Page 1 of 1

Where to go from here to maximize results (Histogram, etc.).

Posted: Thu Nov 08, 2018 1:53 am
by Starry Jack
I have successfully created my 152mm Stage One Mod. I purchased an ERF from Val (Aries) and I love it (anybody want to buy my Lunt 100mm ERF?).

While I am pleased with the results, they are nowhere near what I am seeing others produce so I suspect there is more for me to try with your coaching OR...the really pretty photos are all accomplished in post processing.

EQUIPMENT:
-152mm achromat
-Val's Aries 150mm external ERF (homemade cell out of $4 in wooden knitting hoops!)
-->Glasspath 1.7x in front of the PST Etalon
-->PST Black Box with Standard 5mm BF
-->Baader 1.25" UV/IR Cut filter (Val recommended the Astronomic LV1 but I already own the Baader...let me know if the results should be superior with the Astronomic)
-->Mallincam AG 1.2C camera
* Sony IMX224 CMOS sensor (class 1)
* Sensor Sensitivity: 2350 mv
* 3.75X3.75 pixels size
* 1280 X 960 resolution and 640X480 resolution mode
* 1.2 Mp sensor, 1/3" size (6 mm) diagonal

I have a different camera, but until I can make the Stage 2 mod, I am limited to the Stage 1 1.25" nosepiece holder on the PST.

Given the above:
1) What should the histogram look like? Where do I want to spread my values on the left or right or where do I want the peak?
2) Where would you suggest GAMMA? I see noise with higher gamma but better contrast.
3) Where would I want CONTRAST and BRIGHTNESS?

I have attached a video to show you might first light. I have also attached screenshots of the HISTOGRAM and other SETTINGS (lots of poor seeing that day).
152mm Stage One PST Val ERF.mp4
(213.01 KiB) Downloaded 11 times
HISTOGRAM SETTINGS 3.png
HISTOGRAM SETTINGS 3.png (1.2 MiB) Viewed 220 times
HISTOGRAM SETTINGS 2.png
HISTOGRAM SETTINGS 2.png (1.34 MiB) Viewed 220 times
HISTOGRAM SETTINGS 1.png
HISTOGRAM SETTINGS 1.png (1.35 MiB) Viewed 220 times
Thanks,
Jack

Re: Where to go from here to maximize results (Histogram, etc.).

Posted: Thu Nov 08, 2018 2:49 am
by TheSkyBurner
Looks good but i cannot offer anymore suggestions except to try different locations of the glass path corrector. The focal length must be exactly -200mm for the first pst lens , with an input beam at exactly f/10.

Anything more or less and the image goes south. The glasspath corretor makes things more difficult to find the right area, but not impossible.

Also take into account that some PST etalons are no good and you will never achieve the results of valery who has an etalon from 2006. (before meade took over coronado)

Re: Where to go from here to maximize results (Histogram, etc.).

Posted: Thu Nov 08, 2018 3:48 pm
by MalVeauX
Heya,

Image in B&W (mono), no need to capture color data here. And then just look at your histogram in monochrome. That will help with simply judging things with the histogram, since imaging in color is going to show a lot more red channel than green or blue. It won't change the sensitivity, but it does make the histogram easier.

Regarding the histogram, I fill the histogram to the right just enough to not clip data. This is to get as much signal as possible recorded without losing data. I image with 200+ gain sometimes. Don't be afraid of noise. It mostly stacks out anyways (random noise does). I will go as slow as 10ms for exposure time on full discs and small scale imaging. But at higher resolution and larger scale, my minimum is 5ms exposure time, and shorter is better (2ms, 3ms, etc). For the surface, don't use any gamma, leave it neutral. Just fill the histogram. For prominences, I do use gamma (maximum value) with the same filled histogram. I capture both the surface & proms in one capture with maximum gamma use too. There's several ways to do it. But starting out, I would just focus on the surface and use no gamma (neutral) and get used to filling the histogram.

I don't use FireCapture, so I cannot really suggest much with the settings beyond basic exposure values and gamma use. I do not recommend using any of its internal processing as it's not likely as good as a good pixel editor (photoshop for example). But take that with a grain of salt, as I'm using SharpCap. I just get the RAW data and fill the histogram. I don't do any editing with brightness, contrast, etc, in the capture phase, I do that separately after I've stacked the data and have a good file to process.

I can't tell if you're in focus, or if seeing is really bad. So you'll have to confirm it's reaching focus appropriately, versus just bad seeing.

Here's how my histogram looks, filled without clipping data, in mono:
Histogram_Demo_01.jpg
Histogram_Demo_01.jpg (154.65 KiB) Viewed 183 times
Very best,

Re: Where to go from here to maximize results (Histogram, etc.).

Posted: Thu Nov 08, 2018 4:05 pm
by Starry Jack
Gents,
Thank you both for your input. I will fiddle with the Glasspath, but there are moments of good seeing in a video clip which indicate a sharp focus.

Most of my work is live video outreach so post processing is a luxury when I have a moment using Registax. I'D LOVE some pointers on Registax as well since I have blown away the settings I liked and can't seem to recreate good stacking to save my life.

Marty,
Your histogram advice is EXACTLY what I was looking for. Good stuff and I will practice immediately (sun permitting). I use the Mallincam Sky software designed for their cameras (some of the folks do use Sharpcap as well) rather than FireCapture. Regarding focus, the screen grabs above simply moments of poor seeing, but if you click on the MPG link you see when it snaps into decent focus.

Best,
Jack

Re: Where to go from here to maximize results (Histogram, etc.).

Posted: Thu Nov 08, 2018 5:05 pm
by MalVeauX
Hi Jack,

Yea, the video illustrates a much better image than the screen captures, thanks!. Still looks like either really challenging seeing, or perhaps not at critical focus to me though. But again, hard to know, a short video.

