Jump to content

Banner.jpg.b89429c566825f6ab32bcafbada449c9.jpg

flats and darks


planetdnb

Recommended Posts

sorry if this has been answered elsewhere!

Is it acceptable to build up a library of Darks that can be used on my images.

For example when the nights are cloudy can i make a collection of darks that I can use at a later date. I appreciate each collection of darks would need to be taken in similar temperatures/conditions. So would it be ok to once a month do a collection of darks so that I don't have to take them on the night?

If it is acceptable would it be better to take 1 or 2 darks on the night and then combine them with the library, or is this idea just a waste of time and I have to take my darks on the night of imaging?

With regards to flats I don't like the sound of doing them on the horizon so would this method be acceptable:

Put a sheet of white paper over the front of my scope and aim it at a halogen floodlight I have conveniently mounted on my shed. I could put the front of the scope literally a foot away from the light source.

Also, when taking flats, is it the same process as taking darks where they have to be the same settings? Exposure time, ISO, temperature etc or do I just have to take a quick snap at the same ISO?

Thanks for the help.

Link to comment
Share on other sites

In theory, darks can be taken at any time so long as they are of the same exposure time and at the same temperature. However, be aware that over time, the characteristics of a CCD can change a little eg more hot pixels, different distribution of 'dark' current, which can have an impact on the final image because the darks no longer 'match' the lights. Although over say a month, it shouldn't be a problem. I maintain a library of master darks and refresh them evry month or so.

IMO flats require a little more care. You need a fairly constant light source and to ensure that when you take them, you don't saturate your CCD. Make sure that the ADU level on the flats is somewhere in the range 30-60% of the max ADU of the camera. So if its max is 65K, then aim for say 20k-40k on the image. With a halogen source, I suspect you will need to reduce the light significantly. Good quality paper will suffice but i suspect many sheets will be required. Diffuser material is a good alternative. Alternatively a clean white t-shirt or something similar over the scope will also do. Make sure that the exposure time is at least 0.5sec and add more paper/t-shirt if necessary. You will need to refresh your flats more often though eg everytime you remove or rotate the camera. But since flats take so little exposure time in comparison to darks, IMO it's less of a chore

Good luck!

Steve

Link to comment
Share on other sites

thanks for the prompt response, the darks was as I thought. I will spend a few evenings once a month building a library so that the conditions will be mostly the same through the month.

With regards to the flats, I am using a canon 400D not a CCD. I'm sure the theory is the same but being a noob I am a bit lost at ADU. What is ADU and how do I verify that I am in the 30-60% range?

Thanks

Link to comment
Share on other sites

ADU's are 'Analogue to Digital Units' and can be thought of as the levels of brightness for image processing. You might find this value in your camera's handbook or someone else here may know the value. Alternatively, you might find other values such as 'Well Capacity', ie how many electrons can be stored in each pixel and 'Gain' ie how electrons are converted into ADU's. Dividing Well Capacity by Gain will produce the ADU for your camera. eg 48k Well Capacity with a Gain of 0.8 will give the camera's max ADU as 60k

I'm not familiar with your camera but at this stage aim for about 25K ADUs on your flat image.. Depending on which software you're using to take the images, you should find some tool which will allow you to inspect the image and provide the information.

HTH

Steve

Link to comment
Share on other sites

Taking just one or two darks is a waste of time. There will be random noise elements in them that will add noise to your image.

Take at least 20 of them and then stack them. This will even out any randomness and give you a true representation of the noise.

Darks will also contain any bias noise from the camera electronics etc so if you apply darks, you don't need to apply bias also.

With flats, don't use too bright a lightsource, and you will need to apply bias frames as the flats will contain bias noise, which, if it isn't removed, will add noise to your final image.

I can't speak for other programs, but with Maxim, if you don't bias subtract your flats, they don't apply properly and they either under or overcompensate.

Bias frames are taken with no light entering the camera, and at your shortest possible exposure. I redo my darks and bias about every 6 months.

