PixInsight + Orion StarShoot DScI's + Debayering

[GvnrRickPerry]

Hello all!

This is my first post on the forums, and I must first say - I gave PixInsight a try a while back and couldn't get to it before the trial ran out. I sent an email explaining this to their support team, and they reactivated my license so I could give it a shot! Well, I was able to get a couple things done before that license expired, but a day or two before it expired I started watching Harry's Astroshed videos and examples, and it completely changed me! I went from manually stacking and touching up photos in other software(s) to having an all-in-one application that has more tools than any of the other methods and softwares I was using. I had NO clue what I was missing out on! Well, once my license expired, I went ahead and purchased it and it, so far, has been the wisest purchase I've made for my astronomy endeavours. It's a truly awesome tool, once you know how to use it! It's daunting at first, but WELL worth the effort! Thank YOU PixInsight Team, for creating something so great!

Now, onto my question...

I've been shooting with 3 different cameras; Canon T2i DSLR, Orion StarShoot Deep Space Color Imager Pro v2 (hereafter known as the "Orion V2), and the Orion StarShoot Deep Space Color Imager II (hereafter known as the "Small Orion". My DSLR is becoming a pain (it's not modified, so shooting anything with H-Alpha in it is becoming increasingly difficult) and my Orion V2 shot a puff of electronics smoke out the exhaust port the lat time I plugged it in (it's going back to the manufacturer), so that leaves me with the Small Orion...

It's been a tough learning curve with the Small Orion... It's got a 1/2" chip and I've had problems with guiding (stars being HUGELY bloated), so it presents it's own challenges... Right now, I'm running into an issue in PixInsight where I can't find the correct debayer pattern for this camera! This has me thinking that I'm possibly running the incorrect debayer pattern for my Orion v2 as well...

(MOVED UP FROM BELOW LAST IMAGE:

Is there a different debayer routine that I should be running in PixInsight for my Orion StarShoot CCD cameras? I've read some people mention a "CMYK" debayer method, but I figured I'd just ask here instead of trying a million things... I figured someone has already done this but couldn't find any relevant information in a search.

Thank you so much for any assistance you can provide! I look forward to hearing from y'all!
-Duane)


Here's an example image that I was working with the other day:







As you can see from the image, ALL of the debayer patterns produce these weird pixel patterns in horizontal fashion! I do know now that should be debayering in VNG and not SuperPixel or Bilinear mode  (VNG is the same resolution while the others are a lower resolution), but none of the options are debayering like my image capturing software does... Here's an example from the other application (same zoom factor and reopened in PixInsight after debayer):

[msmythers]

Duane

My suggestion would be to take a daytime picture of something you know what the colors are. Then try the 4 debayer patterns and see which is correct.


Mike

[GvnrRickPerry]

Duane

My suggestion would be to take a daytime picture of something you know what the colors are. Then try the 4 debayer patterns and see which is correct.


Mike

I don't think I can take a picture of something in the daylight with a CCD camera, thank you though...

[Alejandro Tombolini]

Hi Duane, try debayering one uncalibrated image and let me know if it works. Saludos, Alejandro

[NGC7789]

I don't think I can take a picture of something in the daylight with a CCD camera, thank you though...

What exactly is stopping you from taking a daylight picture? Can't you point your telescope at something far enough to focus on? In fact for this test it doesn't really need perfect focus. Or is it some other problem. A daylight picture really is the best way to identify the correct Bayer patter because color issues can be identified easily.

By the way google for the Sony ICX413AQ (the sensor in your camera) identifies the Bayer pattern as RGGB

[pfile]

if you can't make the shutter speed slow enough to capture a daylight image, you can make a mask for the aperture with a pinhole in it. that should help and as a bonus the focus will pretty much be fixed at infinity that way.

the single undebayered image almost looks like a mono fits file, which is weird. but the zoom in your screenshot is not enough to really tell. can you post a raw sub from the camera?

rob

[GvnrRickPerry]

I don't think I can take a picture of something in the daylight with a CCD camera, thank you though...

What exactly is stopping you from taking a daylight picture? Can't you point your telescope at something far enough to focus on? In fact for this test it doesn't really need perfect focus. Or is it some other problem. A daylight picture really is the best way to identify the correct Bayer patter because color issues can be identified easily.

By the way google for the Sony ICX413AQ (the sensor in your camera) identifies the Bayer pattern as RGGB

Hello NGC7789 - I would just rather not chance messing up my sensor. I am sure that I could take a pictre of something (hell, I've taken plenty of t-shirt flats in broad daylight...), I would just rather use that as a last resort. I don't trust myself enough to not mess up this camera (I'm already operating on my last-resort camera) and I just don't want to chance accidently blowing the sensor out if I can do it another way! Plus, since it is a CCD camera, won't the color balance be off anyways? I mean, there's no h-alpha filters on the CCD cameras, like there are on most DSLR'. I just don't want to chance messing my last camera up!

Hi Duane, try debayering one uncalibrated image and let me know if it works. Saludos, Alejandro

Hello Alejandro, I did try debayering (RGGB + VNG method) a RAW sub from the camera and had the same result (pictured at the bottom of this post). I have also attached various files to this post for you to look at, if you'd like.

if you can't make the shutter speed slow enough to capture a daylight image, you can make a mask for the aperture with a pinhole in it. that should help and as a bonus the focus will pretty much be fixed at infinity that way.

the single undebayered image almost looks like a mono fits file, which is weird. but the zoom in your screenshot is not enough to really tell. can you post a raw sub from the camera?

rob

Hello Rob!
RAW Sub has been posted! BTW, this set of images that I posted is of M27 and the final product has turned out really well so far, minus the debayering issues.


