Yet another flats question...

Discuss how you are able to get those fantastic images!!!
User avatar
Larry 1969 United States of America
Milky Way Ambassador
Articles: 0
Offline
Posts: 1481
Joined: Fri May 17, 2019 12:33 am
4
Location: Pittsburgh PA USA
Status:
Offline

TSS Photo of the Day

Yet another flats question...

#1

Post by Larry 1969 »


Sorry if I seem to be asking the same question over and over again...
I'm going to try to be very specific and detailed about the situation.
I had a surprise clear night and a new moon! A perfect opportunity to redo my Horsehead and Flame that I did previously under a full moon and suffered from terrible gradients and weird vignetting that I couldn't seem to overcome. I managed to get 88 x 3 min frames before my target fell behind the trees.

I shot 50 flats at 30,000 ADU with my Pegasus flat panel. I did not use a Tee shirt or paper to diffuse the light because the flat panel isn't bright enough. I shot 50 matching dark flats as well. I stacked the images with APP using default settings except I unchecked "adaptive pedestal / reduce amp glow" and "neutralize background" because that is the preferred settings for post-processing with StarTools and came up with this:
Screenshot HH and Flame original.png
Not only was the vignetting uncorrected, but I also lost substantial detail in the image...

So I shot 50 new flats at 20,000 ADU and matching dark flats. I combined them with APP using the same settings and got this:
Screenshot HH and Flame new flats.png
Much better but the vignetting is still out of hand...

I stacked the 88 light frames with no calibration frames with the same settings and got this:
Screenshot HH and Flame no cal.png
Any ideas what I'm doing wrong?

Larry
For visual:
10" Skywatcher collapsible goto dob, various EP's and a Celestron StarSense auto align.

For imaging:
Orion 8" astrograph 800mm @ F3.9
Eq6-R Pro controlled by APT via EQmod with an OTA mounted mini PC
Tele Vue Paracorr Type 2 coma corrector
Altair Hypercam 26C
Image
User avatar
Larry 1969 United States of America
Milky Way Ambassador
Articles: 0
Offline
Posts: 1481
Joined: Fri May 17, 2019 12:33 am
4
Location: Pittsburgh PA USA
Status:
Offline

TSS Photo of the Day

Re: Yet another flats question...

#2

Post by Larry 1969 »


I forgot to include my master flats...

Here's the 30,000 ADU master:
Screenshot 30,000 ADU flat.png
And here's the 20,000 ADU master:
Screenshot 20,000 ADU flat.png
I don't see much of a difference...

Larry
For visual:
10" Skywatcher collapsible goto dob, various EP's and a Celestron StarSense auto align.

For imaging:
Orion 8" astrograph 800mm @ F3.9
Eq6-R Pro controlled by APT via EQmod with an OTA mounted mini PC
Tele Vue Paracorr Type 2 coma corrector
Altair Hypercam 26C
Image
User avatar
bobharmony
Local Group Ambassador
Articles: 0
Offline
Posts: 2029
Joined: Sun May 12, 2019 1:11 pm
4
Location: Connecticut, US
Status:
Offline

TSS Photo of the Day

Re: Yet another flats question...

#3

Post by bobharmony »


Interesting, Larry. I am beginning to think that flats are more of an art than science. 30K seem like the proper ADU fro the 26C from everything I have read, but the 20K flats are doing a better job of correcting the overall image and the dust motes. I have been having issues with some M78 data I shot last week. At first I thought I had a wicked gradient with my usual 50% flats, but on closer inspection I came to the conclusion (not proven yet) that the flats are overcorrecting. I need to shoot some lower exposure flats and see if that helps. I am also considering replacing the dark flats with bias frames to see if that helps at all.

Does the Hypercam 26C pick up Ha data on the sensor? I couldn't really tell from the specs I looked at. I am getting the suspicion that StarTools doesn't do well with Ha data. Did you try processing either of these through StarTools to see if it would correct the vignetting?

Bob
Hardware: Celestron C6-N w/ Advanced GTmount, Baader MK iii CC, Orion ST-80, Canon 60D (unmodded), Nikon D5300 (modded), Orion SSAG
Software: BYE, APT, PHD2, DSS, PhotoShop CC 2020, StarTools, Cartes du Ciel, AstroTortilla

Image
User avatar
SkyHiker United States of America
Local Group Ambassador
Articles: 0
Online
Posts: 2293
Joined: Sat May 11, 2019 8:40 pm
4
Location: Santa Barbara, CA
Status:
Online

TSS Awards Badges

TSS Photo of the Day

Re: Yet another flats question...