I think the obvious limitation here is likely the tiny blocking filter of the PST. It's robbing you of transmission and giving big vignetting. I would look into upgrading to a Stage II soon with a 15mm Coronado blocking filter. It would make a significant difference in general for everything.

I think FireCapture has an option for real-time flat calibration, which would be super helpful for live streaming outreach to remove dust motes and gradients and vignetting. I would look into this, especially for your purposes.

I've attached some examples with a histogram as above, the actual footage (forgive me if I killed the detail compressing it, but uploading 4gigs of video as 10Mb probably took away from the quality for sure, but, if you watch around 20seconds, the seeing improves and you can see individual spicules, so some good seeing in moments and definitely in focus). Attached also a stacked result of that video. And a colored version.

RAW video (compressed obviously though) (look around 20~25 seconds):
Example_Seeing_Focus_Chromosphere.mp4
Seeing improves around 20 seconds in, example of spicules, real time seeing
(9.57 MiB) Downloaded 14 times
Example Flat Frame (defocused with the disc filling the FOV):
Example_Flat_HA.jpg
Example of a flat frame
Example_Flat_HA.jpg (14.02 KiB) Viewed 175 times
Stacking results & histogram (no processing yet):
Example_Histogram_PostStack.jpg
Example of stacked unprocessed image with flat calibration with histogram from the RAW video
Example_Histogram_PostStack.jpg (52.25 KiB) Viewed 173 times
Processed the Stacked Image (a little histogram stretching, contrast and sharpening/deconvolution):
StackResult_01.jpg
B&W stacked result with minor processing
StackResult_01.jpg (169.48 KiB) Viewed 177 times
StackResult_02.jpg
False colored
StackResult_02.jpg (234.73 KiB) Viewed 177 times
Very best,

Re: Where to go from here to maximize results (Histogram, etc.).

Posted: Thu Nov 08, 2018 5:42 pm
by Starry Jack
Marty,
I love the progression of the raw video and the post processing. That is really helpful to gauge where I am. I am encouraged.

Regarding the Glasspath...I can't wrap my brain around where to place it. It is currently sitting on top of the etalon so I cannot move it closer to the BF. However, the BF is spaced according to the PST Black Box. If the beam is f6.5 and the Glasspath is 1:1.7, how do I figure the math for placement?

Can I ask for your workflow in your stacking program?

Spending money on Val's ERF means no Stage 2 mod for some time but a guy can dream!

Jack

Re: Where to go from here to maximize results (Histogram, etc.).

Posted: Thu Nov 08, 2018 6:02 pm
by MalVeauX
Hey Jack,

Hopefully someone else can chime in about your optical train to make sure an F10 or more narrow beam is hitting that PST. It doesn't have to be exactly F10, it just cannot be faster than F10 (ie, it cannot be F6 or F8). F10 is where it's supposed to be. And I think it's fine to go to F12 or even more narrow and it will still work fine (just dimmer image, but probably will tighten up bandpass a bit too if I recall correctly).

Work flow from above:

RAW video acquisition via SharpCap with histogram filled to around 90% or so, filled but not clipping data.
Defocused video acquisition to create a flat frame from the disc filling the FOV (not on a limb) at same exposure value.
Flat frame created in Autostakkert!2 with the defocused video.
Flat frame loaded on the RAW video in Autostakkert!2 and analyzed.
I stacked 60 of the best frame from the 2,000 that were in that video.
I load the stacked result in IMPPG and do a little histogram stretch and some deconvolution (sharpening).
I load the resulting image from IMPPG into Photoshop for final edits.
Final edits are just a levels adjustment to increase contrast; contrast adjustment; high pass filter x 2 (sharpening) and a light unsharpen mask (sharpening).
Coloring is done with levels adjustment.

Very best,

Re: Where to go from here to maximize results (Histogram, etc.).

Posted: Fri Nov 09, 2018 4:29 am
by Starry Jack
Thank you for the workflow, Marty. Makes sense. In fact, you've got me going back to Autostakkert again.

Attached is the video going through one of my trials in Autostakkert this afternoon. Much improved. Room for more to be certain.
Stack 29 Noise 4 Last Stack_0001_g4_ap331_conv SMALL.png
Stack 29 Noise 4 Last Stack_0001_g4_ap331_conv SMALL.png (1.2 MiB) Viewed 128 times

Re: Where to go from here to maximize results (Histogram, etc.).

Posted: Fri Nov 09, 2018 5:51 am
by TheSkyBurner
Starry Jack wrote:
Thu Nov 08, 2018 5:42 pm
Spending money on Val's ERF means no Stage 2 mod for some time but a guy can dream!


pretty sure the h-alpha pst uses 2" , 50.8mm threads. You can try using a cheapo SCT extension and that might make it really easy and cheap.

I havent made a pst mod in 5 years though so i do not entirely remember what the threads are.


update, just checked some notes.
The rear thread is the same as the front thread = M50 x1

So using some teflon pipe fitters tape, , and some patience with sand paper might get you that stage two real quick...

Re: Where to go from here to maximize results (Histogram, etc.).

Posted: Fri Nov 09, 2018 6:00 am
by TheSkyBurner
https://www.ebay.com/itm/BFB-SCT-Female ... 2916416148

something like this. then just need need your 2" eyepiece holder....

Re: Where to go from here to maximize results (Histogram, etc.).

Posted: Fri Nov 09, 2018 12:27 pm
by MalVeauX
Starry Jack wrote:
Fri Nov 09, 2018 4:29 am
Thank you for the workflow, Marty. Makes sense. In fact, you've got me going back to Autostakkert again.

Attached is the video going through one of my trials in Autostakkert this afternoon. Much improved. Room for more to be certain.
Looking good!

Very best,