Aaaarrrrgggg, Help

General discussion about StarTools.
MadMaxwellSmart
Posts: 20
Joined: Thu Jun 26, 2014 1:12 am

Aaaarrrrgggg, Help

Post by MadMaxwellSmart »

My images of late have been terrible, I read a post about a haze that shows up after the first wipe and auto dev, I have had a terrible time with this and copied a file list of processing to get the haze out and it did not work. I understand that other peoples processing might not work for the image I've got but, it was worth a try. I need some definite help, and I'm willing to upload my fts files for some to look at and play with, but I do not know how to upload to this forum! I need to ask another question, is there any type of info on the different channels associated with star tools,(ie) what does decon mean and what exactly does it do? HDR? and the rest, I'm still going off a work sheet that a good friend gave me and it seems not to be working as good as it used to, I just can't get the purple and white haze out of my pictures. When I take a wide field shot, I have a purple haze that seems to originate from the top middle of the image and it comes down midway into the image, very annoying.....any how, thanks in advance....Glenn
User avatar
admin
Site Admin
Posts: 3367
Joined: Thu Dec 02, 2010 10:51 pm
Location: Melbourne
Contact:

Re: Aaaarrrrgggg, Help

Post by admin »

Hey Glenn,

Upload your data to Google Drive or Dropbox and share the link with us. I'm more than happy to have a look!

It sounds like you might still have stacking artifacts in your image. If so, crop the image until they are no longer in the data. Wipe - by design - backs off when it sees pixels that are darker than the real interstellar background (such as those caused by stacking artifacts, dust donuts or dead pixels). It does this because it does not want to clip your data; it cannot be sure that those 'dark anomalies' are not the real interstellar background. Alternatively you can use a gap in the mask to make Wipe ignore the offending pixels.

Deconvolution ('Decon') - in StarTools - is a mathematical operation that attempts to undo the blurring that atmospheric turbulence (aka 'bad seeing') causes in your images. It's not really a sharpening tool, but an 'unblurring' tool. Deconvolution is traditionally a very tricky procedure to use, but in StarTools it is made somewhat easier through StarTools' ability to automatically create masks for your stars (deconvolution cannot operate near overexposed areas such as star cores), as well as StarTools' ability to figure out how far it can go in the face of noise in your data (traditional deconvolution is very, very sensitive to noise and quickly falls apart in the presence of much noise).

Looking forward to the data!
Ivo Jager
StarTools creator and astronomy enthusiast
MadMaxwellSmart
Posts: 20
Joined: Thu Jun 26, 2014 1:12 am

Re: Aaaarrrrgggg, Help

Post by MadMaxwellSmart »

admin wrote:Hey Glenn,

Upload your data to Google Drive or Dropbox and share the link with us. I'm more than happy to have a look!

It sounds like you might still have stacking artifacts in your image. If so, crop the image until they are no longer in the data. Wipe - by design - backs off when it sees pixels that are darker than the real interstellar background (such as those caused by stacking artifacts, dust donuts or dead pixels). It does this because it does not want to clip your data; it cannot be sure that those 'dark anomalies' are not the real interstellar background. Alternatively you can use a gap in the mask to make Wipe ignore the offending pixels.

Deconvolution ('Decon') - in StarTools - is a mathematical operation that attempts to undo the blurring that atmospheric turbulence (aka 'bad seeing') causes in your images. It's not really a sharpening tool, but an 'unblurring' tool. Deconvolution is traditionally a very tricky procedure to use, but in StarTools it is made somewhat easier through StarTools' ability to automatically create masks for your stars (deconvolution cannot operate near overexposed areas such as star cores), as well as StarTools' ability to figure out how far it can go in the face of noise in your data (traditional deconvolution is very, very sensitive to noise and quickly falls apart in the presence of much noise).

Looking forward to the data!
Hey Ivo, Thanks for the quick response, I may have to go to work very early in the morning, I will up load the link as soon as I get more time, they will be the fts. files...Thanks again my friend, looking forward to seeing what you can do with them....Glenn
ChrisLX200
Posts: 90
Joined: Tue Apr 22, 2014 10:25 am
Location: Macclesfield, UK
Contact:

Re: Aaaarrrrgggg, Help

Post by ChrisLX200 »

Hi Glenn, in case you missed it there is a PDF StarTools Manual (unofficial but still an excellent resource). Well worth a read :)
http://www.startools.org/download/StarT ... ficial.pdf

ChrisH
MadMaxwellSmart
Posts: 20
Joined: Thu Jun 26, 2014 1:12 am

Re: Aaaarrrrgggg, Help

Post by MadMaxwellSmart »

I hope this works, going to try and copy the link to FTS file https://www.dropbox.com/s/comwyqx0hutui ... e.fts?dl=0 Let me know if this don't work. The file should be a wide FOV of the Orion region, M42 running man, horsehead and flame.
MadMaxwellSmart
Posts: 20
Joined: Thu Jun 26, 2014 1:12 am

Re: Aaaarrrrgggg, Help

Post by MadMaxwellSmart »

