M27 with ASI120MC

A place to share images that you have taken with SharpCap.
Forum rules
Please upload large images to photo sharing sites (flickr, etc) rather than trying to upload them as forum attachments.

Please share the equipment used and if possible camera settings to help others.
User avatar
turfpit
Posts: 1779
Joined: Mon Feb 13, 2017 8:13 pm
Location: UK
Contact:

M27 with ASI120MC

#1

Post by turfpit »

A recent image of M27, captured with a ZWO ASI120MC under an 80% waning moon. I didn't have this all my own way, see some of the problems I had to overcome - detailed below.

M27.png
M27.png (39.97 KiB) Viewed 5435 times

Capture:
  • SharpCap 3.0, settings file attached.
  • Celestron C8 on CG5 equatorial mount, TS f6.3 reducer/flattener, no filters.
  • Mount alignment = polar scope (eyeball!), 2 star alignment + 4 calibration stars carried out with f6.3 reducer and camera fitted to telescope and SharpCap reticule turned on.
  • Laptop LENOVO T400, CPU 2.4GHz Core 2 Duo, 8Gb RAM, 120Gb SSD, Windows 10 64-bit.
  • 7 frames @ 30s exposure.
  • Capture Area=1280x960.
  • Colour Space=RAW16.
  • Gain=50%.
  • Output to FITS file.
  • Darks taken in SC but not applied during capture.
Post processing:
  • Deep Sky Stacker - 7 lights + 1 dark stacked, RAW/FITS digital development process settings > FITS > debayered=GRBG, stacking=intersection mode, lights & darks Stacking Mode=Median Kappa-Sigma Clipping, output saved as TIFF image (32 bit/ch rational).
  • GIMP v9.5 (can process 16-bit with no loss of data) - Levels & Curves, export as PNG.
  • Nebulosity - star sharpening.
  • FastStone Image Viewer - reduce image size and bit depth for upload to forum.

The (undocked) capture histogram:
M27-histogram.PNG
M27-histogram.PNG (29.26 KiB) Viewed 5435 times
  • Camera setting Brightness adjusted to 10 to bring histogram away from left-hand side (ensures no data lost).
  • Histogram hitting right-hand side. Looks like ~10 pixels over exposed so data loss is minimal.
  • Colour balance acceptable.
  • Classic histogram shape for a DSO using logarithmic vertical scale.
I am trying to convince Robin that being able to save the histogram graphic automatically as part of a capture is really useful. See viewtopic.php?f=17&t=31 - any support here welcomed!


One of the 7 captures had a satellite passing through it. This was processed out (without intervention by me) during the stacking. I think the Median Kappa-Sigma clipping helped with that.
FITS-image.PNG
FITS-image.PNG (61.31 KiB) Viewed 5435 times
One of the FITS captures displayed in FITS Liberator. Notice the checked pattern - this is how a raw image looks before being debayered. The small bright green dots are hot pixels - which are removed when the dark frame is applied to the image. The histogram looks reasonable for a DSO. The nebula can be seen faintly in the centre - it is surprising that stacking only 7 x 30s images can produce the image at the top - I wonder what improvement using 100 or 500 images would make - one for the future?

The first attempt at post-processing was to debayer the frames using PIPP. I ended up with a green image even though I had chosen GRBG. This has worked before, so some sort of user error here which requires further investigation.
M27-bad.PNG
M27-bad.PNG (70.31 KiB) Viewed 5435 times
I managed not to take any flat frames during the capture :roll: (a race against clouds/rain), which I did not find out about until the next day. I resolved this in GIMP by using the Clone/Stamp tool to remove the dust motes.
:idea: 8-) :lol:

Here is my attempt at M27 taken around 1 year ago, same telescope/mount using an LN300 style video camera:
M27-1-year-ago.PNG
M27-1-year-ago.PNG (28.96 KiB) Viewed 5435 times
So it looks like persistence is the key!

