Differences

This shows you the differences between two versions of the page.

--- un:siril [2026/01/25 21:37] – [Generalized Hyperbolic Stretch (GHS)] shahal
+++ un:siril [2026/01/31 00:32] (current) – [1. Data preparation] shahal
@@ Line 6: / Line 6: @@
 ===== - 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.
@@ Line 143: / Line 144: @@
 {{:un:siril_spcc.jpg?|}}
 ===== Stretching =====
+{{:un:siril_unstretched.jpg?|}}
 **Histogram stretch**
@@ Line 148: / Line 150: @@
 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.
+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 a basic histogram stretch is applied, the Generalized Hyperbolic Stretch (GHS) can be used for finer control:
+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 stretched 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|}}
@@ Line 175: / Line 181: @@
 For denoising, there are three options:
-* ''Full'' applies noise reduction to the entire image.
+ ''Full'' applies noise reduction to the entire image.
-* ''Luminance'' applies denoising only to the brightness channel.
+ ''Luminance'' applies denoising only to the brightness channel.
-* ''Both'' applies denoising to both luminance and color channels simultaneously.
+ ''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.
+ ''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.
+ ''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.