How to Process and Stack Seestar FITS Files in Siril

Discover the power of stacking FITS with Siril and the Seestar S50 Smart Telescope. Enhance your astrophotography like never before!

By Tim Trott | Astrophotography | February 12, 2024
2,371 words, estimated reading time 9 minutes.

What is Stacking and Why Is It Important?

In astrophotography, stacking refers to the process of combining multiple images of the same astronomical object to improve the overall quality of the final image. This technique is particularly useful for reducing noise, enhancing detail, and increasing the signal-to-noise ratio.

Astrophotographers take several exposures of the same object using their camera and telescope setup. These exposures are often taken over a period of time to gather more light and increase the signal. These exposures are called "subs" or "lights" and allow for shorter exposure times for each exposure, but when combined have the effect of one very long exposure. This is beneficial for many reasons:

  1. Noise reduction - camera sensors produce noise through ISO settings, sensor type, heat and other factors. By stacking multiple exposures these are averaged out, improving signal-to-noise ratio.
  2. Improved tracking - the longer the exposure the more vulnerable the final image is to the tracking mount. Shorter exposures require less precision in the tracking mount and errors can be easily remedied.
  3. Detail is enhanced due to the improved signal-noise ratio and tracking accuracy.
  4. Satellites, clouds and aircraft can all ruin a long exposure. Say you left a perfectly aligned mount and perfect camera exposing a galaxy for an hour, then an airplane flies into the field of view and your picture is ruined by the navigation lights streaking across it. With individual subs you can easily remove the affected frames without losing the entire exposure.
  5. Light pollution is easier to remove in lots of short exposures, rather than one long exposure where it will often drown out the signal from the target object.

Overall, stacking is a powerful technique in astrophotography that enables photographers to produce high-quality images of celestial objects, even under challenging conditions such as light pollution or limited exposure time.

Configure Seestar for Individual FITS Files (Subs)

The Seestar by default automatically captures 10 second exposures and stacks them on the fly, the result you see on the screen is a "live stack". When you are done it will save the JPEG to your phone and a raw FITS file to the internal storage. You can enable an option in the menu which will tell the Seestar to save each individual sub as a FITS file which you can then later use to stack manually in processing software such as Registax or Siril as described here.

Enable "save each frame in enhancing" to store individual FITS files.
Enable "save each frame in enhancing" to store individual FITS files.

To have the Seestar S50 store each individual FITS file simply head into the Advanced Settings section of the app and toggle "Save each frame in enhancing" on.

What is a FITS File?

FITS Files contain a lot of header data, such as the objects RA and Dec, filters used, GPS coordinates, time and more.
FITS Files contain a lot of header data, such as the objects RA and Dec, filters used, GPS coordinates, time and more.

A FITS (Flexible Image Transport System) file is a standardised file format commonly used to store and exchange astronomical images and data. FITS files are designed to be highly versatile and flexible, allowing for the storage of various types of data associated with astronomical observations, including images, spectra, and tabular data. FITS files are particularly popular in professional and amateur astronomy communities due to their flexibility, interoperability, and support for scientific data analysis. They are used for storing images captured by telescopes, satellites, and other astronomical instruments, as well as for archiving and sharing observational data within the scientific community.

Stacking FITS with Siril Script

The easiest way to stack FITS with Siril  is to use the stacking script as this will automate the process. If you want finer control over the process you can manually register and stack the subs through the tabs in the right hand pane. I'll save those instructions for a later tutorial because this article will focus on the processing of the stacked FITS.

If you wish to work on the stacked FITS from Seestar without stacking the individual frames, please continue to the Post Processing section.

In order to stack the FITS subs we need to download a script for Siril. By default Siril works with "lights", "darks", "flats" and "biases", lights being the actual exposures, darks and biases are for noise reduction from the camera sensor and flats for correcting vignette and uneven field illumination. The Seestar only captures lights and that's all we have to work with, so we need a script that will only work with the lights. Head over to the official Siril scripts repository  and follow the instructions to download and install them, or alternatively for the purposes of what we are after here, click into the preprocessing folder, click on OSC_Preprocessing_WithoutDBF.ssf and download it to your Siril scripts folder. In Windows this is C:\Program Files\Siril\scripts. Once installed, restart Siril.

Next we need to set up a folder to work in, this is the home folder or working folder. I like to copy the MyWorks folder from the Seestar to my data drive. Sestar automatically creates a folder for each target, and when individual subs are on it will create a separate _subs folder, e.g. M81_subs. Within this folder, create a folder called lights and move the FITS in there.

