MIRA 78
Spring 2007


THE ASTRO IMAGING ISSUE


This is a special imaging issue written by Clive Rogers to help the members get to grips with obtaining digital webcam images of the night sky on their computers.   I hope this is a help and inspiration to all the members who would like to have a go.






Capturing the Planets with a Webcam

By Clive Rogers


I use a Philips ToUcam 740K because it contains a CCD chip which is more sensitive than webcams with a CMOS chip although I have been told these CMOS webcams do perform very well.
The following are the programs that I use on a laptop running XP Home.  All are freeware except K3CCDTools which you pay for after your trial period ends.

http://www.pk3.org/Astro/ 

K3CCDTools from Peter Katreniak for capturing images/AVI's.

http://registax.astronomy.net/ 

Registax from the Aberrator for aligning the images/AVI's

http://www.gimp.org/ 

The Gimp for the final tweaking of the images.

http://www.burri-web.org/bm98/index.html 

WcCRTL By Martin Burri for controlling a webcam without continuously opening Video Capture\Video Source etc.

Hints On Stopping The Frame Dropping.
21st January, 2006


For the first time I had frames dropping while using K3CCDTools.  This was quiet serious as it was dropping a frame every second, sometimes 2 frames a second were dropped.  It took me a while to find the solution but here is what I did to stop all the frames being
dropped.
Checked in the BIOS and turned of the S.M.A.R.T technology.
Open the Device Manager and make sure Enable Write Caching on the Disk is ticked.
Shut down the Anti Virus software (I was not on the Internet so this did not matter too much)
Shut down the Firewall again this was not a problem as I was not on the Internet.
Opened the Device Manager and made sure the DMA was checked to make
sure the saving to disk was running at top speed.  Then Reboot the Laptop.
I have also found that Defraging the partition helps here too !
One other thing I now do is to move all the folders and programs etc. and the partition that I am saving my images/AVI's too and format that partition.  Also I have noted that saving to a FAT32 partition is a lot faster than an NTFS partition I now use a FAT32 partition to save to.
After doing all this I was able to capture like I normally do without the frame dropping.  I still don't understand why I had to go through all this as frame dropping had never been an issue before. Computers are fickle things at the best of times.
The sequence of events that I use for a nights capturing the planets are as follows.

Set up my 222mm f/7.2 telescope on the EQ5 mount at least an hour or two before usage to allow the mirrors to cool down to ambient temperature.

Polar align using the polar scope.  This gets quite accurate with practice.

Plug in the handset and turn it on making sure that I have it set to North for northern hemisphere.

I don't use the C Cell batteries as I find they will only run for a couple of nights then die on you when you least want them too and it turns out to be very expensive.  I now use a 240VAC to 6VDC converter plugged into the EQ5 handset. Be sure to check its polarity and that it is set to 6 VDC before plugging it into the handset and switching it on.  I have made the fatal mistake of catching
the voltage switch unknown to me and put 12 VDC through the handset.  Believe me the handset doesn't like it at all. New motors and handset had to be ordered.  Now I have put electrical tape over the slider switch to stop accidentally moving it.
Next step is to check the collimation of the primary and tweak it as necessary.  I have found that when collimation looks good for visual use is not good enough for planetary imaging because of the high magnification used.  Typically in the range of f/28 to f/40.  I have used my Laser collimator which gets it very close to collimation, but if you move the focusing knob you will find that the laser beam will also move and not sit on the primaries centre any more.  Here I note which one, either racking in or racking out gets me closest to the centre point of the primary so that when out under the stars I will go past focus and rack to focus remembering from earlier which one of the two got me closest to central primary position.
Time to get the laptop up and running.  Start a star atlas program and turn on night vision mode.  Then I lower the laptops screen brightness.  Plug in the webcams USB plug and the parallel port plug even if I am not going to use it that session.  It's amazing how many times I have gone to use the long exposure modification only to find I can't do long exposures.  Then the penny drops.......  The parallel port is not plugged in.


