked frames.
- 13. Sort on Quality by clicking the top of the Quality column.
- 14. Next: right-click in the column with the frames and do a 'Calculate Difference' of all checked frames.
Make sure that you use as Threshold a value higher than the background: move the cursor into the image and read the values on the status bar.
- 15. Right click in the frames column and 'Align frames'; I use 'Difference Alignment', 'Align checked'.
- 16. When done: 'Calculate Result Image (Sum)'
Note: do not tell Peter Katreniak (author of K3CCDTools) this, but it is also very nice to use the Planetary Wizard for DSO aligning and stacking:
immediately after step 10 invoke the Planetary Wizard, follow the procedure and come back here :o)
- 17. You might want to play around with the histogram in K3CCDTools if you want to: especially recommended to bring out faint nebulae!
Note: usually I proceed with the next step, but sometimes I save as a PNG16 file and load this file in Registax and use the wavelets to get the fine details, after which I save as TIFF (also see the Planetary Post Processing Tutorial
- 18. Save the result: File|Save as 16 bit TIFF file with the appropriate name (I usually add something like _k3stack60 to indicate what post processing method and channel I used and what the stack size was).
Here you can see what I mean:
Here is the K3CCDTools result image after stacking and aligning.
Due to tracking problems with my mount I had only 5 usable frames to stack.
Of course it is also possible to use e.g. Paint Shop Pro, or The Gimp etc.
Links to Photoshop Tutorials are on my Software page.
- Load the in K3CCDTools created TIFF(s) into Photoshop.
- Rotate/flip the image(s) so that you have North UP.
- Use 'curves' in the style as nicely described by QCUIAG member TG Tan
Start with the section 'Histogram stretching'.
- TG Tan also describes how he uses layers and all kind of functions in Photoshop to achieve stunning images and I suggest you study and try his methods, including the 'marriage' of the Y and RGB channels, so that the colour image is there again.
With the rather high background level of the above imaging procedure I found out that it is not necessary to do the TG's "Positive Curve Stretch" but you can start immediately with the "Negative Curve Stretch".
- During the processing sometimes dark halos (ears) appear around stars. The cure for this is Loreal, see the link on my Software page.
Meanwhile I have RAWed both my SC3 cameras so that the ears have become non-existent.
- To suppress noise I use NeatImage: see the link on my Software page.
- Vignetting and Sky Gradient Correction - e.g. caused by light pollution - I cure with the Dynamic Background Extraction tool of PixInsight LE: see the link on my Software page
July 31, 2006 21:19 - 21:32 UT.
Dynamic Background Extraction in PixInsight LE
Vignetting and light pollution are the enemies of astro-imagers.
When imaging from urban or suburban locations we have to work knowing that the sky background will never be truly neutral nor uniformly illuminated and coloured in our raw images, especially at low and intermediate heights above the horizon, and varying wildly as a function of local conditions of transparency.
This is what we know as sky gradients.
The Bubble Nebula before and after Dynamic Background Extraction in PixInsight LE.|
Sometimes the DBE might be too agressive to your liking and e.g. nebulosity you wanted to preserve has been removed.
Then a reduced subtraction in PixelMath helped me: I set the Operand RGB parameters to 0.5 (you should experiment per image to get the right values!). as can be seen
in this screen print
6. Improved layering method
QCUIAG member Simon Smith has written Astrowave: with this software you can calculate the rotation angle difference between two images.
Originally intended by Simon for 'wide field deep sky mosaics' I found another usage: combining frames from different sessions and/or camera orientations.
My problem always was to rotate a the layer just so much that it matched perfectly with the other layers underneath: a lot of fiddling was required to get it right.
With Astrowave this is much easier: Astrowave tells me the rotation angle.
In Photoshop rotate the frame with the rotation angle found by Astrowave: Image|Rotate Canvas|Arbitrary.
No only horizontal and vertical moves are required to position the layers which can be best done with the arrow keys (hint: use Blend Mode Opacity).
Simon THANK YOU for sharing this tool !
7. Finally: the result
April 4, 2006 21:59 - 22:41 UT and May 2, 2006 21:32 - 22:43 UT.|
Combination of the colour and black-and-white images of the above days.
I used Astrowave to determine the scale and angle between the two images and layered them in Photoshop whereby the b/w image has been used as luminosity layer.
On the upper left hand side Galaxy IC4278 is just visible.
May 2, 2006 21:32 - 22:43 UT.|
Imaged with LX200 10-inch f/10 telescope with 0.33 Focal Reducer operating at f/4.06 [focal length 1015mm],
Vesta Pro-SC3a B/W RAW camera, Baader Infra Red Blocking Filter.
104 autoguided exposures of 30 seconds.
To get more from your images: CLICK HERE
IMPORTANT: for LRGB Imaging & Post Processing .. CLICK HERE !!!!