Noise, Firecapture and imaging questions

this is the main message area for anything solar :)
User avatar
Montana
Way More Fun to Share It!!
Way More Fun to Share It!!
Posts: 18968
Joined: Mon Oct 17, 2011 5:25 pm
Location: Cheshire, UK

Noise, Firecapture and imaging questions

Post by Montana » Mon Jan 01, 2018 2:06 pm

Happy New Year! to everyone here! may 2018 be full of clear skies and lots of solar activity to upset the scientists theories :)

Now for some questions I hope some of you wise folks can answer.

Last October I bought a Celestron C11 in the hope of imaging the Moon and Sun in close up. I have noticed that at prime focus I see the same level of resolution as I do from my 140mm refractor. However this could be seeing conditions, but what I don't understand is that through my refractor the image has lots of contrast and the histogram is spread over the whole range. When I use my Grasshopper 3 on the C11 the brightness is the same on the screen but the histogram is all shoved the left. This results in not seeing any contrast to aid focusing. It is OK in post processing as you can stretch out the histogram and normalise it, but this doesn't help using on screen capture or stacking as the image is dark and featureless.

1. Why should the output of the camera be so different between two telescopes? surely it sees only what light comes in. Why is the histogram all the way to left in the C11 and spread out with the refractor? Can the histogram be stretched out manually on screen to see during capture? why only in post processing?

Next, noise. When imaging the Sun or Moon through my refractor I can take 1000 and stack 5% and get virtually a noise free image with the Grasshopper 3. Through the C11 I have to take 3000 and stack 15% to get even remotely a decent image, but it is still full of noise (no gain used in either).

2. Why should noise appear just by changing telescopes? the camera is the same.

So I bought a power cable from Rupert as Mark swears by battery power being clean thus reducing noise so you can use more gain. So I tried this a couple of days ago and I was very disappointed. For a start Firecapture 2.3 kept crashing every 3 minutes so if you didn't get your capture within that 3 minutes you lost the ser file. This does not happen when powered by the computer (just tested it). So I changed to Firecapture 2.5 ( the lastest version) and this stopped the crashing. However I don't like Firecapture 2.5 as it does illogical things. You can set the exposure in 16 bit to be correct:

3. Then if you close the program down and restart then the same exposure level is either far too bright or far too dark? why? the exposure should be the same, so the on screen view should be the same every time?

I noticed this because over half an hour or so the frame rate gradually kept dropping from 26 fps down to 18, so a restart was necessary every 15 minutes or so to get the frame rate back. Again, this never happens under computer power.

I tested the noise reduction of computer power vs Battery power, same mag, no gain, stack of 15%. I can't see any difference?
Noise comparison.jpg
Noise comparison.jpg (117.34 KiB) Viewed 1056 times
Only difference seems to be it crashes Firecapture and frame rate drops off, so powering from a battery is a waste of time. I haven't even tried to add gain but if there is no difference without gain I can't see it helping.

Now for the problem of output and contrast (see question 1). All this seems illogical to me.

Firecapture 2.5 - image brightness looks good on screen (depending on how the program randomly opened) but the histogram shows a big bulge only in the left, if you increase exposure the lump on the left just gets higher not wider to fill the histogram, it doesn't move to the right?

Icon on the screen of the ser file looks great
Icon shot.PNG
Icon shot.PNG (42.23 KiB) Viewed 1056 times
or even playback is perfect
Playback.PNG
Playback.PNG (719.1 KiB) Viewed 1056 times
Open in ASK3 and it looks black and it is difficult to get alignment features
ASK3.PNG
ASK3.PNG (692.2 KiB) Viewed 1056 times
The output is also black
ASK3 output.PNG
ASK3 output.PNG (340.44 KiB) Viewed 1056 times
Place it is Registax and suddenly it is fine
Registax output.PNG
Registax output.PNG (1002.92 KiB) Viewed 1056 times
or in Photoshop it is black but the histogram is skewed all the way to the left
Photoshop levels.PNG
Photoshop levels.PNG (72.11 KiB) Viewed 1056 times
What on earth is going on? I noticed this before when using Firecapture 2.5 when imaging Jupiter and every file is different, some are dark, some are light, it seems to be random even though you don't change the exposure. Firecapture 2.3 does not do this, what you see on screen is what you get and what you see in ASK3. I don't get it! it is very annoying indeed and why I choose not to use Firecapture 2.5 unless I really have to. If Genika would work for all my cameras I would use that, but it doesn't.

