Go Back   IceInSpace > Beginners Start Here > Beginners Astrophotography

Reply
 
Thread Tools Rate Thread
  #1  
Old 03-02-2019, 08:19 PM
mynameiscd's Avatar
mynameiscd (Andy)
Registered User

mynameiscd is offline
 
Join Date: Jan 2017
Location: Langkoop, Victoria
Posts: 250
NGC 2264 stetched to the limit

I think I've found the limit of my camera.
I've taken this over a few nights for a total of 3.5 hours and had to stretch and stretch until I can see the nebula but start to see strange artifacts like bars in my images.
After a bit of processing like saturation etc the bars are a yellow orange colour.
At first i thought it was from a mistake I made in DSS but after looking at individual subs which are over stretched and you can see them as well. I had a look at old data and after stetching them to the max you can see the horizontal stripes through each image.
I've come to the conclusion that I'm looking at the sensor and you cant squeeze blood out of a stone!
I think with the 450d I have to try brighter objects or get guiding happening and try longer subs or go down the rabbit hole and upgrade.
Anyway heres the image.
60◊60sec 400iso
60x60sec 800iso
60x90sec 800iso
Darks, flats, and bias subtracted
Stacked in DSS
Overstretched in PS

Cheers
Andy
Attached Thumbnails
Click for full-size image (Cone-Nebula-001.jpg)
178.5 KB109 views
Reply With Quote
  #2  
Old 03-02-2019, 08:46 PM
raymo
Registered User

raymo is offline
 
Join Date: Sep 2011
Location: margaret river, western australia
Posts: 4,875
Great effort Andy. May I suggest that you combine a night with a reasonably
low ambient temp and ISO 1600. Most lower range Canons handle 1600
quite well. Almost every image of mine that you may have seen here was taken at 1600, even though my 600Ds sweet spot is 800. Some of mine are taken at 3200, so don't be afraid to up the ISO a bit.
raymo
Reply With Quote
  #3  
Old 04-02-2019, 06:53 AM
mynameiscd's Avatar
mynameiscd (Andy)
Registered User

mynameiscd is offline
 
Join Date: Jan 2017
Location: Langkoop, Victoria
Posts: 250
Thanks Raymo,
I'll try 1600 a go tonight (Monday) as its going down to 9į overnight and skippy sky looks good.
Its just up to me and how tired I am after work.
The sweet spot for my camera seems to be 400 ISO so 1600 might be a bit grainy but then again processing might hide it.
Cheers
Andy
Reply With Quote
  #4  
Old 04-02-2019, 10:42 AM
sil's Avatar
sil (Steve)
Not even a speck of dust

sil is offline
 
Join Date: Jun 2012
Location: Canberra
Posts: 1,265
indeed the bars may be the sensor, people dont often realise the sensor is a highly polished metallic rectangle, ie a mirror and photons dont hit the sensor and vanish, they do reflect back and get re-reflected back from the last lens element in the optical train (glass is reflective too not just transparent). Depending on the shape of that final glass surface the reflection of the sensor can be distorted or even highly magnified. It's often difficult to try to see this reflection as its usually lost in the photo. So yes you may be at about the limit of your camera but I dont think anywhere near the limit of the imaging by a long way.

There are tutorials online for enhancing nebulosity without affecting the stars. With photoshop I doubt it can be done effectively but it should be doable a bit. A basic method is to take the L channel of your colour image, remove the stars, and apply the result as a mask to the rgb image so you can stretch the nebulosity only or at least mostly, without stretching the background too. I'd guess you are stretch the entire image uniformly, but being more selective lets you enhance and suppress the specific areas that need work. The end goal is basically to enhance the contrast between the interesting features and the background emptiness without bringing all the noise back to swamp everything or blowing out the stars. I do most of my processing in PixInsight and use a similar approach, its a matter of thinking how to solve the problem and understand which tools you can use in combination to achieve that.