#4

Post by SkyHiker »


I'm having the same problems. Two nights lost due to excessive gradients from LP and poorly correcting flats. I took some one night at 30K, against a wall, against the sky, per filter or just LPR, with light shield or without. I see big differences in the flats between 20k at gain 140 and 30k at gain 210 but not between flats against the morning sky or at nigh against a white wall. I haven't pinned it down yet. Tonight I'll try turning the cooler on but frankly it's cold here so I don't expect miracles.
... Henk. :D Telescopes: GSO 12" Astrograph, "Comet Hunter" MN152, ES ED127CF, ES ED80, WO Redcat51, Z12, AT6RC, Celestron Skymaster 20x80, Mounts and tripod: Losmandy G11S with OnStep, AVX, Tiltall, Cameras: ASI2600MC, ASI2600MM, ASI120 mini, Fuji X-a1, Canon XSi, T6, ELPH 100HS, DIY: OnStep controller, Pi4b/power rig, Afocal adapter, Foldable Dob base, Az/Alt Dob setting circles, Accessories: ZWO 36 mm filter wheel, TV Paracorr 2, Baader MPCC Mk III, ES FF, SSAG, QHY OAG-M, EAF electronic focuser, Plossls, Barlows, Telrad, Laser collimators (Seben LK1, Z12, Howie Glatter), Cheshire, 2 Orion RACIs 8x50, Software: KStars-Ekos, DSS, PHD2, Nebulosity, Photo Gallery, Gimp, CHDK, Computers:Pi4b, 2x running KStars/Ekos, Toshiba Satellite 17", Website:Henk's astro images
User avatar
Larry 1969 United States of America
Milky Way Ambassador
Articles: 0
Offline
Posts: 1481
Joined: Fri May 17, 2019 12:33 am
4
Location: Pittsburgh PA USA
Status:
Offline

TSS Photo of the Day

Re: Yet another flats question...

#5

Post by Larry 1969 »


bobharmony wrote: Fri Feb 04, 2022 4:36 am Interesting, Larry. I am beginning to think that flats are more of an art than science. 30K seem like the proper ADU fro the 26C from everything I have read, but the 20K flats are doing a better job of correcting the overall image and the dust motes. I have been having issues with some M78 data I shot last week. At first I thought I had a wicked gradient with my usual 50% flats, but on closer inspection I came to the conclusion (not proven yet) that the flats are overcorrecting. I need to shoot some lower exposure flats and see if that helps. I am also considering replacing the dark flats with bias frames to see if that helps at all.

Does the Hypercam 26C pick up Ha data on the sensor? I couldn't really tell from the specs I looked at. I am getting the suspicion that StarTools doesn't do well with Ha data. Did you try processing either of these through StarTools to see if it would correct the vignetting?

Bob
I too struggle with M78. I've tried it over the past couple of years with 2 different cameras and cannot seem to cope with the gradient. Maybe there is something in that part of the sky that messes with me? I'm pretty much over it... I'll pick a different target...

The Hypercam 26C has an IR/UV blocking sensor window so I'm not sure how it handles Ha data. StarTools has a pre-set in the "wipe" module for vignetting, but I can't seem to get along with it... It just seems to mess up the entire image. I need to look at some tutorials...

I'm going to fire up my old led tracing panel and shoot some old-school flats tonight to see if there is any difference and, if there is, there will be a Pegasus flat panel up for sale shortly! :lol:

Thanks for the reply Bob!

Larry
For visual:
10" Skywatcher collapsible goto dob, various EP's and a Celestron StarSense auto align.

For imaging:
Orion 8" astrograph 800mm @ F3.9
Eq6-R Pro controlled by APT via EQmod with an OTA mounted mini PC
Tele Vue Paracorr Type 2 coma corrector
Altair Hypercam 26C
Image
User avatar
Larry 1969 United States of America
Milky Way Ambassador
Articles: 0
Offline
Posts: 1481
Joined: Fri May 17, 2019 12:33 am
4
Location: Pittsburgh PA USA
Status:
Offline

TSS Photo of the Day

Re: Yet another flats question...