4. How do I get Firecapture 2.5 to stop messing up the histogram? how do I get Autostakkert 3 to unstretch the histogram so that it can see all the features to stack on just like the live view?

I don't like it when cameras and software behave in a totally unpredictable manner, if you have some scientific fact of why all these things are taking place, then I can understand them and correct them. Not understanding is something I really don't like. Any help you can give would be very much appreciated :) If I can't sort these out then I have no hope taking granulation on the Sun with this set up, as last time I couldn't see any contrast at all to focus or even image with.

Let's hope 2018 gets better with more predictable software :) knowledge is key.

Alexandra

User avatar
PDB
Im an EXPERT!
Im an EXPERT!
Posts: 293
Joined: Mon Aug 15, 2016 4:23 pm
Location: Belgium

Re: Noise, Firecapture and imaging questions

Post by PDB » Mon Jan 01, 2018 5:55 pm

Hello Alexandra,

happy 2018 to you too. Lot's of questions to start the year. I don't have all the answers, but might give you some hints ...
Last October I bought a Celestron C11 in the hope of imaging the Moon and Sun in close up. I have noticed that at prime focus I see the same level of resolution as I do from my 140mm refractor. However this could be seeing conditions, but what I don't understand is that through my refractor the image has lots of contrast and the histogram is spread over the whole range. When I use my Grasshopper 3 on the C11 the brightness is the same on the screen but the histogram is all shoved the left. This results in not seeing any contrast to aid focusing. It is OK in post processing as you can stretch out the histogram and normalise it, but this doesn't help using on screen capture or stacking as the image is dark and featureless.

1. Why should the output of the camera be so different between two telescopes? surely it sees only what light comes in. Why is the histogram all the way to left in the C11 and spread out with the refractor? Can the histogram be stretched out manually on screen to see during capture? why only in post processing?
I don't know exactly which refractor you have, but it is probably faster f/ratio than the C11@ f/10. Refractors (good obes) are also known for having more contrast than the SCT where the obstruction in the corrector plate will reduce contrast little bit. I don't know if you have checked collimation on the C11. (Having a C8 myself, I know that needs to be checked and adjusted at "regular times". Some hold collimation fairly well, som others need more adjustment.

So I suppose the difference in contrast causes the difference in historgram. (Also remember the difference in FOV between the 2 scopes has influence on the histogram)

For your focussing: FireCaptore and Genika have features to enhance contrast for viewing only. So you can stretch the image on screen without altering the captured image. (Don't remember from which version in FC this is there)
2. Why should noise appear just by changing telescopes? the camera is the same.
Just a guess, but you described a darker image with the C11, so signal will be lower, noise the same, so S/N ration is not the same, and you need more images.
3. Then if you close the program down and restart then the same exposure level is either far too bright or far too dark? why? the exposure should be the same, so the on screen view should be the same every time?
This looks very weird. (Also the other issues you describe) Do you have the latest software for your camera installed? It might be that FC2.5 expects a more recent SDK for the camera than you have installed. (and therefore FC 2.3 might work better)
As you said, errors and behaviour should be consistent but if different versions are mixed together, strange things can happen. (And lots of the "features" you describe seem to be related to the Grasshopper part of the software) Maybe you shouls send these question directly to Torsten Edelmann.
Open in ASK3 and it looks black and it is difficult to get alignment features
Have you tried Display Options Brightness 1x (up/dow) At least that should help viewing the file in AS!3


Hope this gives you some idea's were to look.

