Jump to content

SkySurveyBanner.jpg.21855908fce40597655603b6c9af720d.jpg

What's am I doing wrong ?

Brian28

By Brian28
in Getting Started With Imaging

Recommended Posts

Brian28

Brian28

Posted
Posted

Last night was the first night I have been outside for months ..  right through  till 5.30 this morning..  ?

I was imaging with my new CEM60 mount and a QHY163c camera .  I was binning 2x2 as recommended , and thought the subs were good .. however I have just been stacking  them and the resulting Tiff file looks awful..  the M81 & 82 look reasonable but the stars inbetween look out of focus ..  but when you look at the subs they appear good .. 

can anyone shed some light on this please , I'm using DSS to stack the subs .. 

the picture below is a saved tiff from DSS , I have not added the Darks , bias or flat frames as I thought that might be the problem , so I restacked  without those subs but I still get this 

thanks Brian 

IMG_3833.JPG

Link to comment
Share on other sites
Thalestris24

Thalestris24

Posted
Posted

Hi Brian

Well, for starters, there's definitely a background image that shouldn't be there! I suspect you've somehow managed to includ some non astro test images in the stack? You may also have not edited the 'raw fits ddp settings' for the correct Bayer pattern. Maybe other problems too....

Louise

Link to comment
Share on other sites
Carl M

Carl M

Posted
Posted

Do you have hot and cold pixel detection ticked on the cosmetic tab of DSS stacking settings? I'm sure I've seen that feature do something like this before. If they are ticked I would untick them and restack.

Link to comment
Share on other sites
vlaiv

vlaiv

Posted
Posted

Since this is image of computer screen - it's hard to tell all the problems with the image.

Missing star centers is stacking problem - one of the settings in DSS can produce this - cosmetic correction or something similar - I can't remember now.

How did you bin your images? With color sensor, first step would be to use super pixel mode rather than bin. If resolution is still to high, you can bin again after.

Link to comment
Share on other sites
LightBucket

LightBucket

Posted
Posted
10 minutes ago, vlaiv said:

Since this is image of computer screen - it's hard to tell all the problems with the image.

Missing star centers is stacking problem - one of the settings in DSS can produce this - cosmetic correction or something similar - I can't remember now.

How did you bin your images? With color sensor, first step would be to use super pixel mode rather than bin. If resolution is still to high, you can bin again after.

Lol....yes I never noticed it was off PC screen either...so could be artefacts from that... :)

Link to comment
Share on other sites
Brian28

Brian28

Posted
  • Author
Posted
4 hours ago, Carl M said:

Do you have hot and cold pixel detection ticked on the cosmetic tab of DSS stacking settings? I'm sure I've seen that feature do something like this before. If they are ticked I would untick them and restack.

Thanks Carl I will have a look tomorrow , I'm outside again tonight , I'm hoping I'm not just taking rubbish ! 

Link to comment
Share on other sites
Brian28

Brian28

Posted
  • Author
Posted
4 hours ago, vlaiv said:

Since this is image of computer screen - it's hard to tell all the problems with the image.

Missing star centers is stacking problem - one of the settings in DSS can produce this - cosmetic correction or something similar - I can't remember now.

How did you bin your images? With color sensor, first step would be to use super pixel mode rather than bin. If resolution is still to high, you can bin again after.

Thanks Vlaiv , I'll check it out tomorrow..trying to make the most of the clear sky period .. I am not bining tonight so I will see how these look  tomorrow.. 

Link to comment
Share on other sites
Brian28

Brian28

Posted
  • Author
Posted
4 hours ago, Thalestris24 said:

Ahhh - I didn't realise that! D'uh!

Lol .. yes that's me taking pictures of the screen !  And the screen flicker    ?  If my subs were that bad I don't think I would bother anymore ...  

Link to comment
Share on other sites
Brian28

Brian28

Posted
  • Author
Posted
4 hours ago, vlaiv said:

Since this is image of computer screen - it's hard to tell all the problems with the image.

Missing star centers is stacking problem - one of the settings in DSS can produce this - cosmetic correction or something similar - I can't remember now.

How did you bin your images? With color sensor, first step would be to use super pixel mode rather than bin. If resolution is still to high, you can bin again after.

Just bining through the Ascom driver for the camera .. in APT 

Link to comment
Share on other sites
vlaiv

vlaiv

Posted
Posted
1 minute ago, Brian28 said:

Just bining through the Ascom driver for the camera .. in APT 

I would personally skip driver based binning. It will almost certainly do the same thing as you can do yourself in software (no hardware binning with CMOS sensor), but doing it yourself leaves quite a bit of flexibility - choosing whether or not to bin, choosing bin method, etc ...