#6

Post by Larry 1969 »


SkyHiker wrote: Fri Feb 04, 2022 3:20 pm I'm having the same problems. Two nights lost due to excessive gradients from LP and poorly correcting flats. I took some one night at 30K, against a wall, against the sky, per filter or just LPR, with light shield or without. I see big differences in the flats between 20k at gain 140 and 30k at gain 210 but not between flats against the morning sky or at nigh against a white wall. I haven't pinned it down yet. Tonight I'll try turning the cooler on but frankly it's cold here so I don't expect miracles.
I've been following that post with interest. Our problems do seem similar but I didn't want to hijack your thread.

Larry
For visual:
10" Skywatcher collapsible goto dob, various EP's and a Celestron StarSense auto align.

For imaging:
Orion 8" astrograph 800mm @ F3.9
Eq6-R Pro controlled by APT via EQmod with an OTA mounted mini PC
Tele Vue Paracorr Type 2 coma corrector
Altair Hypercam 26C
Image
User avatar
Larry 1969 United States of America
Milky Way Ambassador
Articles: 0
Offline
Posts: 1481
Joined: Fri May 17, 2019 12:33 am
4
Location: Pittsburgh PA USA
Status:
Offline

TSS Photo of the Day

Re: Yet another flats question...

#7

Post by Larry 1969 »


So, I pulled out the old LED tracing panel and 3 sheets of paper and shot 20,000 ADU flats.
I combined them in APP using the same settings and I think the result was slightly better. Still some overcorrection and gradient...
Screenshot (74).png
I processed it in StarTools and did some final sharpening and cropping with Topaz DeNoise and came up with this:
HH_and_Flame_LED_flats-RGB-session_1-DeNoiseAI-low-light.jpg
Still not really happy but...

Larry
For visual:
10" Skywatcher collapsible goto dob, various EP's and a Celestron StarSense auto align.

For imaging:
Orion 8" astrograph 800mm @ F3.9
Eq6-R Pro controlled by APT via EQmod with an OTA mounted mini PC
Tele Vue Paracorr Type 2 coma corrector
Altair Hypercam 26C
Image
User avatar
bobharmony
Local Group Ambassador
Articles: 0
Offline
Posts: 2029
Joined: Sun May 12, 2019 1:11 pm
4
Location: Connecticut, US
Status:
Offline

TSS Photo of the Day

Re: Yet another flats question...

#8

Post by bobharmony »


Not bad, Larry. A little light crop and no one would know the difference. You have done a really nice job of controlling Alnitak as well.

Technically, doing lower ADU flats shouldn't make a difference as the purpose of creating flats is to create a ratio of the brightest pixels and the dimmest and that ratio shouldn't change based on exposure length (or number of pieces of paper, layers of t-shirt, etc.), but we know this isn't the case and I am getting ready to take dimmer flats myself. I am also looking into adding a fixed amount to each pixel, which logically should make a difference in the ratio.

I guess I should start a thread with my pains as well so you and Henk don't think you're the only ones suffering from bad flats disease!

Bob
Hardware: Celestron C6-N w/ Advanced GTmount, Baader MK iii CC, Orion ST-80, Canon 60D (unmodded), Nikon D5300 (modded), Orion SSAG
Software: BYE, APT, PHD2, DSS, PhotoShop CC 2020, StarTools, Cartes du Ciel, AstroTortilla

Image
User avatar
Larry 1969 United States of America
Milky Way Ambassador
Articles: 0
Offline
Posts: 1481
Joined: Fri May 17, 2019 12:33 am
4
Location: Pittsburgh PA USA
Status:
Offline

TSS Photo of the Day

Re: Yet another flats question...

#9

Post by Larry 1969 »


bobharmony wrote: Sat Feb 05, 2022 6:57 pm Not bad, Larry. A little light crop and no one would know the difference. You have done a really nice job of controlling Alnitak as well.

Technically, doing lower ADU flats shouldn't make a difference as the purpose of creating flats is to create a ratio of the brightest pixels and the dimmest and that ratio shouldn't change based on exposure length (or number of pieces of paper, layers of t-shirt, etc.), but we know this isn't the case and I am getting ready to take dimmer flats myself. I am also looking into adding a fixed amount to each pixel, which logically should make a difference in the ratio.