Regards,

Paul

User avatar
eroel
Way More Fun to Share It!!
Way More Fun to Share It!!
Posts: 3600
Joined: Mon Dec 19, 2011 10:45 pm
Location: México D.F.

Re: Noise, Firecapture and imaging questions

Post by eroel » Mon Jan 01, 2018 9:36 pm

Alexandra:
First wish you, the hubby and all your family, the best for this starting year, hoping it is full of health, happiness and well being, plus good seeing. :D
I can not help much, but just say that you have a really good corrected refractor made by Yuri Petrunin. I know him since he started, he made my 10" f/20 Maksutov a really optical jewel, you could use this scope at 700X to 1000X during 2 Mars oppositions without the image breaking down.
When using a good refractor you get used to the images it gives, so you will have to get used to catadioptric images too.
Optical resolution depends on optical aperture, so the 11" will have more resolution, (Dawes limit) but as Paul says, you have the big secondary obstruction that lowers contrast too.
As why the Registax gives you a brighter image I don't know, but probably there is a parameter limited on the AS3 set on.
I image the Moon and Planets with a 6" f/12 AP triplet refractor, with the 10" f/20 Maksutov and with a superb optics 12" f/10 Meade SCT, all give different settings when recording images with the same camera, just have to get used to it.
Again, my best wishes to you et al members of this worldwide wonderful forum.
Eric.

User avatar
pedro
Way More Fun to Share It!!
Way More Fun to Share It!!
Posts: 5511
Joined: Sun May 01, 2016 8:26 pm
Location: Portugal
Contact:

Re: Noise, Firecapture and imaging questions

Post by pedro » Tue Jan 02, 2018 9:27 am

HI Alexandra

I had many problems with Firecapture 2.6, but Firecapture 2.5 works fine. I use it all the time.

I never had the histogram issues you mention.

Updating Flycapture SDK (PGR) might be a good idea

User avatar
Montana
Way More Fun to Share It!!
Way More Fun to Share It!!
Posts: 18968
Joined: Mon Oct 17, 2011 5:25 pm
Location: Cheshire, UK

Re: Noise, Firecapture and imaging questions

Post by Montana » Tue Jan 02, 2018 10:22 am

Thanks guys for all the infomation, I'll try updating the Flycapture first, although I am sure I only did this last year. I wish Genika would work properly and I would switch. I need to start bugging him again.

So I think my first port of call is to try and get the cameras working properly, I am sure the battery cable is not working properly either. Then find the histogram stretch in Firecapture/Genika to help focus, then just get used to the C11 not being good for contrast. I need many frames captured to sort out this noise if the noise comes from lack of contrast and signal to noise ratio (that makes sense at least).

Yes I have tried upping the brightness in ASK3 to x2 and this makes viewing better but I am not sure it influences whether it stacks better by seeing more features? I need to ask Emil.
Alexandra

User avatar
pedro
Way More Fun to Share It!!
Way More Fun to Share It!!
Posts: 5511
Joined: Sun May 01, 2016 8:26 pm
Location: Portugal
Contact:

Re: Noise, Firecapture and imaging questions

Post by pedro » Tue Jan 02, 2018 10:40 am

I forgot to mention that I never capture 16bits ser files. For solar and lunar images IMHO 8bit avi files are more than enough.

After stacking (AS3) I save a 16bit Tif file and process it with Registax and PS

I aso have a power cable for my PGR cameras. I used it a couple of times and saw no difference (sorry Mark)

User avatar
marktownley
Librarian
Librarian
Posts: 23509
Joined: Tue Oct 18, 2011 5:27 pm
Location: Brierley Hills, UK
Contact:

Re: Noise, Firecapture and imaging questions

Post by marktownley » Wed Jan 03, 2018 7:46 pm

I get a new laptop at the end of the week, I'll post a proper reply when back on big screen iPhone screen is just too small for longer reply.
Image
http://brierleyhillsolar.blogspot.co.uk/
Solar images, a collection of all the most up to date live solar data on the web, imaging & processing tutorials - please take a look!