Things for me to try:
  • Alternate stacking modes in Deep Sky Stacker (one for a cloudy night).
  • Capturing and stacking more images at 30s.
  • Capturing and stacking at 15s and 60s exposure.
  • More practise with Levels and Curves in GIMP.
  • Carry out the post-processing in Nebulosity (another one for a cloudy night).
  • Repeat the capture with the video camera used last year.
Hopefully some of the things that didn't go right will serve to encourage others.

Dave
Attachments
M27_0001.CameraSettings.txt
(586 Bytes) Downloaded 217 times
User avatar
oopfan
Posts: 1321
Joined: Sat Jul 08, 2017 2:37 pm
Location: New York
Contact:

Re: M27 with ASI120MC

#2

Post by oopfan »

Hi Dave,

This is very useful information, thank you! I have a question about the histogram. You said:
"Camera setting Brightness adjusted to 10 to bring histogram away from left-hand side (ensures no data lost)"

I have not played with this, ever. I must plead ignorance because I never knew that the camera accepted an input called "brightness". I have just been tweaking gain and exposure to get a desired result.

I don't know if you have seen my posts on using OpenCVStarDetector. The problem I am having is that I see 7 stars in the image but it only sees 3. I think what I need to do is a histogram stretch in order to give the star detector more meat to sink its teeth into.

Would it be possible for you to give me a link to the SharpCap documentation for a detailed discussion of "Brightness"?

Thanks,
Brian
User avatar
turfpit
Posts: 1779
Joined: Mon Feb 13, 2017 8:13 pm
Location: UK
Contact:

Re: M27 with ASI120MC

#3

Post by turfpit »

Brian

Thanks for that. Brightness might be called 'offset' depending on the camera manufacturer/model.

Have a look at http://docs.sharpcap.co.uk/2.9/#Using%2 ... %20Quality for some histogram 'guidelines'. Writing this and the research that went into producing it was my 'big breakthrough' in astro-imaging (I had no background in astronomy or photography at the time).

If you can see more stars than the software then try increasing the camera gain. Also, have a look at the black level http://docs.sharpcap.co.uk/2.9/#Setting ... ck%20Level.

With all of this, I have learned that the histogram is my friend, I use that rather than trying to judge what is happening by looking at the display.

Dave
User avatar
turfpit
Posts: 1779
Joined: Mon Feb 13, 2017 8:13 pm
Location: UK
Contact:

Re: M27 with ASI120MC

#4

Post by turfpit »

Brian

With planetary imaging (using the same ASI120MC camera), I set Brightness=0 to get the 'right' shape of histogram for solar/lunar/planetary.

See the histogram in post viewtopic.php?f=16&t=256.

Dave
User avatar
oopfan
Posts: 1321
Joined: Sat Jul 08, 2017 2:37 pm
Location: New York
Contact:

Re: M27 with ASI120MC

#5

Post by oopfan »

Thanks, Dave. For star detection it looks like I should play with Gamma too. I now see that adjusting brightness doesn't really do a histogram stretch -- instead it appears to just shift the histogram left and right (hence the term "offset"). Gamma, on the other hand, appears to have a stretching effect although it is not stretching in the purest sense. I will give it a try! Thanks again.
User avatar
turfpit
Posts: 1779
Joined: Mon Feb 13, 2017 8:13 pm
Location: UK
Contact:

Re: M27 with ASI120MC

#6

Post by turfpit »

A bit on gamma and it's effects here http://docs.sharpcap.co.uk/2.9/#Stretch ... ck%20Image.
User avatar
oopfan
Posts: 1321
Joined: Sat Jul 08, 2017 2:37 pm
Location: New York
Contact:

Re: M27 with ASI120MC

#7

Post by oopfan »

Thanks, lots to learn.