Fig 1  Starting the capture software, in this screen shot its K3CCDTools and WcCTRL

Now I start the capture software (Fig 1) In my case its K3CCDTools and WcCTRL.  Tell WcCTRL to connect a cam and choose which camera your going to use.  Once started I tell K3CCDTools it to use the WDM driver in the Video Capture drop down menu and make sure the Preview is also ticked.  It also pays to turn up the Gain to help in finding that star to help in focusing.
Fit the X4 Barlow lens into the eyepiece holder and then insert the webcam into the Barlow lens.  This gives me an effective focal ratio of f/28.8, I some times pull the webcam out a little bit more there by increasing the focal ratio even more.  By how much I am not sure but lets say around the f/36 mark.


Fig 2  Find a bright star, here it's Betelgeuse

Find a very bright star, here I have chosen Betelgeuse (Fig 2) defocus it so that you can see the shadow of the secondary mirror on screen. Notice where it is and how close it is to the centre of the primary. The closer to central point it is the sharper and more detail your images will have. Collimate as best you can. Now realign that star and focus until I have the 4 needle sharp
spikes associated with the secondary mirrors spider vanes. If the spikes are single and needle sharp you know you are in focus. On the other hand if all you see is twin spikes associated with each of the spider vanes then you are not in focus.
When I have the focus just "so" I now check that the finder scope is set up to look at the same object. When the telescope and the finder scope are not matched perfectly its almost impossible to put the planet on the small CCD chip.
Now to find that planet.
Which ever planet I plan on capturing I now place on the webcams CCD chip.  Believe me this is easier said than done as the webcams chip is very small but with practice this does become easier.


Fig 3  Settings are made in the right hand box

When the planet is being tracked by the telescope I turn my attention to the capturing software again.  Using WcCTRL in the Primary Tab (Fig 3) I set Gamma 0, Gain 50%, Shutter 1/25 and FPS (frames per second) to 5 and I click the Freeze Colour Balance. If Freeze Colour Balance was ticked to start with I tick Auto Colour for a few seconds until the software catches up then click Freeze Colour Balance again.  This stops the software from constantly changing the colours.


Fig 4  Settings are made in the right hand box

On the Secondary Tab (Fig 4) of WcCTRL make sure Colour Mode is ticked.  Saturation is 100% and Cont is 0%.  Next in K3CCDTools either click the tools icon at the top or click Options then come down to settings.  Frame Rate is set to 5.00 fps.  Enable Capture Time Limit: here I tell the software how long to capture for.  If its Jupiter I capture for around 90 seconds and no more because the Planet rotates very quickly (10 hours) and it will blur the final image.  Saturn and Mars you can go longer because on Saturn there
is not much detail that will blur and Mars rotates in just over 24 hours.
Directory is where I store the captured AVI's.  Finally I give the AVI a name and make sure the Incremental is ticked so that the name will have consecutive numbers after the name.  Then click the Camera tab and make sure the SC Long Exposure Modified Camera is not ticked.  Now check to see what the image looks like on screen and adjust the settings until the image is just nice but slightly on the dark side.  The reason for this is when you stack the images it will gain a little in brightness as well as detail.  If the image is bright to start with as it is being captured then when stacked the image will be far too bright and all the detail will be washed out and you won't be able to do much with the image.
When happy with what is see on screen I tell the K3CCDTools to capture the AVI.  After each AVI I alter the Gain settings a little on either side of the original settings and capture more AVI's.  The reason for this is bracketing either side of your original settings and then you stand more of a chance of hitting the right settings.  At the end of the nights captures you could end up with many Gigs of AVI's.  What I then do before I do anything else is Zip them up with WinZip.  Astronomy AVI's Zip up quite well.  I have seen 15 Gigs Zip up into under 700 Megs.  I then store these on a USB2 external hard drive.  Also I write them to CD's and DVD's just in case the hard drive crashes which has happened to me on a couple of occasions.  When the weather plays up and stargazing is not happening because of rain or cloud (more likely then not here in the UK) then I turn to processing the captured Avis. For this I use to Registax a free program written by Cor Berrevoets at

http://registax.astronomy.net/

Here is a very quick outline of what I do to process all those Gigs I have captured.
I click Select.  Browse to the directory were I store the AVI's and choose the first one.  If you can't see the AVI's then choose AVI File Type from the drop down menu.  When I have loaded the first AVI, I normally go through it by hand to throw away the bad frames. This is done by clicking the Framelist on the bottom right.  Use the up and down arrows and Space Bar to tick or untick the individual frames.  When I have done this I then choose a box size from the list at the top that completely covers the planet then click in the centre of the planet.  Here there are 2 choices.  Either work Manually or Automatic.  I choose to work manually as it gives you that little bit more control over what is being done.
To start with you could go through the tutorial on my website at