User avatar
marktownley
Librarian
Librarian
Posts: 23509
Joined: Tue Oct 18, 2011 5:27 pm
Location: Brierley Hills, UK
Contact:

Re: Noise, Firecapture and imaging questions

Post by marktownley » Sat Jan 06, 2018 6:52 am

Hello! Back again, with a new laptop with SSD drive - woohoo! Been a bad end of year for breakages for us - laptop packed up (mother board fried) mid december, followed by central heating boiler and then the TV died - throw in being away and getting fixes and replacements has been protracted - at least new year sales saved a bit of £££...

Anyway, back to the thread...

Scopes first:

You have one of the optically finest refractors on the planet and are comparing it to a (relatively!) cheap mass produced off the shelf SCT. Have a search on cloudy nights as there is loads on this comparing decent refractors to much larger SCTs in terms of planetary observation - the frac is the winner every single time! As others have said the central obstruction on a SCT contributes to a lowering of contrast (34% CO). I'm sure i've read in the past that the TEC140 is optically figured to 1/50 wave, the SCT if you are very lucky might get 1/8th wave, probably somewhere between 1/4 and 1/8th wave. With this in mind contrast with the SCT is always going to be considerably lower. eg https://www.cloudynights.com/topic/2002 ... nd-11-sct/ or see what Damian says http://www.damianpeach.com/simulation.htm

What does the SCT give you that the frac doesn't? well 2 things, light gathering power (the bigger aperture lets in more light) and resolving power of point sources. The latter is going to be limited by seeing, if the seeing caps the resolution there is no point using a scope with a larger aperture than the seeing allows as there is no gain. The TEC gives you a maximum resolution of just under 1 arc second, the C11 gives you a maximum resolution of just under half an arc second. So, for example, if the seeing is running at 1 arc second you will always get better views with the TEC as it has higher contrast than the SCT. The SCT will only outperform it when the seeing is less than 1 arc second, and even then you are still getting the lower contrast view even if the resolution is theoretically higher - what will be the result? The SCT is always playing catch up with the TEC. You get more resultant noise with images with the C11 than the TEC as the C11 an an optical system has a much poorer signal to noise ratio. That's why the histogram is squished up with the C11.

How can a SCT / MCT exceed the performance of a frac? If it is very good quality (like Erics 10" MCT) - but then add another zero on the end of the price tag. I went for my HaT over using the C8 that I already have and getting one of Valerys ERFs (sorry Valery!) because Fred hand picks the donor OTAs that he uses for the HaT and these are guaranteed to be a certain optical performance that come with a test report. This cost me a couple of K£ more but is worth it IMHO. To use an analogy, you have got used to drinking very fine wine, and now you have tried a cheaper bottle of plonk off the bargain shelf and it is not going to live up...

Cameras next:

I had issue with my GigE camera not behaving and then removed and reinstalled the latest version of the SDK file and all is good again, give it a try.

Capture:
I'm totally with Pedro here, I see no gain in capturing the 16bit files over 8 bit, all the 16 bit does is take up more disk space and takes longer to process. Getting AS3 to then save the stacked image as a 16bit tiff file and processing this is the way forward. The big learning curve for me when going 'hi-res' with the HaT which is something Valery has always attested too (and I totally agree Valery!) is the camera frame rate, higher fps is essential. With my GigE camera I get ~36fps, with the USB3 camera using 2x2 binning I get pretty much the same effective pixel size but are then able to run at ~120fps, the extra light binning gives means a much shorter exposure time too. 3000 frames at 26fps means you are doing 120 seconds of exposure time roughly. I found doing animations with the HaT with a cadence time of 15 seconds there was quite clear movement between the respective sub frames of the animation. It could be recording so long you are getting temporal blurring which on the moon would manifest on terminator zones where there is a moving shadow. Not really done any high res lunar so not sure what the exposure length rules are for a given aperture. Not really sure what is happening with the brightness issue - see if it happens in 8bit.