Link to Google Drive with RAW Subs:
https://drive.google.com/folderview?id=0B_GY8nm33B8OZjU2WUFnN1p1RU0&usp=sharing



Picture referenced above:

[GvnrRickPerry]

Ok, I guess I've either stumped some people, came across a real issue, or just have no clue, LOL.

Well I have an update. I used another (different) capture software last night... I kind of like it, I've been needing to switch to this software anyways and I'm glad I'm starting the process - the learning curve is pretty big.

Well anyways, I captured an additional 19 X 600second frames last night with this new software... I found it pretty nifty because it embeds the RA and DEC coordinates into the image, which I'm hoping will help with PI's plate solving... But anyways, I tried to debayer these newly captured images and am running into the same problem, with one exception... Now I'm getting prompted in the other capture software when I go to debayer these new images in it... It's asking for the debayer method - I've never seen this screen before, as it has never prompted me. I'm assuming the capture software I *was* using automatically detected the debayer method from the camera and encoded that information into the FITS file, while this new capture software does NOT do that.

I can't get a good looking debayer at all now! I'm going to keep playing with it, but I really don't know enough about how this stuff works to figure it out. Since I'm presented with a new problem with new capture software, I'm assuming the problem is not with the camera itself but with the software(s) I'm using... Problem is, I don't think there's any debayer options I haven't tried! I don't know how to make these images look like they're supposed to!

Any help with this problem would be greatly appreciated!
-Duane

[oldwexi]

Diane!
Which software did you use for calibration and
production of calibration masterbias -flat -dark files?
One of the thoughts i have is maybe you created calibration masterbias -flat -dark files outside of PixInsight.
Lets ensure you got raw uncalibrated lights into PI. And you have created the master file in PI.
?
Gerald

[GvnrRickPerry]

Diane!
Which software did you use for calibration and
production of calibration masterbias -flat -dark files?
One of the thoughts i have is maybe you created calibration masterbias -flat -dark files outside of PixInsight.
Lets ensure you got raw uncalibrated lights into PI. And you have created the master file in PI.
?
Gerald

Hello Gerald, I think you replied to the wrong thread here, but correct me if I'm wrong! I think I was reading another thread where a "Diane" might have been having an issue, which is why I ask!

To answer you questions in regards to my issue - I used PI for everything except debayering, because I'm having the debayering issue. I did run the the Preprocessing steps before debayering the first time, but saw the images weren't looking correct so started messing with them as RAW's to figure out the debayer problems...

Thanks!
-Duane

[oldwexi]

Sorry Duane for the "Diane"...

Could you provide a complete raw light fits file without any processing on it ?

Maybe it is not a debayer issue. Maybe it happens before debayering.

Gerald

[GvnrRickPerry]

Sorry Duane for the "Diane"...

Could you provide a complete raw light fits file without any processing on it ?

Maybe it is not a debayer issue. Maybe it happens before debayering.

Gerald

Hello Gerald!

I have provided a RAW file straight from my camera in the last post with a picture in it from me. Here's the link to the Google Drive with the file:
https://drive.google.com/folderview?id=0B_GY8nm33B8OZjU2WUFnN1p1RU0&usp=sharing

The RAW file begins with "LIGHTS_M27" - the other file there (begins with "recon_LIGHTS_M2") has been debayered using my capture software and appears to be perfect!

As I said in the other post this morning, I used a new capturing software last night and now the old capture software won't debayer the images correctly either! The ONLY thing that would've changed are the FITS keywords and details in the saved RAW files!

Thanks,
-Duane

[oldwexi]

Hi Duane!
Checking the raw gray file with other Software and Pi
shows with any of my sw-packages these artefacts are present.
In Addition:
the undebayered raw has pixel size 8.6 (!!!)  by 8.3 and 752x581 Pixel Image size and
the undebayered raw image has defined CDD-Type 0

The color image has pixel size 8.3. by 8.3 and 777 x 579 pixel image size
the color image has defined CCD Type 2

So, the issue is not in the debayer routine of PI it is in the capturing parameter of your capture software.
When i read in the net about this SW and this camera there are warnings concerning download
speed. To fast download gets artefacts in the images.
May be thats the point ? Dont have this camera and for sure dont use this software,.
so,  i cant give more advice.

Gerald

[GvnrRickPerry]

Hey Gerald,

Thank you so much for taking a look! I guess it could be the cable and/or download speed of the images (I've heard of this being a problem before), but I had assumed since I was able to correctly debayer the image in the N-software (the original capture software I've been using up until last night) that it wasn't a problem with the cable/speed.

Does that make sense? I'll see if I can reiterate it here... Since the n-software I was using was able to capture images and debayer them correctly, I assumed it wasn't a problem with the USB cable or download speed setting.

Also, you are correct, the pixel size is not square, so maybe that's the problem? Does PixInsight have a way to "square" the pixels? I thought this was built into the debayer methods, but maybe I'm wrong?

I can still open that same image (LIGHTS_M27...) in the n-software and debayer it without any issues or any prompts asking for debayer patterns or anything, and it looks exactly like it's supposed to (which is pictured here below on the the bottom-left of the screen capture).
Is there any way that I can figure out what's wrong here?

[pfile]

something is funny here - after debayering the R and G channels still look undebayered in the Y direction. the B channel looks fine.

can you try a different capture software?

rob


edit: whoops, posted late. to be honest i don't think the debayering is correct in the other software either, the image still has a "screen door" look to it.