http://homepage.ntlworld.com/c.rogers/

on Registax to help you get started.  There is no right way or wrong way to capture and process images.  Each amateur astrophotographer has their own technique.  Although I am always willing to learn though!!!
Choose the Method of quality.  For the planets I use Gradient and the Lowest quality I set at 70%.  Next click Align then fetch a cup of tea.  Next click Limit.  You can click several options here including resizing using either of the Resampling or Drizzling and the Factor.  Otherwise click Optimise then sit back and drink the tea you just fetched. When the program has optimised all the frames its time to stack them.  Click the Stack tab again here you can make use of any and all the options but here I normally click stack. After the images have been stacked its time to adjust the image with the Wavelets.  Move the sliders to the right and watch the image.  With careful use you will be able to bring out the faint detail in the image.  Move the sliders a little too much and you can add processing artefacts which is not what is wanted.  Down the right hand side of the screen there are more tabs that can be used to enhance your image.  The art of adjusting the image is to make the final image look as though it has not been processed at all
The final stage is to save the image out.  Once this has been done you can now load the image into your favourite graphics image package.  Here I use The Gimp on my desktop PC because I have found that the LCD screen on my laptop makes all the images very bright.  One other thing I have noticed is that CRT screens tend to be better for graphics work as they tend to show more detail.  This last point is a personal choice.


Fig 5  This image of Saturn was done using this method described but I used RGB filters.





Planet stacking with Registax

By Clive Rogers



The first step (Fig 1) is to load in your AVI or stack of bmp's or jpg images by clicking Select.  Here I am using an AVI of Mars.


Now choose a size of alignment box (Fig 2) that fits around your planet. Here I am using 128.
Make sure that the Colour box is checked under Use.  For planets I have found it best to use the Gradient and I chose to keep the top 30% (Lowest Quality 70).  You can use either Processing on Automatic or Manual.   Here I have chosen to use Manual.  Then click Align.


After the initial alignment click the Optimise tab along the top (Fig 3).  For this Demo I left most of this untouched.  Search area 2 and Optimise until 1.  Next I set the Create to around 50 and hit Create.  What this does is create a reference frame that the program will use to optimise and stack the other frames too.


Here you are on the Wavelet tab for the first time (Fig 4).  You can make alteration here to your image so that it looks good.  When you have finished here click the Continue button.


Most of my planetary images have the red and blue channels out of alignment (Fig 5). Here click the RGB shift tab running down the right hand side and when the RGB shift box opens click Estimate.  Sometimes the program gets it wrong so you have to click Estimate a second time.   When finished close the RGB box and now move the Wavelet sliders to enhance the images details.   Be careful not to introduce false artefacts here. Remember to click the Continue button at the top.


This next screen (Fig 6) tells you to continue to process the rest of the frames with either Optimise or Optimise & Stack.  Now you can Resample or Drizzle the resultant image from X1 to X4 in .1 steps.  Now press Optimise.


Here the program is sorting through the rest of the frames (Fig 7) in the AVI to find the best frames for stacking.  Wait till the program has finished and click the Stack tab at the top.


There are a number of setting on this tab (Fig 8) but I mainly leave them alone as I have not found them to be helpful as yet so I just click Stack.  When done click the Wavelet tab at the top to move onto the next step.


Now I am back at the Wavelets again (Fig 9). Again I hit the RGBshift tab along the right hand side because now that I have more frames stacked it shows more of the red and blue that has been shifted.


Next step is to sort out the Histogram (Fig 10).  Sometimes the camera can get the colours very strong so I tweak the Histogram to try and make the image look more natural.


In this image you can see that the Wavelet sliders (Fig 11) have been moved a greater amount to the right.  This is because I have changed from Default to Gaussian filter. This allows more control over the Wavelets.  If you are happy with your image you can now save your image to hard drive by clicking the blue Save Image button at top right.