Power:

Let me ex[;ain how, why and for what reason I power my cameras. In another life when i'm not spending money on solar i'm an audiophile and spend money on high end hifi. I'm not going to dwell on a load of science behind it but the quality of the power supplies, cabling and mains going into the hifi makes a huge difference, greater than changing to higher quality(more expensive) source components to a certain degree. The same science applies to the cameras. I power my cameras so that I can run at a higher gain setting than I could if the camera was not powered; this allows me to run with considerably shorter exposure time (higher fps) and allows me to freeze the seeing. If I turn the camera gain up when it is powered by USB there is a lot more 'dancing snow' on the screen (noise) than if I run it with gain and powered by battery, indeed I always run with the gain slider about half way along with the camera powered with no issues with noise. The quality of USB power will vary greatly from laptop to laptop, pc to pc and so will the resultant effect. YMMV.

Sorry for delay in replying, I couldn't have typed all that on the Iphone! :D

Mark
Image
http://brierleyhillsolar.blogspot.co.uk/
Solar images, a collection of all the most up to date live solar data on the web, imaging & processing tutorials - please take a look!

User avatar
Merlin66
Librarian
Librarian
Posts: 3039
Joined: Tue Dec 13, 2011 1:23 pm
Location: St Leonards, Australia
Contact:

Re: Noise, Firecapture and imaging questions

Post by Merlin66 » Sat Jan 06, 2018 7:35 am

Mark,
You’ve obviously taken some time to prepare your post, but I’m not sure it actually responds to Alexandra’s post.

I use a USB3 and SSD drive with a ASI174 to achieve >500 fps.
"Astronomical Spectroscopy - The Final Frontier" - to boldly go where few amateurs have gone before
https://groups.yahoo.com/neo/groups/ast ... scopy/info
"Astronomical Spectroscopy for Amateurs" and
"Imaging Sunlight - using a digital spectroheliograph" - Springer

User avatar
marktownley
Librarian
Librarian
Posts: 23509
Joined: Tue Oct 18, 2011 5:27 pm
Location: Brierley Hills, UK
Contact:

Re: Noise, Firecapture and imaging questions

Post by marktownley » Sat Jan 06, 2018 8:53 am

Me again... Spotted something that may or may not affect things; 3rd image down bottom right says it is a 14bit mono image, 4th image down AS3 screen grab shows you have this set as 16bit - maybe a mismatch as it also shows you have autoscale on? Maybe set AS3 to a 14bit range.
Image
http://brierleyhillsolar.blogspot.co.uk/
Solar images, a collection of all the most up to date live solar data on the web, imaging & processing tutorials - please take a look!

User avatar
Montana
Way More Fun to Share It!!
Way More Fun to Share It!!
Posts: 18968
Joined: Mon Oct 17, 2011 5:25 pm
Location: Cheshire, UK

Re: Noise, Firecapture and imaging questions

Post by Montana » Sat Jan 06, 2018 8:03 pm

!!!! well spotted Mark, I have never ever seen this before. If you click autoscale off then go down to 14 bit suddenly I have full contrast and bright photo. I didn't know the Grasshopper was 14 bit. I am just going to try and process and see if it makes a difference.
Very interesting post Mark, I really do prefer and enjoy the TEC so much more, but I so dearly wish to witness angular granulation rather than round ones.
I disagree about capturing in 8 bit though. When I capture WL with the TEC in 8 bit vs 16 bit it is like night and day, the 16 bit is so smooth and processing is a dream. There is so much more to play with, it is in Halpha too but not as noticeable as the WL.
I am not sure I like the power cable, the mount seems to totter about and the camera also drops frames and coughs and splutters. I don't think the power of the battery can stand it, whereas the computer seems to be so stable. The noise seems the same between the two and I can't raise the gain even a small amount without it making it worse. Maybe my laptop power from my USB is very good?
Well it has finished processing and the output is a miracle! proper scaling and huge increase in sharpened detail and far less noise. WOW!! what a difference :hamster:
:bow2
Alexandra

User avatar
Montana
Way More Fun to Share It!!
Way More Fun to Share It!!
Posts: 18968
Joined: Mon Oct 17, 2011 5:25 pm
Location: Cheshire, UK

Re: Noise, Firecapture and imaging questions

Post by Montana » Sat Jan 06, 2018 8:17 pm

This is unbelievable!!! where has the noise gone :hamster: :hamster: :hamster: I have to process more...........