I guess I should start a thread with my pains as well so you and Henk don't think you're the only ones suffering from bad flats disease!

Bob
Thanks Bob!
If you start a thread I will be following...

Larry
For visual:
10" Skywatcher collapsible goto dob, various EP's and a Celestron StarSense auto align.

For imaging:
Orion 8" astrograph 800mm @ F3.9
Eq6-R Pro controlled by APT via EQmod with an OTA mounted mini PC
Tele Vue Paracorr Type 2 coma corrector
Altair Hypercam 26C
Image
User avatar
bobharmony
Local Group Ambassador
Articles: 0
Offline
Posts: 2029
Joined: Sun May 12, 2019 1:11 pm
4
Location: Connecticut, US
Status:
Offline

TSS Photo of the Day

Re: Yet another flats question...

#10

Post by bobharmony »


I will mention here that I am doing a new set of flats now using an extra two sheets of paper in front of the LED panel. The ratio of brightest to lowest pixel for the new dimmer flats is about 1.7:1. The previous brighter flats had a ratio of 1.9:1. After thinking about this more the pedestal value of the camera would impact this ratio, so I withdraw my earlier statement that dimmer flats wouldn't be any different than brighter flats. In short I am hoping the the dimmer flats will reduce the overcorrection in a useful way.

The pedestal value is the minimum value set by the camera for each pixel at the beginning of each exposure. It protects against any random change in the pixel value that drops the value to 0, which would be a bad thing mathematically speaking. AFAIK the value is set by the camera firmware and cannot be modified by the user. It is also the reason dark flats or bias frames are used to calibrate the flats, one of the things they do is to reduce the pedestal value from the final stacked image.

Bob
Hardware: Celestron C6-N w/ Advanced GTmount, Baader MK iii CC, Orion ST-80, Canon 60D (unmodded), Nikon D5300 (modded), Orion SSAG
Software: BYE, APT, PHD2, DSS, PhotoShop CC 2020, StarTools, Cartes du Ciel, AstroTortilla

Image
User avatar
SkyHiker United States of America
Local Group Ambassador
Articles: 0
Online
Posts: 2293
Joined: Sat May 11, 2019 8:40 pm
4
Location: Santa Barbara, CA
Status:
Online

TSS Awards Badges

TSS Photo of the Day

Re: Yet another flats question...

#11

Post by SkyHiker »


I took several sets of flats of exposure times 8,4,2,1,0,5,0.25 seconds at gain 210 and offset 24 of the white wall (ADUs went from saturated 16 bit at 8 sec to rather thin at 0.25) and after stacking with the individual sets of flats, the results were near-identical except perhaps with the 8 seconds one that got saturated. The results were not as bad as before but still not good. Maybe using a nonzero bias makes a difference?

So, it looks like the ADU level does not matter a whole lot. At an earlier occasion I took flats against the morning sky with all of the LRGB and the results across filters were near-identical modulo a constant corresponding with the filter. I did that by loading them up in Scilab and looking at the flats numerically, plotting them in 3D, dividing them by one another to see if their ratio is approximately constant across the sensor (yes). Taking flats against the sky or against my white wall also seems to result in near identical flats.

The RGB channels all have the same value for a given master flat from DSS. I plan to stack them in Scilab using the DSS method, applying them to lights and see if I get the same as in DSS. Unfortunately, Scilab loads jpg images in 8 bits so I need some bright subs but given the amount of LP that may not be too much of a problem. Mayb APLab can help but the ZWO cameras are not supported and adding a new one takes some work, while I'm not sure if I need it.
... Henk. :D Telescopes: GSO 12" Astrograph, "Comet Hunter" MN152, ES ED127CF, ES ED80, WO Redcat51, Z12, AT6RC, Celestron Skymaster 20x80, Mounts and tripod: Losmandy G11S with OnStep, AVX, Tiltall, Cameras: ASI2600MC, ASI2600MM, ASI120 mini, Fuji X-a1, Canon XSi, T6, ELPH 100HS, DIY: OnStep controller, Pi4b/power rig, Afocal adapter, Foldable Dob base, Az/Alt Dob setting circles, Accessories: ZWO 36 mm filter wheel, TV Paracorr 2, Baader MPCC Mk III, ES FF, SSAG, QHY OAG-M, EAF electronic focuser, Plossls, Barlows, Telrad, Laser collimators (Seben LK1, Z12, Howie Glatter), Cheshire, 2 Orion RACIs 8x50, Software: KStars-Ekos, DSS, PHD2, Nebulosity, Photo Gallery, Gimp, CHDK, Computers:Pi4b, 2x running KStars/Ekos, Toshiba Satellite 17", Website:Henk's astro images
User avatar
Larry 1969 United States of America
Milky Way Ambassador
Articles: 0
Offline
Posts: 1481
Joined: Fri May 17, 2019 12:33 am
4
Location: Pittsburgh PA USA
Status:
Offline

TSS Photo of the Day

Re: Yet another flats question...