Siril Folder Structure for Stacking with Scripts
Siril Folder Structure for Stacking with Scripts

Now, inside of Siril click on the blue home icon, then select the M81_subs folder. This will set the home folder, from which we can create a stacked image.

Running Siril Script
Running Siril Script

When the script runs the console windows will show logs of logging data which will scroll pretty fast. When it is done it should say something along the lines of "Script execution finished successfully, total execution time 15.92 s". If not, check the output for errors, correct them and try again.

Within the home directory you will now have a result.fit file which is the result of the frames aligned and stacked. You can now move to the Post Processing section where we will stretch and enhance to bring out the details.

Post Processing FITS for Maximum Results

When opening FITS files don't worry that they are dark, this is the RAW data that needs stretching.
When opening FITS files don't worry that they are dark, this is the RAW data that needs stretching.

If you haven't already set the home folder, do so now by clicking the home icon. Then, click the open button and select the result.fit from above, or the pre stacked FITS from Seestar. This will open the file and you can see the image in the left hand side of Siril. Don't worry that it is very dark, this is normal. We need to perform some actions to stretch and bring out the details before you can see anything.

Autostretch allows us to quickly see what we are working with without altering the original data.
Autostretch allows us to quickly see what we are working with without altering the original data.

Because the image is so dark we need to brighten it so we can see what we are working with, but we need to do it in a non-permanent way so we don't affect the final result. We can do this by auto-stretching the preview image by clicking on the Linear drop down and changing it to Auto-stretch with High Definition. Most likely the image will turn green, this is normal. Simply uncheck the linked button next to the auto stretch drop down to correct it.

Image Plate Solving compares your image with star catalogue data by matching star patterns.
Image Plate Solving compares your image with star catalogue data by matching star patterns.

Next we need to plate solve the region which will correctly orientate the image and allow for photometric colour calibration in the next step. Click on the Hamburger Icon in the top right, then Image Information, then click Image Plate Solver. Just click Go on this window and check the console. Most, if not all the time it reports success and flips the image. If it is unable to plate solve, uncheck the Auto checkbox under catalogue parameters and select each of the star catalogues in turn. Usually one will work. Once compete, close the image plate solver window.

Cropping prevents unwanted data from affecting the stretching algorithm.
Cropping prevents unwanted data from affecting the stretching algorithm.

If you need to, this is a good time to crop the image since the next few operations are going to work on the entire image data and we don't want any bad or unnecessary data affecting the results. Right click the image and from the Selection menu select the aspect ratio or free selection. Then drag a box around the area to keep. Once you are happy, right click again and click on Crop.

Background Extraction helps correct noise, gradients and banding which are all ugly and distracting.
Background Extraction helps correct noise, gradients and banding which are all ugly and distracting.

Next we are going to perform a background extraction. This will help remove some of the background noise, gradients and other undesirable effects in the background of the image. Background extraction can be found from the image processing menu and will open a new window. The options can be left as default for now. Click on Generate and it will put a grid of red squares on the image. You need to ensure that these are not on stars or your target. If any are then right click to remove. You can left click to add a square. Now click Compute Background and it will show a preview of the new background. You may need to alter the grid of squares to further enhance. If you have ugly colour banding you can enable the "Add dither" option which should help even it out.

Photometric colour calibration re-balances the RGB channels to get as close as possible to the true colours of the object.
Photometric colour calibration re-balances the RGB channels to get as close as possible to the true colours of the object.

Now we are going to apply some colour correction using photometric data. If you have already plate solved the image the values will be pre populated, otherwise you can search for an object and plate solve from within this screen. Head to the Image Processing menu, then Color Calibration then Photometric Color Calibration. Click on Ok and it will apply colour correction to the image. This can be helpful when light pollution has created an orange hue on stars and galaxies.

Now we can finally tackle the noise in the image using some noise reduction. There is a Noise Reduction option in the image processing menu, however I find that the Deconvolution produces better results. On the Deconvolution window take all the defaults and click apply. It may take a few seconds to do this part and you'll see a few messages in the console. When complete it should say something like Execution time 11 s. Close the window.

Now is a good time to save the FITS file. Save it using a unique name so we don't ruin the original or stacked data.

StarNet Star Removal will create two files - one with the stars only and another with just the object.
StarNet Star Removal will create two files - one with the stars only and another with just the object.