User avatar
marktownley
Librarian
Librarian
Posts: 23509
Joined: Tue Oct 18, 2011 5:27 pm
Location: Brierley Hills, UK
Contact:

Re: Noise, Firecapture and imaging questions

Post by marktownley » Sun Jan 07, 2018 5:08 am

Well that's good if the brightness issue is fixed.

Very interesting about your 16bit experience with the WL, maybe it is because contrast levels are lower than with CaK or Ha, which is where i've done my 8/16 bit comparisons in the past. I shall certainly give it a re-try... Could well be your laptop has stable power, at the end of the day the important thing is you do what works for you.
Image
http://brierleyhillsolar.blogspot.co.uk/
Solar images, a collection of all the most up to date live solar data on the web, imaging & processing tutorials - please take a look!

User avatar
PDB
Im an EXPERT!
Im an EXPERT!
Posts: 293
Joined: Mon Aug 15, 2016 4:23 pm
Location: Belgium

Re: Noise, Firecapture and imaging questions

Post by PDB » Sun Jan 07, 2018 8:47 am

Hi Mark & Alexandra,

interesting discussion. First question: which Grashopper camera are you using? There are several models with different sensors and some of them seem to have a 14bit ADC (ICX285, ICX674) while others like the IMX174 model have only 10bit (highspeed) or 12bit (normal speed) ADC's. When there is a 14bit ADC and you don't loose speed, 16 bit will give you more options in processing. In good conditions, you can use the same recording to get the proms and surface, and then combine the two processed stacks in one image. (So carefull consideration of what is most important speed or dynamic range should be considered)

That 14 bit value is really strange, but I have seen some SER files recorded with the same camera where the value sometimes indicates 10, 12 or 14 bit instead of 16. (It should be in the header of the SER file, so it woul be good to have a look at that) All depends probably how the camera interprets the values, and also how that is written to disc. (Had some discussion about that with Torsten Edelman some time ago, and he confirmed he does not manipulate any of the camera data. Image data coming from the camera is written as is to the data-stream) I did some experiments long time ago and if I remember correctly, it looked as if ZWO will stretch the 14-bit into 16bit (need to retest that). No idea how Point Gret does.

This could be tested by making a short video with forces full well and then look into the SER file how the data really looks.

Regards,

Paul

User avatar
marktownley
Librarian
Librarian
Posts: 23509
Joined: Tue Oct 18, 2011 5:27 pm
Location: Brierley Hills, UK
Contact:

Re: Noise, Firecapture and imaging questions

Post by marktownley » Sun Jan 07, 2018 5:41 pm

Interesting indeed. It got me looking at my cameras on the PGR website: Seems my 2.3Mp Blackfly can operate in 10 or 12 bit depending on setting https://www.ptgrey.com/blackfly-23-mp-m ... ius-imx249 and my Chameleon 3 can operate in 12 bit https://www.ptgrey.com/chameleon3-32-mp ... sonyimx265

I've never twiddled with any of the settings like this in AS3, I think I will have to. Maybe this is why in my previous 'trials' i've not seen any difference between 8 and 16 (well what we think is 16 ;) ) bit recordings.

Looks like we have the basis for a good experiment in the spring and summer when the sun is higher in the sky ;)
Image
http://brierleyhillsolar.blogspot.co.uk/
Solar images, a collection of all the most up to date live solar data on the web, imaging & processing tutorials - please take a look!

User avatar
Valery
Way More Fun to Share It!!
Way More Fun to Share It!!
Posts: 2694
Joined: Sun Nov 18, 2012 3:13 pm

Re: Noise, Firecapture and imaging questions

Post by Valery » Sun Jan 07, 2018 5:47 pm

Hi all,

Very interesting discussion.