If you're willing to bundle up your subs I could chuck them through AstroPixelProcessor which does a great job not too far from what I can do with PI and see what it comes out with, at least it'll give a good indication if your processing is close to the limit or its time to step away from DSS and PS. The offers there if you like, I'm keen to see and suspect the image can be much improved with the data you captured.
Reply With Quote
  #5  
Old 04-02-2019, 06:25 PM
mynameiscd's Avatar
mynameiscd (Andy)
Registered User

mynameiscd is offline
 
Join Date: Jan 2017
Location: Langkoop, Victoria
Posts: 250
Thanks Steve for a heap of info.
Thanks also for offering to have a go with my data and help me out.
I just need to find somewhere to upload the files. Ive got Google drive but im not quite sure on how to share files.
Just got home from work so i might have a bit of time later on tonight.
Cheers
Andy
Reply With Quote
  #6  
Old 04-02-2019, 08:06 PM
RyanJones
Registered User

RyanJones is offline
 
Join Date: Jun 2018
Location: Melbourne,Australia
Posts: 627
My effort on the same target Andy. There's a reason why I didn't post mine
To be fair I got a little way into processing then decided that it wasn't good enough data. That's when I started " playing " with what I had.

Turns out this target is a little more difficult that it looks like it should be.

Good luck with yours and I'll have another go again soon too.
Attached Thumbnails
Click for full-size image (image.jpeg)
157.8 KB30 views
Reply With Quote
  #7  
Old 04-02-2019, 08:14 PM
rcheshire's Avatar
rcheshire (Rowland)
Registered User

rcheshire is offline
 
Join Date: Apr 2010
Location: Geelong
Posts: 2,580
Beautiful target but probably more data in there. This would be fun to process if you decide to upload.
Reply With Quote
  #8  
Old 04-02-2019, 08:40 PM
Michael_A
Registered User

Michael_A is offline
 
Join Date: Jan 2019
Location: Brisbane, Australia
Posts: 3
Hi Andy, I reckon thatís a great image and is what Iím trying to achieve with my new telescope. Iíve just purchased a Celestron NexStar Evolution 6Ē SCT and have an Olympus Micro Four Thirds Camera. Any tips? I think I might have to swap out the relatively heavy duty alt-azimuth mount that it came with and noticed most people here are using some sort of equatorial mount?
Reply With Quote
  #9  
Old 04-02-2019, 09:33 PM
RB's Avatar
RB (Andrew)
Moderator

RB is offline
 
Join Date: Aug 2005
Posts: 21,605
Quote:
Originally Posted by Michael_A View Post
Hi Andy, I reckon thatís a great image and is what Iím trying to achieve with my new telescope. Iíve just purchased a Celestron NexStar Evolution 6Ē SCT and have an Olympus Micro Four Thirds Camera. Any tips? I think I might have to swap out the relatively heavy duty alt-azimuth mount that it came with and noticed most people here are using some sort of equatorial mount?
Michael, instead of asking for advice in other people's threads, it would be better if you started your own thread asking for help and advice in the Beginners forum.
That way people can respond to your questions and you can ask whatever relevant questions you may have.

RB
Reply With Quote
  #10  
Old 05-02-2019, 09:26 AM
mynameiscd's Avatar
mynameiscd (Andy)
Registered User

mynameiscd is offline
 
Join Date: Jan 2017
Location: Langkoop, Victoria
Posts: 250
Thanks Ryan, Rowland, Michael, and RB.
Work is interfering with my astrophotography.
I was going to image last night but I was so tired and having to be up at 5:30 I just couldn't do it. Maybe tonight.
I've collected all the data and it's about 4.1 gig.
I'll zip it up into manageable pieces and upload to Google drive tonight.
Cheers all and I'd better get back to work.
Andy
Reply With Quote
  #11  
Old 05-02-2019, 11:45 AM
xelasnave's Avatar
xelasnave
Gravity does not Suck

xelasnave is offline
 
Join Date: Mar 2005
Location: Tabulam
Posts: 11,855
Andy that is a great effort.
Work is where you go to catch up on sleep☺.
Alex
Reply With Quote
  #12  
Old 09-02-2019, 03:27 AM
sil's Avatar
sil (Steve)
Not even a speck of dust

sil is offline
 
Join Date: Jun 2012
Location: Canberra
Posts: 1,265
Andy took me up on my offer and provided subs for me to chuck into APP.


