Show Posts

This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.

Messages - astroedo

Pages: 1 [2] 3 4 ... 11
Gallery / Re: 2017 eclipse - HDR
« on: 2017 August 28 12:28:10 »
Wonderful image!

I'm trying to process a few images from a friend (unfortunately I wasn't in the US  :'( )
I can align images with FFTRegistration and easily create the HDR image, but, when I try to compress the dynamic range with HDRMultiscaleTransform I get a Bright ring aroun the Sun.

Probably the wavelet based HDRMT gives a sort of ringing effect cause by the fast drop in luminosity near the moon edge.
If I reduce the number of layers, the bright ring, shrinks but never disappears.

As a reference I'm uploading a sample image (it's a single shot not the HDR one, but the effect is the same).

Do you have any hint on how to manage this problem?

Thank you


General / Re: asking advice about ubuntu vs Windows 10 version
« on: 2017 August 21 08:59:40 »
Thank you Juan for your precious information.
What do you think about UBUNTU with GNOME interface?
I've practically no experience with linux

PixInsight should run on this machine, but it will put everything to the very limit, especially with version 1.8.5. I strongly recommend you try to upgrade your processor and RAM. Even a Core i5 will run much faster with 8 GB of RAM.

Consider that this notebook is the "on the field processing" machine.
I use it only for fast processing on the field or for the workshops that I perform in northern Italy.

My processing PC is a I7 7700 with 32 GB of RAM and a SSD Hard disk.

General / asking advice about ubuntu vs Windows 10 version
« on: 2017 August 19 08:31:57 »
Hi all,

I'm formatting and reinstalling my old notebook.

It is an ASUS notebook with a INTEL CORE DUO processor and 4 GB RAM + 1 GB Graphic RAM.
My main O.S. is Windows 10 64 Bit where PixInsight runs decently and I'm happy about it, but I was evaluating the idea of making a multiboot system with Ubuntu (or Xubuntu, Lubuntu that seems lighter for my old machine).

- Has someone tested Linux PI Version compared with windows version on the very same machine?
- Which system has better performances? (If one in better than the other)
- In your opinion, is my idea good or fool?
- Are there other Free Linux distributions more suitable for this job?

Thank you in advance for your help.


General / Linear "L" RGB Combination
« on: 2017 March 05 10:27:54 »
Hi All,

it is a bit of time that I'm trying to find a way to make an "L" RGB combination at linear stage.
We all know that the real LRGB combination must be done on NON linear images, matching the luminosity profile of the L image with the RGB one and combining in with the LRGBCombination Process (or Channel Combination); this because the CIE Lab color space is nonlinear, so we cannot use a linear L.

Googling on the internet I haven't found any solution, except an old post here on the forum

where Juan suggests a method to combine a linear L with RGB.
Unfortunately I haven't managed to make it work: on my images this method creates horrible Green and magenta artifacts (don't know if I'm doing something wrong).

So I've tried to find a solution by myself, and, maybe, I've found it.

This is my workflow:

I have 4 images: R G B and "L"

  • RGB combination with ChannelCombination
  • BackgroundNeutralization
  • ColorCalibration
  • Extract I Channel from HSI color space with ChannelExtraction
  • Use LinerFit on "L" using  I  as reference
  • Insert fitted "L" in RGB image with ChannelCombination (Use HSI color space deselect H and S checkboxes and use "L" as I )
Thats all folks!  :D

Processing the image the final result is practically Indistinguishable from a normal LRGB.
The I component of the HSI color space is a linear combination of RGB so it is linear and can be combined at linear stage.

We can call it IRGB Combination instead LRGB.  ;)

What do you think about this workflow?
Am I doing somthing wrong?



General / Re: Process Container - AutoHistogram + HT
« on: 2016 September 27 10:08:03 »
Hi Roberto,  you can use the clipping feature of AutoHistogram.

At the top of the process dialog you can enable the clipping value.

Experiment with the value to get the desired result.
Clear skies


It's very simple indeed! ;-)

You must use it for denoising at linear stage: activate the mask checking the "Linear Mask" checkbox.
Create a Preview on your image to test the masking and denoising parameters.
Deactivate the Screen Transfert Function and Check the "Preview Mask" flag.

Set the "Amplification" factor: this is a multiplicative factor applied to the image that enhance it to create a protective mask.
Typical values range from 60 to 110.
Drag and drop the blue triangle of MMT process on your preview: you will preview the mask that will be applied.
Trim the amplification factor and the smoothness factor to get the desired masking effect.