#12

Post by Larry 1969 »


SkyHiker wrote: Sat Feb 05, 2022 9:59 pm The RGB channels all have the same value for a given master flat from DSS. I plan to stack them in Scilab using the DSS method, applying them to lights and see if I get the same as in DSS. Unfortunately, Scilab loads jpg images in 8 bits so I need some bright subs but given the amount of LP that may not be too much of a problem. Mayb APLab can help but the ZWO cameras are not supported and adding a new one takes some work, while I'm not sure if I need it.
Maybe download the trial of APP and test that against DSS? I seem to have the issue with both programs though...

Larry
For visual:
10" Skywatcher collapsible goto dob, various EP's and a Celestron StarSense auto align.

For imaging:
Orion 8" astrograph 800mm @ F3.9
Eq6-R Pro controlled by APT via EQmod with an OTA mounted mini PC
Tele Vue Paracorr Type 2 coma corrector
Altair Hypercam 26C
Image
User avatar
bobharmony
Local Group Ambassador
Articles: 0
Offline
Posts: 2029
Joined: Sun May 12, 2019 1:11 pm
4
Location: Connecticut, US
Status:
Offline

TSS Photo of the Day

Re: Yet another flats question...

#13

Post by bobharmony »


I have stacked with DSS and ASTAP and both have the issue. I checked the ASTAP flats in Photoshop and they are all consistent with the same values in RGB.

I tried a couple of things today. I ran a dimmer set of flats with the median value close to 2000 instead of my usual 8K. The result showed a slightly larger dark circle in the center of the image but the corners were just as bright as the stack with the higher median value.

I also tried a stack using bias frames instead of flat darks to calibrate the flats. It showed no improvement over the flat dark version.

Still hunting for the magic setting!

Bob
Hardware: Celestron C6-N w/ Advanced GTmount, Baader MK iii CC, Orion ST-80, Canon 60D (unmodded), Nikon D5300 (modded), Orion SSAG
Software: BYE, APT, PHD2, DSS, PhotoShop CC 2020, StarTools, Cartes du Ciel, AstroTortilla

Image
User avatar
starfield United States of America
Milky Way Ambassador
Articles: 0
Offline
Posts: 1022
Joined: Fri Aug 30, 2019 3:22 pm
4
Location: San Diego
Status:
Offline

TSS Awards Badges

TSS Photo of the Day

AP Processing Challenge

Re: Yet another flats question...

#14

Post by starfield »


I've been chasing some weird flat problems that only appear when I shoot from my light polluted back yard. My green flat tends to overcorrect and red one under corrects. I'm beginning to wonder if I'm not getting some type of reflection from the inside of my scopes dew shield. Why I only have this problem when I shoot from my LP backyard really has me stumped. My flats I shoot out in the desert (Bortle 3) are always spot on.
Scopes: Esprit 100, 12.5" Telekit Dob
Camera: 294 MC Pro, 224 mc, 2600mm
Guiding: ZWO 290 mini on 120mm guide scope
Mounts: EQ6R-Pro, EQ Platform.
Filters: Optolong L-Pro & L-Enhance, Chroma 36mm LRGB, 5nm HA, 3nm OIII, 3nm SII.
Software: SharpCap, SGP, StarTools 1.7, Photoshop, Pixinsight
The Death Star Observatory
Astro Photos:https://www.astrobin.com/users/starfield/
User avatar
SkyHiker United States of America
Local Group Ambassador
Articles: 0
Online
Posts: 2293
Joined: Sat May 11, 2019 8:40 pm
4
Location: Santa Barbara, CA
Status:
Online

TSS Awards Badges

TSS Photo of the Day

Re: Yet another flats question...