This is NOT a "my processing is better than yours" exercise. Andy is at crossroads of where to improve and hopefully I can help him decide. My original thought was APP could greatly improve the nebulosity. More like its shown how badly I understood the target brightness.


So the piccies. Attached are three pics.



First pic is a rotated and cropped image to match original posted pic. This image was processed from Andy's first set of subs he provided, using AstroPixelProcessor with its default settings except using sigma rejection. Its the default autostretch when you view a file (in this case the stacked integrated image) and then saved to JPG. No other processing except resizing to get small enough filesize to upload here. Lights, flats, darks, bias were used.


Second pic is same result autocorrected in photo editing program.


Third pic is source FITS from APP processed in Pixinsight with only a histogram stretch just to show how much nebulosity and structure is hiding in the data still.


My thoughts are I was hoping by default to see more nebulosity than I got but still i think APP with defaults did a great job of producing a cleaner result to work with than DSS. There are more and better options for processing and noise reduction during the regiostration and integration process available before you can start post processing. I find for my needs the default autostretch of APP gives me a great result with good colour but if I want the best I can do I can take the fits file into PixInsight but usually I prefer the whole workflow in PI myself and use APP more to do a quick process of a new data set to make sure I got what I was after and its worth the time and effort of PI.


Interesting the ugly streaking didnt really exagerrate at all so I think its more from a source out of shot rasther than sensor reflection so perhaps a shroud is neededto keep from optical element in shadow?
I dont think noise is a major problem with the cameraeither as the unfeasable stretch in PI still shows fairly clean nebulosity even though other noise is vastly obvious. To me this shows with some masking techniques and star reduction there is room to bring more nebulosity structure and colour forward which will be tricky but doable.


My conclusion is processing is where improvement is most likely, though of course better gear will help as always but then you have the processing limitation still.


Anyway its 3:25am and I still have more of Andy's data to play with and respond to him with but for now wanted initial thoughts re original post put down before i forget them. Discuss ?


steve
Attached Thumbnails
Click for full-size image (matched0_0.25.jpg)
141.3 KB28 views
Click for full-size image (matched_auto0.25.jpg)
162.9 KB28 views
Click for full-size image (St_avg_5400_0s_SC_1_3_0_none_x_1_0_LZ3_NS_full_eq_add_sc_BWMV_nor_AAD_RL_noMBB_St_clone0.25.jpg)
172.5 KB33 views
Reply With Quote
  #13  
Old 09-02-2019, 10:30 AM
mynameiscd's Avatar
mynameiscd (Andy)
Registered User

mynameiscd is offline
 
Join Date: Jan 2017
Location: Langkoop, Victoria
Posts: 250
Thanks so much Steve.
Ill have a good look tonight when i get home.
Cheers
Andy
Reply With Quote
  #14  
Old 09-02-2019, 02:35 PM
RyanJones
Registered User

RyanJones is offline
 
Join Date: Jun 2018
Location: Melbourne,Australia
Posts: 627
Great comparison Steve.

It really is amazing what experienced processors can bring out in an image. To be honest I'm not surprised to see what PI was capable of. I've used a trial version and it is amazing dispite not being overly intuitive. I think it's going to have to go back on my list.

Cudos to both you and Andy for getting together on this one and showing the rest of us what is possible. Nice work gents
Reply With Quote
  #15  
Old 10-02-2019, 08:59 AM
rcheshire's Avatar
rcheshire (Rowland)
Registered User

rcheshire is offline
 
Join Date: Apr 2010
Location: Geelong
Posts: 2,580
Hi Andy.

Curious to see if the image can be improved.

There are a few ways to tackle the image set that might help.

I don't know precisely the breakdown of bias darks etc, so reading between the lines...

DSLR data suffers badly when bias and darks are used together to calibrate lights. PI lets you handle this a little better with dark optimisation but it's not always successful with DSLR images.

Create master frames first but don't bias subtract the darks - just stack them.

Load your master dark, master flat (no bias frames) and light frames (3 for a dry run). The bias is in the dark so you get bias subtraction anyway.