First is the question to Ken. How have you reached the 500fps mark? Very small ROI?

To Mark: for lucky imaging and further massive post processing the optical quality of a scope and central obsetruction plays much less role than for visual planetary observing. All the contrast loss due to C.O. we can easily compensite by just contrast increasing.
You have had no gain due to hand picking of a scope. SCT already have about 1/3 to 1/4 wave of spherical aberration. So, for the final image to deconvolve there is no real difference were the images taken with hand picked or just a standard SCT (assume no astigmatism in both). And a standard (not a picked up) SCT can have even advantage over a picked up one! In the case the standard one has overcorrected optics - in such a case it can be close to null correction in a red light! More so, if you have a standard SCT, not a HD, you can move the focal point further outward and find the position where the red is null. This is impossible for a HD versions. Having a separate DERF, you always can choose a better scope if you meet it and then sell your previous one and a telescope is not limited to use in a red light.

To Alexandra: you are correct when telling us that a 16bit recording is better. This is really true when we have a very high difference of brightness we recording. This is especially true for lunar landscapes near the terminator, for imaging the sun flares, ative areas in a Ca II K or H lines, prominences at the limb when we wish to image also the surface details. Disk space and processing time is a third priority here.
"Solar H alpha activity is the most dynamic and compelling thing you can see in a telescope, so spend accordingly." (c) Bob Yoesle.

Largest full size 185 - 356mm Dielectric Energy Rejection Filters (D-ERF) by ARIES Instruments.

User avatar
PDB
Im an EXPERT!
Im an EXPERT!
Posts: 293
Joined: Mon Aug 15, 2016 4:23 pm
Location: Belgium

Re: Noise, Firecapture and imaging questions

Post by PDB » Sun Jan 07, 2018 5:59 pm

Looks like we have the basis for a good experiment in the spring and summer when the sun is higher in the sky ;)
Hopefully my Quark is back by then :(
Still some programming to do for my Solar Scintillation monitor program, after that probably some time to do more testing.
Interesting indeed. It got me looking at my cameras on the PGR website: Seems my 2.3Mp Blackfly can operate in 10 or 12 bit depending on setting
This is triggered by the "High Speed" setting in FireCapture (Highspeed marked = 10 bit, if not 12 bit ADC) In Genika i think is set by Mode 7 in the camera settings.

P.

User avatar
Montana
Way More Fun to Share It!!
Way More Fun to Share It!!
Posts: 18968
Joined: Mon Oct 17, 2011 5:25 pm
Location: Cheshire, UK

Re: Noise, Firecapture and imaging questions

Post by Montana » Sun Jan 07, 2018 8:01 pm

This is so interesting guys, I have not ever seen such a difference in processing coming out of ASK3 before, getting the right setting gives a whole leap in extra sharpness and no noise in my moon shots. I didn't even know that button existed before. I think it is a must have check box now for everyone.
I have looked at my Grasshopper, it does say 14 bit ADC https://www.ptgrey.com/grasshopper3-28- ... 674-camera
Well what a revelation :) I must try reprocessing some of my Daystar images that were always dark, full of noise and no contrast :)
Alexandra

User avatar
Montana
Way More Fun to Share It!!
Way More Fun to Share It!!
Posts: 18968
Joined: Mon Oct 17, 2011 5:25 pm
Location: Cheshire, UK

Re: Noise, Firecapture and imaging questions

Post by Montana » Mon Jan 08, 2018 9:37 pm

So I am now re-processing and half are 14 bit Ser (as per manufacturer) and half are 15 bit ser? Now the 14 bit process amazing images, the 15 bit won't process as ASK3 only has 12, 14 or 16 settings. The 16 are extremely dark and when using the 14 bit, the 15 bit images have strange black artefacts. Why are some 14 and some 15? is there some way I can stop it recording 15 bit and always keep it 14. No good really if it is random.

