Abekta

This page is read only. You can view the source, but not change it. Ask your administrator if you think this is wrong.
====== Siril ======
Siril is an image processing software designed for amateur astrophotography.

{{youtube>v74ZJJXdXVI?large}}
Siril download link: https://siril.org/download/

===== - Data preparation =====
Test OSC data link: https://drive.google.com/drive/folders/1ULlncwRit-lAi5jrdkFx1ytKaImHEuOl?usp=sharing

Generally there are three kinds of calibration files. Darks, Flats and Biases. 
Collect the calibration files for your setup. 
Then sort them into folders(eg. put the darks in a folder named Darks). 
Now be careful about the script that you are going to use to stack the images. Sort them how the script demands. (eg. it might tell you to name the folders with just the first letter).

{{:un:osc_files.png?600|}}

The light files are the actual exposures of the target object or sky. These also need to be sorted according to the script requirements. For the two scripts used in this getting-started tutorial, the light files must be placed in a folder named lights if the data were taken with a one-shot color camera. If the images are multi-filter data, they should instead be organized into separate folders named L, R, G, and B.

{{:un:lrgb_file.png?600|}}


===== - Stacking =====

In this step, we will stack the calibrated images using Siril’s preprocessing scripts. The exact script you use depends on whether your data are narrowband or broadband (one-shot color).


1. ''Narrowband data''

Download the Siril preprocessing scripts for narrowband imaging from the following link:  
https://n8h9z8x9.delivery.rocketcdn.me/wp-content/uploads/2023/08/RemoteAstrophotography_Siril_PreProc_Scripts.zip

After downloading:
 Unzip the file.
 Move the extracted scripts into Siril’s default scripts directory.

To check or locate the scripts directory in Siril, go to:  
<code>Hamburger menu (☰) → Preferences → Scripts</code>


2. ''Broadband / One-Shot Color (OSC) data''

For broadband or one-shot color data, Siril already provides a built-in script.  
Go to:  
<code>Scripts → Get scripts → OSC_Preprocessing.ssf</code>

Note that this script may already be available in your scripts list, so check there first before downloading it again.

{{:un:siril_script_repo.png?600|}}
==== Running the stacking script ====

Once the scripts are ready:

 Click the ''Home'' button in Siril.

{{:un:siril_home.png?600|}}

Select the main project folder that contains all the required subfolders, including:
 Calibration frames (bias, darks, flats)
 Light frames
 After setting the home directory, go to the ''Scripts'' menu.
 Select the appropriate preprocessing script based on your data type (narrowband or OSC).

Siril will now begin the calibration, registration, and stacking process automatically. You can monitor the progress in the ''terminal panel'' on the right side of the interface, where each step of the process is displayed in real time.

{{:un:siril_script_sucess.png?400|}}
==== Opening the stacked image ====

Once the script has finished running successfully:
 Click the ''Open'' button in the top-left corner.
 Load the final stacked FITS file into the image viewer. Then click on the ''Linear'' button in the bottom. Select ''Auto Stretch'' . You will see the an auto stretched image.

At this point, you will have a single stacked image ready for further processing, such as background extraction and color calibration.

{{ :un:siril_qyxy1kyrry.webp |}}
===== - Foreground subtraction (background/true sky extraction) =====

At this stage, the images have already been stacked. The next step is to remove unwanted background gradients using ''Background Extraction'' in Siril.

To perform background extraction:
 Go to the ''Image Processing'' section.
 Select ''Background Extraction''.
 Click the ''Generate'' button to place background sample points across the image.

The number and distribution of sample points can be controlled by adjusting the following parameters:
 ''Samples per line''
 ''Grid tolerance''

{{:un:siril_background.jpg|}}

When placing or generating sample points, it is very important to ensure that they lie only on the ''background''—that is, the darker regions of the sky. Avoid placing points on:
''The main subject''
''Stars''
''Bright nebular or galactic structures''

You can manually refine the sample placement:
 ''Left-click'' to add a sample point
 ''Right-click'' to remove a sample point

Manually place the points on the places where you feel like some glow or artifact is present which is not part of the object or sky.

Once the points are correctly placed on the sky regions, Siril models the background (foreground gradient) and subtracts it from the image. This process removes sky gradients and results in a much cleaner and more uniform final image.

{{:un:siril_background_done.jpg|}}


===== Star position calibration (astrometry) =====

After stacking, the next step is to perform ''star position calibration (astrometry)'', which assigns accurate sky coordinates to the image. In Siril, this process is not fully automatic and requires user input to guide the astrometric solver.


{{:un:siril_astrometry.png?400|}}

To begin, make sure the stacked image is open in the viewer. Then go to the ''Image Processing'' section and select ''Astrometry''. For the solver to work correctly, you must provide approximate values for the imaging system, specifically the ''focal length'' of the telescope or lens and the ''pixel size'' of the camera. These parameters are required by the Siril solver to estimate the image scale and search for a valid solution.

You must also specify the approximate sky position of the image. Enter the ''name of the target object'' in the search field, perform the search, and select the correct object from the results to define the pointing coordinates.

Once these inputs are set, run the astrometric solver. Siril will attempt to match detected stars in the image with stars from an astrometric catalog based on the provided scale and position, and the progress will be shown in the terminal panel.

{{:un:siril_astrometry_pane.jpg|}}

If the solver fails to find a solution, several options can be used to improve the chances of success. You can enable ''Force split solving'', which allows Siril to solve the image in smaller sections, or use ''Downsample image'' to reduce the resolution and make star matching easier. For wide-field images, enabling ''Auto crop'' can also help by removing distorted edges that may prevent a successful solution.