Last edited by rcheshire; 11-02-2019 at 08:10 PM.
Reply With Quote
  #16  
Old 10-02-2019, 12:41 PM
mynameiscd's Avatar
mynameiscd (Andy)
Registered User

mynameiscd is offline
 
Join Date: Jan 2017
Location: Langkoop, Victoria
Posts: 250
Thanks again Sterve,
Ive finally had a good look and you did wonders to my data. Like you said you have more control on how the data is interpreted so you have created another learning curve for me which i dont mind one bit
Looks like moving on and trying PI and APP and learning a new workflow.
Thanks for your input Rowland and Ryan even though Steve did all the hard work.
Rowland, this is my work flow and what data Steve had to work with and I'll upload for you as well and you can have a play.

3 x different nights. Each data set has lights, darks, flats and bias accordingly for each ISO.
I stacked each individual nights data into three separate images and then stacked the three together.
It now makes sense that the sum of all the images can increase the noise dramatically.
I might have a go later tonight with the method that Rowland suggests.
I'll send a PM to you Rowland with the link to my data.
I'll put all my data up for public when i get my head around how Google Drive works and how you can set permissions etc.
Cheers
Andy
Reply With Quote
  #17  
Old 10-02-2019, 03:59 PM
sil's Avatar
sil (Steve)
Not even a speck of dust

sil is offline
 
Join Date: Jun 2012
Location: Canberra
Posts: 1,265
Ok final image attached below. This was All of Andy's images processed at once by APP with its process. The integrated master FITS I then put into PixInsight to enhance nebulosity and reduce the stars. I didn't crop and rotate to match Andy's original pic.


With AstroPixelProcessor I changed to my usual settings with regards to processing iterations and pixel rejection. I'm no expert with APP and it doesnt have the fine control of PixInsight so its still close to the program defaults and i dont know What its internal process does but it can take all subs in multiple sets and go through its process with them all. As I've said before its more a step up from DSS in that its closer to a one click program without the learning curve or complexity of the other astrophotography programs like PixInsight, Nebulosity, AstroArt etc. More a bridge program and I find it invaluable for quickly throwing subs together to make sure i got my target and feel out the processing potential before i lose a week or two in PixInsight. My experience is that a properly processed image in PixInsight is only a little better than what APP produces more simply by default. Colour calibration is very good too in the result.


The little work I did on this image in PI was pretty minimal as the steps require good attention to detail and adjustments VERY specific to the image being worked on; in other words they are not settings apply to another image as it will have unique strengths and weakness that need to be taken into consideration. The star removal is overdone while the nebula stretch is underdone. As it is it was several hours instead of days work to do those steps.


Andy's lights were varying exposure durations and iso settings, but APP just swallowed them up no problem and its result is pretty damn good I feel. I also chose the "mosaic" registration mode so you can see jagged edges in this last pic. Its a great feature of APP and I use APP instead of the BPP script in PI to get me a master integration file to stretch. One of the sets Andy supplied was upside down compared to the other two and APP just figured it out and dealt with it.


I guess my nose is brown enough so no more package name dropping. I'm no expert in any of this but with a disability my time and efforts mean much to me so I like to try to use my hardware and software at the limit of what it can achieve since I dont havemoney to throw at new gear. So my opinion here with Andy's query is complicated. I think he has yet to find the limits of his camera and finding those will help his images improve. Of course a newer camera will help too, it always will but then you have its limits to find.I think I've shown there is processing room for improvement too. DSS isn't built for great stretching/post processing and neither is Photoshop really. And I think its time for Andy to move beyond those. I know some people can make them sing and dance but I've no idea how, so maybe Andy will work it out? More subs will also help regardless and lets not argue that here (its been done and overdone elsewhere many times) and while its a matter of diminishing returns, technically it still makes a measurable if not noticable improvement. As with all the options its a personal decision as to what option to follow, some are easier than others and some people have better aptitude than others, So maybe others can help Andy decide and mention options I haven't figured on. I better go now and upload files for Andy to play with and examine beyond these lowres jpegs.


