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 - dhb2206

Pages: [1] 2
1
Bug Reports / Re: Crash on Debayer
« on: 2019 August 27 00:24:38 »
Update: Changed my workflow to move the files to be debayered from one external drive to the destination drive before executing debayer. No more error. I believe this could have something to do with one or t'other drive being in sleep mode in windows 10.

2
Thanks Alejandro, been off-piste for a while. I'll give that a bash!

3
General / Re: PixelMath Help
« on: 2019 February 15 13:58:20 »
And no, that really didn't work..... I'll have a look and see if I can find a pattern to work with.

4
General / Re: PixelMath Help
« on: 2019 February 15 13:53:07 »
Thanks guys, I'd tried all sorts but forgot about square versus round brackets!

5
General / PixelMath Help
« on: 2019 February 15 09:36:04 »
I've noticed in my images that areas with blown out highlights after PCC have a green value that is almost half of that for red and blue, which leaves star core in linear form looking somewhat magenta. So, I was wanting to have a play with PixelMath and see if I can set a ratio for the green channel where red and blue are above a certain threshold, but I'm stumped. I've gotten as far as defining the functions for R and B as $T (admittedly that bit isn't too complex), but I have no clue as to how to refer to individual colour channels to construct my iif statement for G (IIF(myredvalue>0.85,$T/5*8,$T))?

Addendum: I suppose I could utilise a symbol definition and iterate through the individual pixels using myR = pixel($T, x, y, 0), but then we hit iteration - how does one loop through x and y from 0 to with and height respectively, or does the process just do this anyway?

I've no idea if it'll work, but no try, no know!

Cheers,

David


6
Bug Reports / Re: Crash on Debayer
« on: 2019 February 13 10:10:42 »
At least it isn't just me. There is just one machine that this happens on, so I reckon there may be some setting somewhere causing the drop out - more likely to be something Windows than PI I fear.

7
General / Re: Debayered files suddenly deep red
« on: 2019 February 12 10:03:53 »
Try adding Background Neutralisation and Photometric Colour Calibration to your flow and see what that does. My images start off like the hulk, very green!

8
Image Processing Challenges / Re: HDRMT
« on: 2019 February 11 09:56:52 »
This might be stating the bleeding obvious, but you have checked the star in question is in the star mask? Sometimes the big fellas get left out.

Cheers,

David

9
Image Processing Challenges / M42 and M43 Additional Process? #2
« on: 2019 February 11 09:40:40 »
Hmm, first post seems to have developed a fault - probably my attempt at adding a JPEG to it, so I'll give it a slot in dropbox (https://www.dropbox.com/s/mo8b9q25gg4eux3/2019-02-10-101.jpg?dl=0).

As winter is upon us in the northern hemisphere Orion is a decent target. I've been building subs for a couple of months (i.e. about 4 nights as the weather hasn't been playing ball), all with a 300mm lens attached to a Canon DSLR riding on a Skywatcher Star Tracker. There are 282 1 minute light subs at F5.6 and ISO 1600 making up the integrated image, no darks, no flats and no bias. I'd guess about 25% of these are really good quality. Anyway, I've gotten to the end of a post processing run from a drizzled integration of a square 3500 x 3500 preview and would appreciate any tips on where I may have digressed or whether there is more that could be done to improve the image.

My run consisted of:

DBE - default sample generation at a tolerance of 2, samples shunted to avoid stars and deleted where there is obvious nebulosity, reined back to 0.85 for normalised subtraction
BN
PCC
MLT (Multiscale Linear Transform) for noise on layers 1 to 4 - 3 iterations at 0.5 application and thresholds: 4, 3, 2 and 1 masked by an inverse of a lightness abstract with applied STF stretch
Morphological Transformation - Erosion of 1 on a 7 structure applied to a star mask generated from the lightness mask
Histogram Transform on highlights to even out blown out R, G and B values (following a masked stretch and pink stars!)
Masked stretch on default values with a background preview active
MLT for noise stepped back on thresholds to 3, 2, 1, 0.5 masked by lightness extract auto clipped for shadows and highlights
ACDNR on defaults
Histogram Transform to reset black point with minimal clipping
Curves Transform - tiny tweak of RGB for contrast and a reasonable saturation increase
HDRMT (HDR Multiscale Transform) at 9 layers to lightness on a mask revealing just the core of M42

I'm pretty chuffed with the result though the jpeg doesn't do it justice (if I can get it to attach).....

I've stuck the files in a dropbox in case anyone fancies a play, but be aware, the files are large (575MB each)!

