Jump to content

SkySurveyBanner.jpg.21855908fce40597655603b6c9af720d.jpg

Calibration in DSS..... Is it doing it wrong?....RESPONSE FROM LUC OF DSS


RobH

Recommended Posts

  • Replies 95
  • Created
  • Last Reply

As one who does not do a great job of calibrating his images but is getting better I thought

I thought it went like this

Subtract a median bias from everthing

subtract the relevent median darks from the lights and median flat field images

then divide the corrected lights by the corected flats

then stack the corected lights

(wave a magic wand in photoshop and you are done)

when this is all sorted out please let me know cos I'm doing some darks, flat fields and flat darks at the moment and would love to know how to apply them in the correct way

Steve

Link to comment
Share on other sites

It is all sorted out Steve.

It's done the way myself, NBP, and Roundycat, plus almost everyone else who'se been doing it a while does it. Not the way DSS does.

Apply Darks or Bias. Not both.

Do not apply bias to the darks as the darks already have bias information in them which will be subtracted when the darks are applied to the lights. If you bias subtract the darks, it means you've double subtracted bias from your lights.

Apply bias to the flats.

Calibrate your lights with darks (or bias) and bias subtracted flats

Job Done.

Cheers

Rob

Link to comment
Share on other sites

Ok, so the way I see it now to use DSS correctly would be to use a light frame and add all the flats and bias frames and run the process, you would then end up with a Master Bias and a Master Flat, the Master Flat will be bias subtracted. Do the same with the dark frames(this time with no bias) to create a Master Dark.

Then when you come to process your images proper just use a Master Dark and your Master Flat - job done!

Link to comment
Share on other sites

This has been bugging me all night, and his statement about why would you not subtract bias from all frames, and it sort of makes sense if you do it in the right order...

Subtract Bias from the Flat - thats ok because the flat is not further subtacted.

Subtract Bias from the Dark so you have (Dark(with no Bias)).

Subtract (Dark(with no Bias)) from Light.

Subtract Bias from Light - thus avoiding the double subtraction.

Link to comment
Share on other sites

Yes Martin, if you do that it'll work, but you end up adding noise....see Narrowbandpauls post earlier in the thread about exactly that.

He seriously know's the theory of all of this, having studied it at uni.

I must say though, that the amount of noise it will add will be miniscule, it's just not the best way.

Cheers

Rob

Link to comment
Share on other sites

Hi, have found this very interesting to read through, and I will spend more time to really understand all the technical aspects going forward. However, as someone who is just starting out and wants to use something freely available and quick/simple to use is the conclusion of this that DSS does not work in the perfect way, but putting light, dark, flat and bias into DSS and letting it do its thing is better than not using bias at all? Thanks.

Link to comment
Share on other sites

Yes and no......

All frames will contain bias information and we want to get rid of it by subtraction, so -

The dark frame contains hot/cold pixel info + bias info, so by subtracting the dark frame we are getting rid of the hot/cold pixel info AND the bias info in one hit - so no need to further subtract bias.

So the first time you use DSS to create masters DON'T include bias frames...........

I think!!! I'm sure someone will put me straight.

Link to comment
Share on other sites

You need to apply bias to your flats though as they won't apply properly if you don't.

In Maxim, for example, if you don't bias subtract your flats, they will under or overcompensate.

Freddie, I'm glad you've found this discussion useful...there has been confusion over this issue for a long time when applied to DSS, so it's good that we've got to the bottom of it, which is why I started the thread in the first place.

Rob

Link to comment
Share on other sites

So is it better to bias subtract the flats or use flat darks?

It's really best to use flat darks, but they are a pain to do, and as flat darks are normally of very short duration, in practice, bias will do the same job, and you only shoot a master bias once every 6 months or so (to take account in the change in your sensor over time).

I never bother with flat darks myself.

Cheers

Rob

Link to comment
Share on other sites

Calibrate your lights with darks (or bias) and bias subtracted flats

Job Done.

I never take darks with my SX CCD cameras, just BIAS and FLATS so the above is exactly what I do. Mind you, I don't get much imaging done at all thee days so for me it is getting academic :)

Link to comment
Share on other sites

Talking to Ian King a few days ago, he mentioned that it's recommended that you don't take darks with the H9, as the camera is so low noise that darks will increase your noise, not reduce it.

I would think that this applies generally to 285 chipped cameras.

Rob.

Link to comment
Share on other sites

So is it better to bias subtract the flats or use flat darks?

Use bias subtracted flats, darks as they are and no further bias frames.

So to create master sets you need a 2 stage process..

1. Load a light frame and all of your Flats and Bias frames, process as normal. This will produce a Bias subtracted Master Flat.

2. Clear the files and load a light frame and your Darks, process as normal. This will produce a Master Dark.

Thereafter just use the Master Flat and Master Dark when processing.

Link to comment
Share on other sites

Use bias subtracted flats, darks as they are and no further bias frames.

So to create master sets you need a 2 stage process..

1. Load a light frame and all of your Flats and Bias frames, process as normal. This will produce a Bias subtracted Master Flat.

2. Clear the files and load a light frame and your Darks, process as normal. This will produce a Master Dark.

Thereafter just use the Master Flat and Master Dark when processing.

I hope I do not understand you right!

Load a light frame blah, blah, to make a master flat? How does that work?

Load a light frame and your darks to make a master dark?

You make a master dark by combining, say, twenty of them with no alignment and a median output. That's all. Lights don't come into it.

You make a master flat by first shooting all your flats through the various filters. Take all the Red flats and if you do not have flat darks with the same exposure time just subtract bias from each flat. Then combine them as for the darks.

Dennis

Link to comment
Share on other sites

Martin.

Dennis doesn't exactly put it diplomatically I'm afraid, but what he says is right.

You don't need lights anywhere in the equation to make master darks, flats, flat darks, or bias.

The lights are calibrated afterwards with whichever of the above 4 calibration masters you are using.

However, as neither Dennis or I are DSS users, we don't know if you have to load a light in order to make master calibration frames.

Cheers

Rob

Link to comment
Share on other sites

I think this is referring to creating masters only.

Use a light ( could be 2 can't think now ) load flats and bias. Stack. Creates master flat and master bias. Clear. Load light. load darks. Stack creates master dark. Clear. Load all lights, master flat and master dark. There you go.

Interesting thread this.

Dave

Edit - I keep re reading that and hope it's ok. There are too many ways that have been reported on this and other threads. It gets confusing. I'm easily confused :)

Link to comment
Share on other sites

Martin.

Dennis doesn't exactly put it diplomatically I'm afraid, but what he says is right.

You don't need lights anywhere in the equation to make master darks, flats, flat darks, or bias.

The lights are calibrated afterwards with whichever of the above 4 calibration masters you are using.

However, as neither Dennis or I are DSS users, we don't know if you have to load a light in order to make master calibration frames.

Cheers

Rob

In DSS you need at least 1 light frame to make master calibration frames because its written as a one shot process, unlike Maxim where you can just do calibration frames

Link to comment
Share on other sites

I think I said earlier in this thread that I have little time for DSS so my comments on it would not be particularly worthy.

My brain is still going haywire trying to get my head around the idea that you need a light frame (of some sort) to make a master dark. If Maxim doesn't need this approach why would any other software need it?

Aplogies if my frustration with trying to understand some aspects of DSS is making me less than my normally undiplomatic self.

There can't be any copyright on a simple mathematical approach to dealing with image calibration so why the hell do some people go to the trouble of finding ways to do it differently?

Dennis

Link to comment
Share on other sites

Archived

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

  • 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.