Recently I began using Gain of 25 (instead of 50) for DSO's with my ASI120MC and exposures of 120sec. I got the idea from this gentleman's photo of NGC 4565:

http://astrob.in/290513/0/

If you click on the image and then click on the full resolution button and then pan over to the galaxy you will see incredible dynamic range in the dust lane. I am really impressed. Then if you switch back to his Technical Card you will see that his gain setting is only 8.00. What? Most guys with cooled cameras use gains in excess of 100 -- and I would notice that their nebulosity is quantized. So I brought up the user manual for my camera and I saw that I can increase dynamic range by lowering the gain. I did that, and switched over to RAW16 from RGB24. I only had one opportunity to test this out the other day -- it was a full Moon so I am not making any quick judgements.

The default setting for brightness on my camera is 0 (on a scale of 0 to 20) and gamma is 50 (on a scale of 0 to 100). While I continue to have awful weather I am going to redo darks with the brightness set to 10. I am debating whether or not to also set gamma to 75 in order to stretch out the faint nebulosity. Doing darks is an all-night endeavor as I capture frames across a wide range of sensor temperatures and then later move them into separate folders for when I do stacking.
User avatar
oopfan
Posts: 1321
Joined: Sat Jul 08, 2017 2:37 pm
Location: New York
Contact:

Re: M27 with ASI120MC

#8

Post by oopfan »

Hi Dave,

I joined the ZWO User Forum and posted a topic. Who knows if I will get a reply. Of the three inputs (Gain, Offset and Gamma) that I would like to tweak I am most curious about how ZWO implements Gamma "Correction". Here is my supplemental post that expands on Gamma.

Dave, do you have any knowledge on how ZWO implements this?

My post:

Wikipedia states that Gamma correction originated back in the days of the Cathode Ray Tube (CRT). I am old enough to remember CRT's, sad to say! Even before CRT's were used in computer monitors they were used in television sets. Take my word for it -- there was no such thing as "digital" in television sets back then. Historically, Gamma correction was performed in the analog circuitry of the amplifier that deflected the electron beam across the phosphor.

Fast forward to 2017 -- the trend these days is to do everything in software -- which leads to the question:

In the ASI120MC does Gamma correction happen in the analog amplifier before the signal is digitized? Or does Gamma happen in software after the signal is digitized?

If the answer is "in software after the signal is digitized" then that would be counterproductive and a shame.
User avatar
oopfan
Posts: 1321
Joined: Sat Jul 08, 2017 2:37 pm
Location: New York
Contact:

Re: M27 with ASI120MC

#9

Post by oopfan »

Well, I looked at the datasheet for the AR0130CS from ON Semiconductor. The chip has a built-in A/D converter and 12 bits of digital output. There are inputs for gain for each channel but no mention of gamma.

So it looks to me that gamma is being applied after digitization by either ZWO firmware on the camera or within SharpCap. (It is never easy to track down where the magic happens! No one is very upfront about it sadly.)

The bottom line is that there is nothing to be gained by tweaking gamma for DSO capture. It is best to do that in PixInsight. It is a shame.
User avatar
oopfan
Posts: 1321
Joined: Sat Jul 08, 2017 2:37 pm
Location: New York
Contact:

Re: M27 with ASI120MC

#10

Post by oopfan »

Dave, I finally got a clear night. With regards to star detection using OpenCVStarDetector, the number of stars detected increases significantly when I boost Brightness from 0 to 10 and Gamma from 50 to 75. That seemed to be the best combination. That along with Gain of 100. Still, though, I cannot go much below a 2 sec exposure. My goal was 1/2 sec.

That's OK, I was attempting to use a novel approach of using triangle-generation (a la FOCAS) and Generalized Hough Transform to do in realtime (every 1/2 second) essentially what stacking software does (that's not exactly the application but it is a good analogy). But of course I need a minimum of 3 stars, and I wanted to be able to point the telescope anywhere in the sky and have it work! That's a tall order. Anyhow, I have a fallback algorithm that has been working well. I just wanted to try something new.
Post Reply