#15

Post by SkyHiker »


I plotted the master flat and bias, a single sub, the manually corrected sub, and the DSS-calibrated one in Scilab in 3D and included them in a message to the DSS IO board. There is a significant difference between the DSS-calibrated one and the manual one, which reflects the DSS documented math according to the help. See https://groups.io/g/DeepSkyStacker/message/27765 , I don't know if you need to sign up to have access.
... Henk. :D Telescopes: GSO 12" Astrograph, "Comet Hunter" MN152, ES ED127CF, ES ED80, WO Redcat51, Z12, AT6RC, Celestron Skymaster 20x80, Mounts and tripod: Losmandy G11S with OnStep, AVX, Tiltall, Cameras: ASI2600MC, ASI2600MM, ASI120 mini, Fuji X-a1, Canon XSi, T6, ELPH 100HS, DIY: OnStep controller, Pi4b/power rig, Afocal adapter, Foldable Dob base, Az/Alt Dob setting circles, Accessories: ZWO 36 mm filter wheel, TV Paracorr 2, Baader MPCC Mk III, ES FF, SSAG, QHY OAG-M, EAF electronic focuser, Plossls, Barlows, Telrad, Laser collimators (Seben LK1, Z12, Howie Glatter), Cheshire, 2 Orion RACIs 8x50, Software: KStars-Ekos, DSS, PHD2, Nebulosity, Photo Gallery, Gimp, CHDK, Computers:Pi4b, 2x running KStars/Ekos, Toshiba Satellite 17", Website:Henk's astro images
User avatar
bobharmony
Local Group Ambassador
Articles: 0
Offline
Posts: 2029
Joined: Sun May 12, 2019 1:11 pm
4
Location: Connecticut, US
Status:
Offline

TSS Photo of the Day

Re: Yet another flats question...

#16

Post by bobharmony »


SkyHiker wrote: Sun Feb 06, 2022 7:32 am I plotted the master flat and bias, a single sub, the manually corrected sub, and the DSS-calibrated one in Scilab in 3D and included them in a message to the DSS IO board. There is a significant difference between the DSS-calibrated one and the manual one, which reflects the DSS documented math according to the help. See https://groups.io/g/DeepSkyStacker/message/27765 , I don't know if you need to sign up to have access.
Those graphs are a very nice way to visualize what is happening, Henk. I am going to study them some more to see what else can be gleaned from them. BTW, I was able to view the group post (and one reply thus far) without a sign-in to the DSS group. I may also have to look into setting up Scilab as a tool.

Bob
Hardware: Celestron C6-N w/ Advanced GTmount, Baader MK iii CC, Orion ST-80, Canon 60D (unmodded), Nikon D5300 (modded), Orion SSAG
Software: BYE, APT, PHD2, DSS, PhotoShop CC 2020, StarTools, Cartes du Ciel, AstroTortilla

Image
User avatar
Larry 1969 United States of America
Milky Way Ambassador
Articles: 0
Offline
Posts: 1481
Joined: Fri May 17, 2019 12:33 am
4
Location: Pittsburgh PA USA
Status:
Offline

TSS Photo of the Day

Re: Yet another flats question...

#17

Post by Larry 1969 »


SkyHiker wrote: Sun Feb 06, 2022 7:32 am I plotted the master flat and bias, a single sub, the manually corrected sub, and the DSS-calibrated one in Scilab in 3D and included them in a message to the DSS IO board. There is a significant difference between the DSS-calibrated one and the manual one, which reflects the DSS documented math according to the help. See https://groups.io/g/DeepSkyStacker/message/27765 , I don't know if you need to sign up to have access.
I was also able view the post. Great job with those graphs.

Larry
For visual:
10" Skywatcher collapsible goto dob, various EP's and a Celestron StarSense auto align.

For imaging:
Orion 8" astrograph 800mm @ F3.9
Eq6-R Pro controlled by APT via EQmod with an OTA mounted mini PC
Tele Vue Paracorr Type 2 coma corrector
Altair Hypercam 26C
Image
User avatar
Larry 1969 United States of America
Milky Way Ambassador
Articles: 0
Offline
Posts: 1481
Joined: Fri May 17, 2019 12:33 am
4
Location: Pittsburgh PA USA
Status:
Offline

TSS Photo of the Day

Re: Yet another flats question...