If Siril Solver fails, we have other ways to plate solve the image. One of them is Uploading the photo in ''[[https://nova.astrometry.net/upload|nova.astrometry.net]]''. We will choose the stacked file and upload that. Then we will see a page that will refresh in every 10 seconds until a solution is found. Then We will see a page like the image. From there click on the New Fits Image to download the image. It is the exact same image with the world coordinate systems  solution in its header.

{{:un:nova_astro.jpg?|}} 

When the astrometric solution is successful, the image is plate-solved, meaning that each pixel is now associated with precise celestial coordinates. This calibration is required for subsequent steps such as photometric color calibration, image annotation, and scientific measurements.

After clicking the 3rd and 4th button bsedide the Autostretch button you can see the co ordinate planes, and the object identifiers in the image. 

{{:un:siril_astrometric_solution.jpg?|}}

===== - Spectrophotometric color calibration =====

After completing astrometric calibration, the next step is ''spectrophotometric color calibration (SPCC)'', which provides a physically accurate color balance by comparing the measured flux of stars in the image with spectrophotometric reference data from star catalogs. 


To perform SPCC, ensure that the astrometrically calibrated image is open in the viewer. Then go to the ''Image Processing'' section and select ''Spectrophotometric Color Calibration''. Siril uses the solved sky coordinates to identify stars in the image and match them with cataloged stars that have known spectral energy distributions.

You need to select the appropriate ''filter set'' and camera configuration so that Siril can correctly model the response of the imaging system.

Once all required parameters are set, run the SPCC process. Siril will fit the observed stellar colors to the cataloged spectrophotometric data and compute correction factors for each color channel. 

{{:un:siril_spcc.jpg?|}}
===== Stretching =====
{{:un:siril_unstretched.jpg?|}}

**Histogram stretch**

To perform a histogram stretch, open the histogram transformation tool in Siril with the linear, background-subtracted, and color-calibrated image. First, locate the histogram of the image. Most of the data will be concentrated near the left side.

Start by sliding the ''midtone'' slider slightly to the left. This will move the main peak of the linear histogram a little to the right and make it slightly wider. The ''Midtone'' Slider is the middle one in the bar under the histogram curve. And the left most one is the ''Shadows'' slider.
Observe how the image brightens and the faint structures become more visible. 
Repeat this adjustment multiple times, in small increments, gradually moving the midtone slider while monitoring the histogram peak and the image appearance. The goal is to achieve a balance between the highlights and shadows. 
Next, adjust the ''black point'' slider carefully. Pulling the black point slider to the right will shift the histogram to the left, increasing contrast. Be careful not to clip data: the leftmost end of the histogram should not cut into significant pixel values. 
Continue alternating between small midtone and black point adjustments until the histogram is widened appropriately, the faint details are visible, and both highlights and shadows are preserved.

The key is to perform these adjustments gradually and repeatedly rather than making large, aggressive changes in a single step.

{{:un:siril_stretch_1.jpg|}}

** Generalized Hyperbolic Stretch (GHS) **

Once you have understood the histogram stretch tool, we can use more advance tool like this GHS to have finer control over the stretching. Evn you can use both of the stretching one after another to further refine the image.

Open the GHS tool in Siril with the unstretched image. 
Adjust the ''stretch factor'' slightly to enhance faint regions while keeping the bright areas from saturating. Small incremental changes work best.
Move the ''symmetry point'' slider slightly to position the stretch curve according to the main data peak. This helps preserve star cores and highlights while brightening the faint structures.
Monitor the histogram and the image as you adjust. Repeat the adjustments in small steps until you achieve a balanced stretch where faint details are visible and bright areas are not clipped.
Avoid extreme settings: do not pull the sliders to the edges, as this will clip the highlights or shadows and degrade image quality.

{{:un:siril_ghst.jpg|}}


===== - Sharpening and denoising =====
Cosmic clarity: https://drive.google.com/drive/folders/1MKsuA4XgIMd0cJshJHEWdN782cXvzR3Y?usp=drive_link

After stretching and enhancing the image, the next step is sharpening and denoising using ''Cosmic Clarity'', an AI-based tool integrated into Siril via scripts.

To get started, go to the ''Scripts'' section in Siril and add the Cosmic Clarity scripts for both ''Sharpening'' and ''Denoising''. Download the Cosmic Clarity application from the provided drive folder, unzip it, and in the script settings point to the unzipped folder and required files. Once everything is linked, you can run the scripts and adjust the sliders to find a suitable balance for your image.

For denoising, there are three options:  
 ''Full'' applies noise reduction to the entire image.  
 ''Luminance'' applies denoising only to the brightness channel.  
 ''Both'' applies denoising to both luminance and color channels simultaneously.  
Before Denoise: 
{{:un:siril_denoise1.jpg|}}
After Denoise
{{:un:siril_denoise2.jpg|}}


For sharpening, you can choose:  
 ''Stellar sharpening'', which targets only the stars, enhancing their crispness.  
 ''Full sharpening'', which enhances both stars and non-stellar structures such as nebulae and galaxies.
Before Sharpening:   
{{:un:siril_sharp1.jpg|}}
After Sharpening:
{{:un:siril_sharp2.jpg|}}

Cosmic Clarity uses AI models trained on astronomical images to intelligently separate stars from background structures and noise, enhancing fine details while minimizing artifacts for a cleaner and sharper final image.