Here is the problem
15 bit.PNG
15 bit.PNG (395.53 KiB) Viewed 741 times

User avatar
PDB
Im an EXPERT!
Im an EXPERT!
Posts: 293
Joined: Mon Aug 15, 2016 4:23 pm
Location: Belgium

Re: Noise, Firecapture and imaging questions

Post by PDB » Tue Jan 09, 2018 8:54 am

Hi Alexandra,

must be some flaw in some program. It would be interesting to see the header of the SER file. There seems to be some wrong magic somewhere with bit depth in SER files. (don't throw away the file yet, i could analyze it on my machine, but it is probably to big to email?)

Paul

User avatar
Montana
Way More Fun to Share It!!
Way More Fun to Share It!!
Posts: 18968
Joined: Mon Oct 17, 2011 5:25 pm
Location: Cheshire, UK

Re: Noise, Firecapture and imaging questions

Post by Montana » Tue Jan 09, 2018 10:05 am

Thanks Paul, the file is 17GB so even too big for WeTransfer. How do I check the header of the ser file?
I thought I had cracked it as the 50% that show 14 bit now process like a dream ASK3, absolutely noiseless and beautiful sharpness. But the 50% that now show this curious 15bit won't process on either setting in ASK3 :(
Alexandra

User avatar
PDB
Im an EXPERT!
Im an EXPERT!
Posts: 293
Joined: Mon Aug 15, 2016 4:23 pm
Location: Belgium

Re: Noise, Firecapture and imaging questions

Post by PDB » Tue Jan 09, 2018 10:20 am

Think you can look at the header in SER player. (But that is of course the way SER player interprets this) To really see the contents, you should need to view it an hexviewer. Don't think there are any standard tools for that on Windows. There have been changes between SER version 1 and SER version 2, but need to dig back in my old information. If the header really says 15 bit, then I wonder how the capture software decides that. (should be either 14 or 16 but 15?) Did you capture that with FireCapture?

Rgrds,

Paul

User avatar
Montana
Way More Fun to Share It!!
Way More Fun to Share It!!
Posts: 18968
Joined: Mon Oct 17, 2011 5:25 pm
Location: Cheshire, UK

Re: Noise, Firecapture and imaging questions

Post by Montana » Tue Jan 09, 2018 1:38 pm

Yes, Firecapture 2.5
I must have a look at all my other files that v2.3 does, it will be interesting to compare
Alexandra

dvoss
The Sun?
The Sun?
Posts: 12
Joined: Wed Aug 30, 2017 2:50 pm
Location: Silver Spring MD USA

Re: Noise, Firecapture and imaging questions

Post by dvoss » Tue Jan 09, 2018 2:44 pm

Alexandra, would processing in PIPP help? In addition to accepting what the SER header indicates, PIPP has a 15 bit input option.

David

User avatar
marktownley
Librarian
Librarian
Posts: 23509
Joined: Tue Oct 18, 2011 5:27 pm
Location: Brierley Hills, UK
Contact:

Re: Noise, Firecapture and imaging questions

Post by marktownley » Tue Jan 09, 2018 5:52 pm

PDB wrote:
Sun Jan 07, 2018 5:59 pm
This is triggered by the "High Speed" setting in FireCapture (Highspeed marked = 10 bit, if not 12 bit ADC) In Genika i think is set by Mode 7 in the camera settings.

P.
Hi Paul,

If you get a minute can you screen shot me this for me in firecapture, I have looked but I can't find it. I'm usually being blind and it is under my nose so apologies if it is squarely in view :D

I did try going away from my usual 8bit recording in some trials last night, but if I tick 16bit recording option in FCv2.5 it outputs a 16bit ser file, but all this seems to be is an upsampled 8bit file, not a true 16 bit file. I can't get it output a 10bit ser with the 236m blackfly or a 12bit with the PGR CH3....
Thanks

Mark
Image
http://brierleyhillsolar.blogspot.co.uk/
Solar images, a collection of all the most up to date live solar data on the web, imaging & processing tutorials - please take a look!

Post Reply

Who is online

Users browsing this forum: No registered users and 4 guests