#18

Post by Larry 1969 »


There is a setting in APP under the "Calibrate" tab for master flat integration. "Automatic" is the default. The other choices are "average", "median" and "maximum". According to the help, "automatic" will choose between "average" and "median" depending on the number of flats. Less than 20 it selects "median" and more than 20 gets "average". "Maximum" says Use maximum to show all outliers / artifacts (with too high ADU values) in the data like hot pixels, cosmic rays etc.

I restacked my image using "maximum" and it made no visible difference...

I'll keep fooling around I suppose...

Larry
For visual:
10" Skywatcher collapsible goto dob, various EP's and a Celestron StarSense auto align.

For imaging:
Orion 8" astrograph 800mm @ F3.9
Eq6-R Pro controlled by APT via EQmod with an OTA mounted mini PC
Tele Vue Paracorr Type 2 coma corrector
Altair Hypercam 26C
Image
User avatar
Gordon United States of America
Site Admin
Site Admin
Articles: 1044
Offline
Posts: 8366
Joined: Wed Apr 24, 2019 10:52 pm
4
Location: Cottonwood, AZ
Status:
Offline

TSS Awards Badges

TSS Photo of the Day

I Broke The Forum.

TSS EAA Messier awards

Re: Yet another flats question...

#19

Post by Gordon »


I've had pretty good results using the "Flats Wizard" in N.I.N.A. You still need to create settings but it really helps dial it in.
Gordon
Scopes: Explore Scientific ED80CF, Skywatcher 200 Quattro Imaging Newt, SeeStar S50 for EAA.
Mounts: Orion Atlas EQ-g mount & Skywatcher EQ5 Pro.
ZWO mini guider.
Image cameras: ZWO ASI1600 MM Cool, ZWO ASI533mc-Pro, ZWO ASI174mm-C (for use with my Quark chromosphere), ZWO ASI120MC
Filters: LRGB, Ha 7nm, O-III 7nm, S-II 7nm
Eyepieces: a few.
Primary software: Cartes du Ciel, N.I.N.A, StarTools V1.4.

Image
User avatar
SkyHiker United States of America
Local Group Ambassador
Articles: 0
Online
Posts: 2293
Joined: Sat May 11, 2019 8:40 pm
4
Location: Santa Barbara, CA
Status:
Online

TSS Awards Badges

TSS Photo of the Day

Re: Yet another flats question...

#20

Post by SkyHiker »


Keep following that DSS thread, I posted a bit more. Someone offered to analyze it in DSS. Note my comments about the LP and why Blue is stacked near-perfectly, while LPR is a mess. It may very well be caused by sideways incident light during LPR.

The flats for LPR and Blue are identical, normalized, to less than 1% error. Note also that the circular footprint is off to the right - I may have to adjust the axial position of my secondary (the dreaded #1 step of first-time collimation, should be a one-time action).

Bob if you want to play with Scilab, install the IPCV module after installing Scilab. I think it's done by using ATOMS, just search for it and you will find some pointers. Scilab itself is easy to install, and IPCV is not a big issue - just a 3rd party Scilab module.

You will end up writing code that looks something like this in a .sce file (I added some tile averaging functions, and cd to the directory above Flats, Light and Bias, // means commented out):

cd "C:\Users\aling\Pictures\astro\Objects\NGC 2170";

function [x,y,z] = imLoad(fn, skip)
l = imread(fn);
[nx, ny] = size(l);
dx = int(skip/2);
dy = dx;
x = [dx+1:skip:nx-dx]';
y = [dy+1:skip:ny-dy]';
nx = length(x);
ny = length(y);
z = zeros(nx, ny);
for i = -dx:dx do
for j = -dy:dy do
z = z + double(l(x+i,y+j));
end
end
z = z/(2*dx+1)^2;
endfunction

function [x,y,z] = imLoad1(fn, skip)
l = imread(fn);
[nx, ny] = size(l);
x = [1:skip:nx]';
y = [1:skip:ny]';
l = l(x, :);
l = l(:, y);
z = double(l);
endfunction

skip=20;

fn = "Flat\Blue\MasterFlat_Gain210.tif";
[x,y,zfb] = imLoad(fn, skip);
fig1 = scf(1); clf(); plot3d(x,y,zfb);
fig1.children(1).data_bounds(1,3)=0;
//fig1.children(1).isoview="on";
title("Master flat Blue");

