Jump to content

Banner.jpg.39bf5bb2e6bf87794d3e2a4b88f26f1b.jpg

Should flats TOTALLY elimate all gradients in an image?


swag72
 Share

Recommended Posts

Last night for the first time I took flats, having remembered that I already had a light panel in the back of my wardrobe that I'd bought years ago for another purpose. Strange thing to have hanging around in your wardrobe I know!!

So, the flats were taken for the 2 filters I bagged yesterday. They were around 28K ADU and showed gradients. Today I did a quick stack of red data, one with flats and one without. There was a definite improvement in the gradient, but there was still a little there. This was easily sorted out with DBE in PI, normally the gradient removal takes super human effort!

My question is whether correct flats should eliminate ALL gradients, or does it just make them easy to deal with? If the gradient isn't totally removed does that mean that my flats are not quite right? Perhaps my ADU is a little high?

Would welcome your thoughts on this, as I would say I have achieved a high percentage improvement, but not 100% which is what I would have expected.

Link to comment
Share on other sites

Gradients caused by outside influences - light pollution or the moon for example will not be corrected by flats.

Flats will only compensate for the differences in illumination in the optical path and remove imperfections such as dust bunnies.

Flats do much good, but for everything else there's DBE!!

Edited by johnrt
Link to comment
Share on other sites

I suppose that some of the gradient could be caused by light pollution (particularly wide-field). Or, in the flats, possibly the shutter if the exposure time is too short? (not sure about this one).

Can't say I've ever had a flat deal with all the gradient perfectly myself.

Edited by Shibby
Link to comment
Share on other sites

So the fact that I have some gradient left shouldn't be a worry then - Thanks for that, the flats have obviously worked!!!

It is a good idea to use a bias control frame to remove the bias from all your flat subs before stacking and making your flat control frame. That way, any gradient caused by the camera bias will be eliminated from the flat. My Canon 1000D shows a gradient across the ccd in a bias frame.

Edited by FrankieValley
Link to comment
Share on other sites

Getting complicated now!! I just chuck it all into DSS and tell it what the files are, lights, flats or bias!! Just had a quick run through and the bias took out a little more of the gradient that I tend to get along the bottom (like a bright line) - Definitely looks better using calibration frames than not!!!

Link to comment
Share on other sites

Not sure how DSS works in practice, but essentially, subtract bias from darks and flats to make master dark and master flat. Then subtract bias, master dark and master flat from your lights.

It can be even more complicated..... but I tend to leave it there!!

I'm sure you mean 'Then subtract de-biased master dark and DIVIDE BY master flat on your lights.

Mike

Link to comment
Share on other sites

Not sure how DSS works in practice, but essentially, subtract bias from darks and flats to make master dark and master flat. Then subtract bias, master dark and master flat from your lights.

It can be even more complicated..... but I tend to leave it there!!

This is not correct.

Darks contain bias information, so you don't subtract bias from darks, nor do you subtract bias from lights if you are dark subtracting them.

If you do so, you are performing a double subtraction.

Subtract bias from your flats only.

DSS has you throw everything in, but this isn't required....I had a long conversation with the chap who designed DSS about this as we had a thread going about a year ago to get to the bottom of all of this.

In all other programs apart from DSS, tcalibration is done as I describe, and it works that way in DSS too....you just uncheck some boxes I believe.

Rob

Link to comment
Share on other sites

Rob, that is not correct either

For full wack, darks should have bias removed before use otherwise any scaling applied for temperature or exposure variation also scales the bias pedestal as well.

Application of the bias to the light is a straight subtraction. Application of the bias-subtracted darks to the light is a second subtraction.

Division by the flat field which has itself been bias and dark-subtracted is the last operation to the light frame.

however if the bias and dark are taken close together in time, you can ignore bias-subtracting the darks since you won't be doing any scaling

Cheers

Mike

Link to comment
Share on other sites

Can I ask that you stop picking holes (or not), in people's understanding and help Sara overcome the problem she posted. Thanks.

Correcting something that is incorrect is not picking holes. Your advice was wrong. If she'd gone by that, her calibration would be messed up and she would not overcome the problem.

@Skybadger. What you say is correct if you are using darks that are not temperature matched to your lights. What I said was correct if one is doing what should be done and using temperature matched darks, which should not be bias subtracted, as you pointed out.

Rob

Edited by RobH
Link to comment
Share on other sites

Correcting something that is incorrect is not picking holes. Your advice was wrong. If she'd gone by that, her calibration would be messed up and she would not overcome the problem.

@Skybadger. What you say is correct if you are using darks that are not temperature matched to your lights. What I said was correct if one is doing what should be done and using temperature matched darks, which should not be bias subtracted, as you pointed out.

Rob

whatever

Link to comment
Share on other sites

Sara....You've already had the correct answer to your question....don't let yourself get sidetracked by esoteric discussions about how many angels can fit on the head of a pin :D.....even with flats applied, there can still sometimes be a bit of gradient left, and also LP gradients will still be there....DBE in Pixinsight makes short work of these (can't figure the rest of PI out though!!)

Rob

Link to comment
Share on other sites

I agree with Rob and since Sara is using a set point camera she'll be fine with his advice.

DSS is very worthy but I feel out of control with it, though I suppose I could do the reading. Personally I'd rather make a master flat first along with a master dark. Stack the flats using a bias master as the dark frame. This means you can chuck the gazillions of subframes involved off your computer, for one thing. It also makes it easier to find out what is wrong if something IS wrong.

Olly

Edited by ollypenrice
Link to comment
Share on other sites

Thanks folks! It all went above my head a little way in! Didn't realise that flats won't cure LP - That must be what I have along the bottom of my subs then. Damn the streetlight!!!

Edited by swag72
Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
 Share

  • 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.