So, in a nutshell....dark subtract and flat subtract your light frames, and make sure you bias subtract your flats. Darks and Bias can be used for a few months but flats must be reshot every time the optical train is disturbed.

Take at least 20 of each type of calibration frame and stack them to remove any random element.

Cheers

Rob

Link to comment
Share on other sites

thank you so much for the excellent advice. I will have a go at finding a suitable flat method.

With my darks I am building up a library of 25 of each type (Exposure/ISO etc) and combine them in Neb.

Now onto the Bias frames...

If darks contain the Bias do I just subtract the darks from the flats or do I need to take specific bias frames which I subtract from the flats. If its the latter how do I take the bias? Is it the same as a dark but with an absolute minimal exposure time?

Thanks again

Link to comment
Share on other sites

You should read the manual for Nebulosity as all that should be explained there.

Yes, a bias is a shortest exposure dark. The temperature does not matter much.

Yes, you need to stack bias frames as they are all a bit different and you're only interested on what DOESN'T vary from frame to frame.

Link to comment
Share on other sites

On this occasion, it depends .....!

When the lights and darks have the same exposure, you're correct in that the darks already contain the bias so there's little point in removing the bias from the dark and then the bias from the light. It's the same result. you may as well leave the dark including the bias

However, bias frames really come into their own when the darks don't have exactly the same exposure time as your lights. Then, some software will 'scale' the darks to the correct exposure after the bias has been deducted. Under that situation the bias has to be removed from the light. This approsch to scaling certainly happens in Maxim but I'm not sure whether Nebulosity treats scaling in the same way. At this stage, you either need a Nebulosity expert or a read of the Help/Manual!

Again, it's best to take a series (25 or 30) of bias frames rather than rely upon 1 or 2. They're even quicker that taking lights! Treat them as Darks but set the exposure as low as your software will allow. You're trying to capture the 'dark current' which is present in the chip merely from pressing 'start'

HTH

Steve

Link to comment
Share on other sites

Hi

I agree with everything said above but would like to add my thoughts.

For flats I don't subtract bias but I do subtract darks with the same exposure length as the flats. My theory is that as bias is effectively a dark frame of minimal exposure I also need to add in the dark current associated with the duration of the flat exposure. Since this is very short it will be similar to a bias but a little higher current. Given that the flats I take are under a second I take darks for them at the same time.

Now, in processing I subtract normal darks but divide the result by the dark adjusted flats. Think of the flats as an inverse multiplier, making the darker areas lighter with respect to the rest of the field.

Hope that helps.

Robbie

Sent from my iPhone using Tapatalk

Link to comment
Share on other sites

lol. now i am just confused.

If the darks contain the bias surely subtracting the darks and the lights from the main image is all I need to do. If i subtract the darks from the lights and then the darks from the main images am I not subtracting the bias twice?

Or are you suggesting that I just subtract the darks from the lights and then the lights from the main images?

Link to comment
Share on other sites

Lights are your actual images.

Robbie is correct....what he subtracts from his flats are known as flat darks.

If you subtract flat darks from your flats, then you don't need to use bias frames at all, as the bias is contained in the flat dark and the dark.

So.....subtract darks and flats, and either bias subtract or flat dark subtract your flats, but not both.

You aren't subtracting bias twice, only once, and that is with the darks.

The bias (or flat darks) subtract bias from your flats.

When the flat is applied to the light (the actual image), bias isn't involved as it has already been removed from the flat.

HTH.

Rob

Link to comment
Share on other sites

Rob is spot on!

Either take darks the same length as your flats, and subtract the flat darks from the flats, or take bias frames and use them. Bias are almost identical to flat darks for short exposures.

Your your real darks, they must be at the same temp, ISO, and duration are your astro-images (or lights as they are called)

From these astro images, subtract either the darks or bias frames, but not both. Then apply the flat (with the flat dark or bias already subtracted). Technically the image is divided by the flat, but thats neither here nor there.

hope that helps

paul

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.