fn = "Flat\LPR\MasterFlat_Gain210.tif";
[x,y,zf] = imLoad(fn, skip);
fig8 = scf(8); clf(); plot3d(x,y,zf);
fig8.children(1).data_bounds(1,3)=0;
//fig8.children(1).isoview="on";
title("Master flat LPR");

fn = "Flat\Blue\MasterFlat_Gain210.tif";
zfi = ones(zf)./(zf/mean(zf));
//zfi = 1 + exp(2*(zfi-1));
fig7 = scf(7); clf(); plot3d(x/10000,y/10000,zfi);
//fig7.children(1).data_bounds(1,3)=0.9;
//fig7.children(1).isoview="on";
title("Master flat inverse normalized");

fn = "Bias\MasterOffset_Gain210.tif";
[x,y,zb] = imLoad(fn, skip);
fig2 = scf(2); clf(); plot3d(x,y,zb);
fig2.children(1).data_bounds(1,3)=0;
//fig2.children(1).isoview="on";
title("Master bias");

fn = "Light\Blue\NGC_2170_Light_001.tif";
[x,y,zl] = imLoad(fn, skip);
fig3 = scf(3); clf(); plot3d(x,y,zl);
fig3.children(1).data_bounds(1,3)=0;
fig3.children(1).data_bounds(2,3)=5000;
//fig3.children(1).isoview="on";
title("Single sub");

fn = "Light\Blue\AutosaveNGC_2170_Light_001.tif";
[x,y,za] = imLoad(fn, skip);
za = za - min(za);
za = za/max(za);
za = (2^16-1)*za;
za = min(2^16-1, 5*za);
fig6 = scf(6); clf(); plot3d(x,y,za);
fig6.children(1).data_bounds(1,3)=0;
fig6.children(1).data_bounds(2,3)=5000;
//fig6.children(1).isoview="on";
title("DSS-stacked single sub");

z = (zl-zb).*(zfi-0.8);
//z = (zl-zb).*zfi;
z = min(z, 2^16-1);
fig4 = scf(4); clf(); plot3d(x,y,z);
fig4.children(1).data_bounds(1,3)=0;
fig4.children(1).data_bounds(2,3)=5000;
//fig4.children(1).isoview="on";
title("Corrected single sub");

//z = (zl-zb).*(zfi-0.6);
z = (zfb/mean(zfb)).*zfi;
fig9 = scf(9); clf(); plot3d(x/10000,y/10000,z);
//fig4.children(1).data_bounds(1,3)=0;
//fig4.children(1).data_bounds(2,3)=1.;
//fig4.children(1).isoview="on";
title("Comparison with LPR flat");

z = z - min(z);
z = z/max(z);
z = (2^16-1)*z;
z = min(2^16-1, 5*z);
z16 = uint16(z);
scf(5); clf(); imshow(z16);
fig5.children(1).isoview="on";
title("Corrected single sub");

return;
... Henk. :D Telescopes: GSO 12" Astrograph, "Comet Hunter" MN152, ES ED127CF, ES ED80, WO Redcat51, Z12, AT6RC, Celestron Skymaster 20x80, Mounts and tripod: Losmandy G11S with OnStep, AVX, Tiltall, Cameras: ASI2600MC, ASI2600MM, ASI120 mini, Fuji X-a1, Canon XSi, T6, ELPH 100HS, DIY: OnStep controller, Pi4b/power rig, Afocal adapter, Foldable Dob base, Az/Alt Dob setting circles, Accessories: ZWO 36 mm filter wheel, TV Paracorr 2, Baader MPCC Mk III, ES FF, SSAG, QHY OAG-M, EAF electronic focuser, Plossls, Barlows, Telrad, Laser collimators (Seben LK1, Z12, Howie Glatter), Cheshire, 2 Orion RACIs 8x50, Software: KStars-Ekos, DSS, PHD2, Nebulosity, Photo Gallery, Gimp, CHDK, Computers:Pi4b, 2x running KStars/Ekos, Toshiba Satellite 17", Website:Henk's astro images
Post Reply

Create an account or sign in to join the discussion

You need to be a member in order to post a reply

Create an account

Not a member? register to join our community
Members can start their own topics & subscribe to topics
It’s free and only takes a minute

Register

Sign in

Return to “Image processing”