help with m31

[gonzostar]

I am new ro stacking and useIng DSS. Ihave taken lights (13), darks (22), flats bias(40). When i fed all data into DSS i got 6 * 2 min lights stacked out of 13 light frames.

Seemed to have stacked ok with the  adjustments that dss provides.

Then  shock horror m31 around the core was is a "ring going all around the core, with a dark gap in between.

Any suggestions on settings, techniques would  be gratefully accepted.

Iam using a 8"SCT (f6.3) and a canon 450d camera.  

[Freddie]

Posting up the pic may help us to better understand the issue.

[gonzostar]

Hope this helps. Any tips will be accepted. I have a feeling more light frames? Also how do you select  light frame Scores on DSS, for the ones to choose for stacking?

Dean

[Stub Mandrel]

I'm a beginner. I check my lights by eye, get rid of any with trailing or other faults and tell DSS to stack 100%.

If its rejecting frames and the stack percentage is less than 100%, click the advanced button and set the star detection threshold to a smaller number.

Your image seems very dark for 2 minute exposures of M31, what ISO were you using?

[gonzostar]

I have restacked the process.  Getting rid of bad flats, I get a threshold of 71  stars, I took these pics with iso 800. It was last saturday when moon was 95% full. However i did use a astronomik cks filter. Maybe this makes a big difference? 

Dean

[dph1nm]

Looks like that  image has got posteriorization to me i.e. at some stage it has been saved/converted to an 8-bit image.

NigelM

[spaceman_spiff]

Hi,

Don't worry about the very dark image - this is a natural consequence of stacking many frames containing mostly black.

In the processing stage of DSS you need to go to the luminance tab and alter the line (looks like a curve cutting through the histogram). Alter it using the dials on the left (changing the Darkness, mid tone and highlights) so that the steep upwards inflection of the curve occurs in the middle of the peak of the histogram. This should maximise the contrast of the image.  Once you have done that, you can fiddle with it to get a nice dark sky or smooth some of the noise.

Hope that makes sense.   

[gonzostar]

Thanks for the info. The issue seems to arise when open upon up my image from DSS to gimp. Can i convert the 16bit image from DSS to 8bit so gimp can use it properly? I am on vista and is not allowing me to use photoshop! 

The image i get on DSS is not to bad(certainly no way as impressive as alot of images on the site) Is there an alternative good processing site like gimp?

Thanks

Dean

[spaceman_spiff]

I'm not too sure about gimp (I used to use it for work years ago). I find that DSS can be used to get pretty good images without reducing the bit depth or transferring the image to other programs. I have added a before and after print screen of DSS to show what improvements can be made. Try to get the curve on the bottom left to rise steeply where the peak in the histogram is.   

[dph1nm]

Thanks for the info. The issue seems to arise when open upon up my image from DSS to gimp. Can i convert the 16bit image from DSS to 8bit so gimp can use it properly?

Yes - but you need to adjust it in DSS first (as the post above) and save it with the adjustments applied (which is one of the settings in DSS - can't remember where off-hand though).  This is because GIMP works in 8-bit and DSS is in 16-bit. To import it, GIMP just scales the image down by 256 (2^8) - if you haven't saved those DSS adjustments this results in a very dark image with quantised layers (as you have found!).

NIgelM

[gonzostar]

Wow thats absolutely amazing what you have done with the data. Now i shall give it ago

cheers

Dean

[gonzostar]

[gonzostar]

Slowly bur surly think nearly getting there 

Any more hints gratefully received, 

Thanks 

Dean

[spaceman_spiff]

Looking better - fiddling with the settings is always a good way to see the faint details hidden in the picture. You can see much more of the core of M31. To see details further out, you may really need to push the contrast (moving the curve to the left). If the picture gets too noisy but no more detail can be seen then consider taking longer exposures (5 minutes) or more light frames using higher ISO. Only problem is you will need new darks!

I think the best thing to do is just play around with it!   

[wimvb]

If you want to process the image in an 8 bit application, you need to stretch it as much as possible in DSS, before importing it in the other app. Use the steepest part of the stretching curve.

This will enhance noise, vignetting, dust bunnies, etc. But you'll have to take care of that later on.

Remember, between any two levels in an 8-bit image, a 16 bit image has 256 extra levels. Going from 16 to 8 bit, you loose these.

Before I settled on Pixinsight, I tried IRIS (google: iris astronomy). It can handle 16 bit TIF, but I found the learning curve too steep (pun intended)

[gonzostar]

Yes will try moving curve slightly more to left. Also in future will do more light frames. The reason why iso 800 is that was recommended in every phon count book, but like most of this it comes with experience. Will also try iso1600. As for 5 minute tracking the image on dslr would get to "foggy". Also i am trying to  get used to the synguider    

Will have to wait till after christmas to get  pixinsight or it will have to come out of her ladyships present! and she wont be to impressed   

Dean

[gonzostar]

Yes you are of course correct to far left noisy. So need more lights and trying longer exposure in future.

Deanm31blankleft1.tif