Link to comment
Share on other sites
Brian28

Brian28

Posted
  • Author
Posted
2 hours ago, vlaiv said:

I would personally skip driver based binning. It will almost certainly do the same thing as you can do yourself in software (no hardware binning with CMOS sensor), but doing it yourself leaves quite a bit of flexibility - choosing whether or not to bin, choosing bin method, etc ...

Hi Vlaiv , I don't understand what you mean .. , could you elaborate a little please .. 

Link to comment
Share on other sites
vlaiv

vlaiv

Posted
Posted
2 minutes ago, Brian28 said:

Hi Vlaiv , I don't understand what you mean .. , could you elaborate a little please .. 

You are using QHY163c camera right?

CMOS sensors don't support hardware binning like CCD sensors do.

When you select bin mode in ASCOM driver for that camera what happens is that camera firmware reads every pixel as it normally would and adds adjacent pixels together - via mathematical addition, not some sort of physical process like in CCD (where electrons are "bunched up together" prior to read).

Since binning is done mathematically - no need to do it in camera, you can just download full sub and decide how to bin it your self in software.

Depending on method used you can get either same result as in camera binning, or you can even get better results - depending what type of binning you employ.

Just to give you idea of differences, here are couple of ways you can bin your subs:

1. Regular 2x2 pixel bin - you simply take 2x2 pixels and add their value to produce one pixel in their place - this is what is considered "normal" software binning. It works for Mono sensors, color sensors can't do it because of bayer matrix.

2. If you are using color sensor, you can do "super pixel". This creates color image out of single sub debayering in the process. It takes red pixel and produces red channel image, blue pixel for blue channel, and remaining two green pixels are averaged for green channel - this is again done on 2x2 matrix of pixels (repeated over image)

3. You can do weighted average on 2x2 - this works for Mono sensors. CMOS sensors suffer from telegraph type of noise - this means that some pixels have larger read noise values than other pixels, and often this read noise is not following Gaussian distribution, but rather two or three peaked distribution (like sum of gaussians each offset by some value). Such pixels are "isolated" or come in pairs. This means that out of 2x2 pixels there is a chance that one or max two pixels will have this type of noise. You can examine read noise distribution from darks and optimize average by assigning weight based on this read noise value for each pixel.

4. This is probably the best way to "bin" images. It works for both mono and color sensors, and it's not proper binning, but effects are the same or better than regular binning. Sub frame splitting. With mono and 2x2 bin you will in effect produce 4 sub frames for each light frame, thus quadrupling number of frames to stack - you do this by taking "every other pixel" and putting it corresponding sub frame (every other both vertically and horizontally, it also works for "higher" bin modes like 3x3, or 4x4, but it produces corresponding number of sub frames - like 9 or 16). With color sensor you do similar - again split into sub frames, but this time there will be one red sub frame, one blue sub frame and two green sub frames.

All above techniques apart from regular binning (for mono) or super pixel mode - require software support. First two are generally implemented (I believe).

You can also choose to do something different altogether - stack images at native resolution and bin resulting stack 2x2.

This is why I think is better to do it your self rather than letting camera firmware do it for you.

Link to comment
Share on other sites
carastro

carastro

Posted
Posted

Yes agree regarding the stars, turn off the cosmetic settings in DSS, this happened to me before.  Only other reason I found for holes in the stars was being out of focus.

I don;t understand why you are binning, isn't that counter productive? 

Should I presume the pink and green lines over the whole screen (that look like fabric lines) are not in the image but a reflection on the screen?

Carole 

Link to comment
Share on other sites
wimvb

wimvb

Posted
Posted
33 minutes ago, carastro said:

Should I presume the pink and green lines over the whole screen (that look like fabric lines) are not in the image but a reflection on the screen?

They are a so called Moiré pattern, caused by pixels on the screen interfering with pixels in the imaging camera that was used to photograph the screen. That pattern is not part of the problem. It's just the stars that are the problem, as I understand it.

Link to comment
Share on other sites

Archived

This topic is now archived and is closed to further replies.

Go to topic listing

  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...

Important Information

We have placed cookies on your device to help make this website better. You can adjust your cookie settings, otherwise we'll assume you're okay to continue. By using this site, you agree to our Terms of Use.