When done Deactivate the "Preview Mask" flag, reactivate the Screen Transfer function and proceed as usual with denoise.

The denosing effect will be applied with the built in masking.

Clear Skies

General / Using the transformation matrix of star alignment.
« on: 2016 August 25 03:46:59 »
Hello to all.

As you know, after the star alignment process, the transformation matrix for the align operation is shown in the process console.

Here my question: can I use this matrix to apply the very same transformation to another image?

For example I have Image A, image B and a starless copy of B (let's call it C)
I want to register C on A: I can register B on A and then I want to use the same transformation on C.
Is it possible to use the transformation matrix to replicate the same transformation from B to C?

If yes: HOW?

Thank you.

Are you sure that the other images are greyscale?
You can read the color space in the top bar of the image.

Bug Reports / Re: DBE bug?
« on: 2016 January 07 12:02:24 »
Very happy to hear about it Juan!

I can't wait for the new tool!

By the way I've experimented the same problem: Latest PI Release on Windows 10 Home, Intel Core Duo processor, 4 GB RAM.



Hi all

During the Halloween week end I got some Narrow Band data on the Flaming Star Nebula.
Waiting for the OIII data I've tried a weird experiment: a false color composition of Ha and SII

Ha and SII emisson lines lays both in deep red so it is impossible to render color differences in a real way: the only option is a false color composition.

In this image Hydrogen emission has the usual pinkish color due to the Ha and Hbeta emission lines.

As said before the SII emission line lays in deep red, but on Earth the cristalline sulfur has yellow hues so my goal was to render the SII in yellow.

To accomplish My Goal I've used the PixelMath tool in PixInsigth.

The composition formula was:

R Channel: Ha
G Channel: SII
B Channel: 0.25*Ha - 0.75*SII
The B channel formula is very simple the 0.25*Ha acconts for the Hbeta emission giving to the Hydrogen emission the usual pinkish color, subtracting the SII emission shifts the color hue of the Sulfur rich zones toward yellow the 0.75 factor is an arbitrary value chosen to render the wanted color.

See also on my site

and Astrobin

Tutorials and Processing Examples / Playing with Drizzle
« on: 2015 August 30 02:43:41 »
Hi all!

I've finally managed to try the Drizzling feature of PI and I'm impressed of the result.
It produces a REAL increase in resolution.

Look here.

Not a very good image indeed, but a nice Drizzle example.

I've used it with the default settings.

Image Processing Challenges / Re: Trouble getting blue stars
« on: 2015 August 23 12:16:59 »
Yes, no preview for the white reference: my goal was to use the stars of the whole image as white reference.
at linear stage the nebula is very feeble and do not compromise the white reference.

I usually use a preview when I need to limit the area for the WB for example selecting a spiral galaxy in the field to use as a white reference with star detection disabled.

Ask if you have any other question.

P.S. I forgot to write that as a final step of the I have reduced green noise with SCNR at 0.8

Image Processing Challenges / Re: Trouble getting blue stars
« on: 2015 August 23 09:48:05 »
unfortunately I dont know  Regim's Automatic B-V Color Calibration, so I cannot help you on this issue.

Anyway I've tried PixInsight Color calibration and blue stars seems to be bluer than in your image.

Using Channel combination I've created the RGB image.
I've created a preview in an area without nebula (in the dark region in the top left corner) than I've performed a BackgroundNeutralization with upper limit 0.002.

Then I've applied ColorCalibration with default settings except for the background reference (the same preview of the BackgroundNeutralization) and the upper limit set to 0.002.

You can see the result side by side with your image with the same extremely simple processing (Only HistogramTransformation and Color saturation with curves Masked with a luminance mask).

Announcements / Re: PixInsight 1.8.4 Ripley Released
« on: 2015 August 03 22:14:31 »
Edoardo, I'll have to recompile all my modules for compatibility to the new PCL. Give me a few days, please.

Thank you Carlos for the reply.
No problem!  Please notify us when done.

Announcements / Re: PixInsight 1.8.4 Ripley Released
« on: 2015 August 03 12:59:45 »
Hi All.

Thank you guys for hard work.
All has gone well but now I was instaling Carlos Interchannel curves and I got this error

<* failed *>
*** PCL Win32 System Exception: At address 000007FEFD52B3DD with exception code C0000005 :
Access violation: invalid memory read operation at address 00000000000000D8

I have Window 7



Pages: 1 [2] 3 4 ... 11