Base drizzle image is here: https://www.dropbox.com/s/vrcamdn8pqjwyn3/190210%20Drizzle.zip?dl=0
Post processed image is here: https://www.dropbox.com/s/kfdhzzhoqd9m54s/190210%20Drizzle%20Post%20Plus.zip?dl=0

10
Image Processing Challenges / M42 and M43 Additional Process?
« on: 2019 February 11 08:09:12 »
As winter is upon us in the northern hemisphere Orion is a decent target. I've been building subs for a couple of months (i.e. about 4 nights as the weather hasn't been playing ball), all with a 300mm lens attached to a Canon DSLR riding on a Skywatcher Star Tracker. There are 282 1 minute light subs at F5.6 and ISO 1600 making up the integrated image, no darks, no flats and no bias. I'd guess about 25% of these are really good quality. Anyway, I've gotten to the end of a post processing run from a drizzled integration of a square 3500 x 3500 preview and would appreciate any tips on where I may have digressed or whether there is more that could be done to improve the image.

My run consisted of:

DBE - default sample generation at a tolerance of 2, samples shunted to avoid stars and deleted where there is obvious nebulosity, reigned back to 0.85 for normalised subtraction
BN
PCC
MLT (Multiscale Linear Transform) for noise on layers 1 to 4 - 3 iterations at 0.5 application and thresholds: 4, 3, 2 and 1 masked by an inverse of a lightness abstract with applied STF stretch
Morphological Transformation - Erosion of 1 on a 7 structure applied to a star mask generated from the lightness mask
Histogram Transform on highlights to even out blown out R, G and B values (following a masked stretch and pink stars!)
Masked stretch on default values with a background preview active
MLT for noise stepped back on thresholds to 3, 2, 1, 0.5 masked by lightness extract auto clipped for shadows and highlights
ACDNR on defaults
Histogram Tranform to reset black point with minimal clipping
Curves Transform - tiny tweak of RGB for contrast and a reasonable saturation increase
HDRMT (HDR Mutiscal Transform) at 9 layers to lightness on a mask revealing just the core of M42

I'm pretty chuffed with the result though the jpeg doesn't do it justice (if I can get it to attach).....

I've stuck the files in a dropbox in case anyone fancies a play, but be aware, the files are large (575MB each)!

Base drizzle image is here: https://www.dropbox.com/s/vrcamdn8pqjwyn3/190210%20Drizzle.zip?dl=0
Post processed image is here: https://www.dropbox.com/s/kfdhzzhoqd9m54s/190210%20Drizzle%20Post%20Plus.zip?dl=0

11
Bug Reports / Re: Crash on Debayer
« on: 2019 February 04 13:26:08 »
Huh, typical, ran without a hitch, so I think it may be something to do with my super home pc!

What I have noticed is that I get the process window appear, then you usually get the first 8 files listed as going for processing, when it crashes that doesn't happen - PixInsight just disappears altogether. There are no crash messages, no ghost processes, nothing!

12
Bug Reports / Re: Crash on Debayer
« on: 2019 February 04 13:13:12 »
I suppose a couple of my raw files ought to do the job. Tell you what I will do as well though, I'll have a bash on my laptop as that also has PI installed. If it happens there then I can say it is either something Canon raw or something PI, but if not it is likely to be the config of that machine.

I'll run a test in the next hour or so and post what happened.

Sample files are here: https://www.dropbox.com/sh/h2aq09iuznn2mrq/AAAd1JSREYuT_z6zj7oKW3uoa?dl=0

13
General / Re: DSLR-Raw option format Help Please
« on: 2019 February 04 13:00:34 »
Carl - I was in the same place last year. I'm much more confident now (though still a novice)! Have a search for LightVortexAstronomy. There are some good entry level tutorials on there.

Essentially your raw data comes in a format that Pixinsight can't use very well, so it needs to be converted; this is called debayering (using a one shot colour camera has the data in a colour filter array - a Bayer array, hence de-Bayer). I say "can't use very well", because it works, it is just that the results aren't very good at all. In fact, don't be disheartened when you come out beyond integration and have a green monstrosity staring back at you, this is normal! Mr Bayer again - because the sensor has two green pixels for every blue and red combination (i.e. RGGB), the green dominates somewhat (though I have seen red dominated images too). I'm using canon lenses so skip the calibration and darks and biases as they add nothing that I've found (I'm sure I'll be put right - I was probably just doing the wrong thing). Workflow to make the image pretty runs; debayer, register (star align), integrate, save! Then take a trip through the andromeda tutorial on light vortex, that shows the post process route pretty well.

14
Bug Reports / Crash on Debayer
« on: 2019 February 04 00:32:41 »
Since the new update I've been finding that my first attempt at a debayer will crash PI completely; a wipe out - no error messages, just gone. Second time round it works fine. I'm debayering from a Canon 760d in CR2 format set for RGGB to VNG (I used to get a similar behaviour if I left the pattern dropdown on Auto in the previous version, but just for this camera, the 100d was fine to leave on auto).

Not a prohibitve bug, just slightly irritating when you forget it is coming!

Cheers,

David

15
General / Re: Getting the best from a JPEG export
« on: 2019 February 04 00:25:29 »
I have a feeling this may actually be something to do with my computer's JPEG setup. I'm getting the same with good old daylight shots from Lightroom as well. Anything going to JPEG is darkened from what I'm seeing on the screen (then of course there is monitor calibration). I may be tech savvy, but maybe not savvy enough!  ;)

Pages: [1] 2