Now we can start the stretching and enhancing parts. Switch back to the liner view to make everything dark again. We're going to use a program called StarNet to remove the stars so we can enhance the stars and the object separately. Head to the StarNet website  and download the command line tool for your platform. Extract the archive to a folder, I put mine in C:\Program Files\StarNet\. Within Siril, click the hamburger icon on the top, then preferences, then miscellaneous. Under software location browse to the folder for StarNet. On the Image Processing menu, select Star Processing, then StarNet Star Removal. In the window, enable the "Pre-stretch linear image" option then click Execute. This will create two FITS in the home directory. One will have the stars - starmask_ and the other the object - starless_.

Stretching the star mask with modified arcsinh transform.
Stretching the star mask with modified arcsinh transform.

Open the starmask file first, then from the Image Processing menu select Generalised Hyperbolic Stretch Transform. In the window select Modified arcsinh transform in the type of stretch option box and drag the stretch factor to a point where the stars look good. I find a value around 7 works well. Click on apply when you are happy and close the window. Now go to Image Processing and Colour Saturation. Bump this up to get some colour into the stars and click apply. Finally click save to save the starmask.

Setting the symmetry point here to pivot on the object details
Setting the symmetry point here to pivot on the object details

Now open the starless file. Go to Image Processing and Generalised Hyperbolic Stretch Transform again, but this time ensure that the transform type is Generalised Hyperbolic Transform. In the top left hand side of the window there is a zoom control. Set this to 100 and press enter. Now click on the middle of the graph's vertical axis at the point where it meets the curve as shown on the image left. This will set the symmetry point and make the stretch easier.

Stretching the starless file to bring out object details
Stretching the starless file to bring out object details

Adjust the stretch factor to bring out the details, don't worry about the background brightness, we'll remove that shortly. When you have a brighter target, click apply, change the stretch type to linear and adjust the black point to recover the background. You don't want a pure black background, just enough that its dark and you don't lose details. You can then go back to the generalised stretch and repeat if necessary.

Removing the green noise greatly improves the quality of the image.
Removing the green noise greatly improves the quality of the image.

When you are happy with the stretch we can remove the noise that was introduced using the Image Processing and Remove Green Noise function, just take the defaults. You can also use the noise reduction option if the background is still noisy. Finally you can increase the saturation if needed. Save the starless file.

Recombining the starmask and the starless object files to crate the final result.
Recombining the starmask and the starless object files to crate the final result.

Now we get to the exciting part, recombining the starless and star mask for our final result. On the image processing menu, click the star processing menu, then star recomposition. In the background stretch section select the starless FITS file, and in the right hand side star stretch, select the starmask file. You can make a few final tweaks using these sliders if required. Finally close the window to complete the process. Make any final crops then save the file. You can also export to PNG or TIFF to further process in Photoshop.

M81 (Bodes Galaxy) and M82 (Cigar Galaxy)
M81 (Bodes Galaxy) and M82 (Cigar Galaxy)
Was this article helpful to you?
 

Related ArticlesThese articles may also be of interest to you

CommentsShare your thoughts in the comments below

If you enjoyed reading this article, or it helped you in some way, all I ask in return is you leave a comment below or share this page with your friends. Thank you.

This post has 7 comment(s). Why not join the discussion!

We respect your privacy, and will not make your email public. Learn how your comment data is processed.

  1. MG

    On Thursday 21st of March 2024, Mike G said

    Just making the leap over to the dark side of astrophotography from visual and EAA. Just got a Seestar. I’m anxious to try your tutorial when the skies clear and I can gather some data.

  2. NI

    On Tuesday 19th of March 2024, Nick said

    Thanks for this article. I have viewed many tutorials on this subject but you have put it together in such a way that I was able to actually make progress using Siril with the Seestar files. I found that the most important parts left out of most tutorials is well covered in this post.

  3. M5

    On Thursday 14th of March 2024, M 57 said

    Thank you for this!

  4. JB

    On Wednesday 13th of March 2024, Jean-Pierre Bolengo said

    Thanks for your good instrution. I have tried it on Messier M35, so no nebula.
    When I try to save it, it says that it can't be saved due to the Auto-Stretch mode.
    What I'm I doing wrong?

  5. JB

    On Tuesday 12th of March 2024, Jean-Pierre Bolengo said

    Thanks for your good instrution. I have tried it on Messier M35, so no nebula.
    When I try to save it, it says that it can't be saved due to the Auto-Stretch mode.
    What I'm I doing wrong?

  6. PW

    On Tuesday 27th of February 2024, Phil Wilkes said

    Very useful on stacking and processing for Seestar

  7. AN

    On Friday 16th of February 2024, Andrew said

    Amazing and very detailed guide. Thank you for taking the time to write this.