Click the Final tab along the top of the program (Fig 12).  On this page you can rotate the image to correct for telescope positioning as you can see here I rotated Mars.  After any of the adjustments from here on you can save your image to hard drive by clicking the blue Save Image button at the bottom left.


Click the Advanced HSL (Hue, Saturation and Lightness)(Fig 13).   I found that I had to put the mouse pointer in the white area and press the left mouse button and while holding down the button move the mouse to get the lines to appear.  To move the lines just click on and hold one of the red circles.  Adjust to taste.


Should you wish to you can now save your image by clicking the blue Save Image button (Fig 14)


This last image shows that you can now resize your image by clicking Resize (Fig 15).  You can resize from 25% up to 200% using one of the 4 filters.  Again clicking the blue Save Image button allows you to save your image.  From here you can import the image into your favourite graphics package to tweak the final image.

You can download Registax as a free download from

http://registax.astronomy.net





Removing Light Pollution Using the Gimp

By Clive Rogers


Because I live on the edge of a large city most of my deep-sky images suffer from light pollution.  So I have had to learn a trick or two to try and bring back the darker skies to my images.  Looking round the Internet I found that most of the astro imagers use Adobe Photoshop.  The price of this piece of software is very prohibitive so I looked around for an alternative and found one already in my distribution of Linux.  The software is call The Gimp (Graphic Image Manipulation Program)
Having found a graphics package now it was time to translate what I had read from Photoshop to Gimp.  Here is what I found.


Going from this very light polluted image Fig 1.  To this darker skies image Fig 2.


First load your image as in Fig 3, use either File/Open menu or CTRL+O.  Then open the File/Window/Layers dialogue box (CTRL + L).   You should now have 3 box's open on screen.  Organise these to suite yourself.


Under the Layer menu choose Duplicate Layer or in the Layers dialogue box right mouse click and choose Duplicate image Fig 4.   This will give you 2 images in the Layers dialogue box.  Make sure the Background Copy is the highlighted image.


Now click Filters in the top menu and come down to Blur/Gaussian Blur Fig 5.


When the dialogue box opens change it from 5 to around 50-70 and click OK Fig 6.  This will blur your image to such an extent that you wont be able to recognise it.  This is similar to having a flat field image which we want to subtract from the original image.  What you are doing is blurring out all detail so that you are left with nothing but background information.


Now move to the Layer box and change the Normal to Subtract Fig 7 by using the drop down menu.  Here you can subtract as much or as little of the blurred background as you wish.  Sometimes heavy subtraction can take out too much of the original image so use with caution.


Now move the slider to the left until you are happy with the result Fig 8 which will be shown in the large image box.


Click Image/Merge Visible Layers or CTRL+M to merge or Flatten Image the 2 layers of your image fig 9.  This will now combine the two images in the Layers dialogue box.


Click OK at the next dialogue box Fig 10 as this is asking about clipping or expanding your image.  So because we did not use two different images we don't need to do anything.


Now your image should be a little darker and perhaps a little sharper too Fig 11.  If not you can always go back and readjust the blurring by pressing and holding CTRL and tapping the Z key or using the menu system of Edit/Undo until you again have the 2 images in the Layers dialogue box.  Before you start to alter your images again, being back here at the two images make sure your Background Copy is the image that is highlighted in the Layers dialogue box.  There are a few times I have forgotten and worked on the Background image instead of the copy and got unexpected results.


When you are happy with your image and you want to save it out then either click File/Save As or Shift+CTRL+S Fig 12.  One thing here is to type the name you wish to save the image under before choosing the file type.  Sometimes I have found that the image would not save until I typed the name first then choose the file type.
All the above can also be done with right mouse clicking on your image and the Layer dialogue box and choosing the relevant headings but only when you have an image on screen.
One more thing about Gimp is that it is only 8 bit format.  This means its good for BMP and JPG format images etc.  Now I use CinePaint which not only understands 16 bit format like fits, tiffs and png images but 32 bit fits as well.  The good thing about CinePaint is that it is so similar to Gimp it was easy to learn.  Both of these programs can be used either under Windows or Linux but the Windows version is a little behind the Linux at present.

Gimp can be found at

www.gimp.org

and CinePaint can be found at

http://www.cinepaint.org/