ChrisLX200 wrote:Hi Glenn, in case you missed it there is a PDF StarTools Manual (unofficial but still an excellent resource). Well worth a read :)
http://www.startools.org/download/StarT ... ficial.pdf

ChrisH
Hey Chris, I downloaded that manuel, Thank you, I played around a bit with that same FTS file I uploaded, I made a little progress but, nothing I'd be willing to put out for others to see. I used the Develop slider and went to about 90%, then I used the Gamma to tone down the purple haze that is drifting in from the top edge. I hope you see that if you play with the file. And yes, I did crop to make sure the artifacts were gone. But back to the purple and white haze, I did manage to get rid of those, but in turn it seemed that I dumbed the image down, lost the view of the horse head in the end....The image was taken with a Canon 550D with a 250mm lens, 2 Lights@30sec., 15 Lights@240sec., and 2 Darks@30sec., and 8 Darks@240sec., I added some Bias about 15 I think...I was told that they did help any, but they didn't hurt either. all was done at 800iso.
ChrisLX200
Posts: 90
Joined: Tue Apr 22, 2014 10:25 am
Location: Macclesfield, UK
Contact:

Re: Aaaarrrrgggg, Help

Post by ChrisLX200 »

I'm afraid it's going to take someone smarter than me to resolve some of the problems in that image - so I'll let Ivo show how it's done :)
Meanwhile my pathetic attempt is below... (hey, at least you can see the HorseHead now!)

Image


ChrisH
MadMaxwellSmart
Posts: 20
Joined: Thu Jun 26, 2014 1:12 am

Re: Aaaarrrrgggg, Help

Post by MadMaxwellSmart »

Chris, here is the result I got
M42.jpg
M42.jpg (227.93 KiB) Viewed 13162 times
Did you see the Purple haze in the top while processing? I'll see what Ivo can do with it, Ivo, when you look at this, can you work up a file list especially when you get rid of the haze? Thanks
User avatar
admin
Site Admin
Posts: 3367
Joined: Thu Dec 02, 2010 10:51 pm
Location: Melbourne
Contact:

Re: Aaaarrrrgggg, Help

Post by admin »

Glenn,

You've done a good job with the data at hand.

There are a number of issues with the data, with the major problem being the severe vignetting and very uneven illumination. You need to take flats - they are not optional!

Another issue is that the data seems to have been meddled with beforehand - it has, at the very least, been color balanced (try turning off al channel calibration/alignment in DSS!) and it appears you've integrated multiple exposure lengths which further complicates things; your data is far from linear now, with noise and colors varying all over the place. The (presumably) primary reason for integrating multiple exposure lengths hasn't borne fruit either with M42's core still overexposed. If you really must integrate multiple exposure lengths, do it in StarTools and use only two exposure lengths. Using multiple exposure lengths should be unnecessary for all but the brightest objects (really only M42 comes to mind).

As a matter of fact, I did not see any purple haze at the top, but there definitely was some remnant gradient due to very uneven illumination and (what I presume is) some dust on the sensor. Did you crop the image to get rid of any stacking artifacts, as suggested?
Ivo Jager
StarTools creator and astronomy enthusiast
MadMaxwellSmart
Posts: 20
Joined: Thu Jun 26, 2014 1:12 am

Re: Aaaarrrrgggg, Help

Post by MadMaxwellSmart »

admin wrote:Glenn,

You've done a good job with the data at hand.

There are a number of issues with the data, with the major problem being the severe vignetting and very uneven illumination. You need to take flats - they are not optional!

Another issue is that the data seems to have been meddled with beforehand - it has, at the very least, been color balanced (try turning off al channel calibration/alignment in DSS!) and it appears you've integrated multiple exposure lengths which further complicates things; your data is far from linear now, with noise and colors varying all over the place. The (presumably) primary reason for integrating multiple exposure lengths hasn't borne fruit either with M42's core still overexposed. If you really must integrate multiple exposure lengths, do it in StarTools and use only two exposure lengths. Using multiple exposure lengths should be unnecessary for all but the brightest objects (really only M42 comes to mind).

As a matter of fact, I did not see any purple haze at the top, but there definitely was some remnant gradient due to very uneven illumination and (what I presume is) some dust on the sensor. Did you crop the image to get rid of any stacking artifacts, as suggested?
Ivo, Thanks for the info I cannot find the tab that allows me to turn off the (al channel calibration/alignment in DSS) I've looked, could you direct me in that direction. And as far as Flats go, I was told that taking images with just the Camera and a 250mm lens, I would not need Flats or Bias, but I add the bias anyhow, I tried taking Flats a few months ago and wound up with a horrible white haze over the entire image. So now I'm really confused, ( but having fun while learning). So to sum this up, when I take wide FOV shots with the camera piggybacked, I take Darks, Flats and Bias, Basically the whole nine yards? Is there a rule I need to go by when taking Flats for a Wide shot....I really appreciate your help in this matter.....and yes I did crop as much as I could.
Last edited by MadMaxwellSmart on Fri Dec 19, 2014 6:45 pm, edited 1 time in total.
Post Reply