good luck mate!
Attached Thumbnails
Click for full-size image (enh_neb_a.jpg)
187.5 KB15 views
Reply With Quote
  #18  
Old 10-02-2019, 10:45 PM
mynameiscd's Avatar
mynameiscd (Andy)
Registered User

mynameiscd is offline
 
Join Date: Jan 2017
Location: Langkoop, Victoria
Posts: 250
Well Steve its certainly an improvement on my image.
You have pulled the nebulosity out of the data. Thanks for all your work and the encouragement along the way. It looks like i have to go back to astro processing school and understand how to manipulate data and get the most out of it like you have.
I normally just stack in DSS with no processing (i think) or stretching.
I then load the .tiff file in photoshop and stretch the image with levels and curves and a little sharpening.
You have shown me im just scratching the surface on what can be done but its back to square one for me.
Thanks for your epic posts they are a real inspiration and also informative at the same time.
I think im getting reasonable data but im about to introduce a laptop into the chain so i can get guiding together. I hope this isnt going to affect anything but at the data end is where my work,is just beginning.
I've read about your disability and I just take it for granted that i can just put my scope on my mount, hook up the camera and after alignment can start imaging.
Im not sure if you do much imaging lately but your processing is spot on.
Thanks again
Andy
Reply With Quote
  #19  
Old 11-02-2019, 05:20 AM
rcheshire's Avatar
rcheshire (Rowland)
Registered User

rcheshire is offline
 
Join Date: Apr 2010
Location: Geelong
Posts: 2,580
I had time to process the first set last night.

Step by step.

  • Inspect the individual subs - very closely aligned
  • Create master frames
  • Calibrate lights with master dark and master flat, debayer, align and stack

Point 1 seems to be the issue. Dithering between subs cleans the stack up nicely. Older model DSLR images benefit hugely from dithered subs.

Recovery

The individual sets are framed differently so effectively dithered. Long process but you could do this;
  • Align and stack the first image from each se; then
  • Do the same for each subsequent image

1, 1, 1 - 2, 2, 2 and so on - 3 subs, one from each image set. The result should be much cleaner images.

Align and stack the previously combined images.

EDIT: Longer subs will help - move the histogram right so that it detaches from the left side.

Last edited by rcheshire; 13-02-2019 at 08:44 PM.
Reply With Quote
  #20  
Old 11-02-2019, 03:45 PM
sil's Avatar
sil (Steve)
Not even a speck of dust

sil is offline
 
Join Date: Jun 2012
Location: Canberra
Posts: 1,265
Was happy to help Andy. I wont go into processing workflow, its usually different for everyone. I think people assume the camera takes the image they see in magazines and online but thats rarely the case. We are spoilt for software that happily take jpegs from different cameras and combine them to get a high bit depth integrated image. All the "must have" gear gets you is shortcuts in the journey to an end result, but if you are patient you can get close with much more modest gear. Understanding what the gear components do for the process helps with part of the entire process effecting imaging time and sometimes certain image features.

As I said there are people doing awesome stuff with DSS & PS and I'd suggest you look for turoials that use those and run through them. Even if you already know the process the tutorial covers I find some tutorials explain the steps in different ways that give me a better understanding of why they are used at that point in the workflow and what to expect as a result. Not all steps present an obvious effect at the time but may when you reach the final image.

Already your method is good, its not too different from the versions I posted. I think your colour has a cast and you get the signal forward pretty well its not far from what APP did. But neither DSS nor APP have masking capabilties I'm aware of to allow isolating of parts of the image for more focused processing like I did with PI. I never could get my head around using DSS and doubt I could do as good a job.

These programs are just toolboxes and the hard part is understanding when and how to use specific tool and if that means you need to take the image between different programs in order to achieve a step in your workflow then thats fine. I bought pretty much all the astrophotography programs several years ago, couldn't get my head around most or get them to work with my raw dslr files properly. Some expected data only as video from an imaging cam, others wanted filtered mono data etc. But that was more reflective of how images were captured at the time than a problem with the programs.

My own astrophotography journey was hindered by my understanding of the terminology used and how little of my photography knowledge helped in any way. My first steps involved figuring a workflow That let me use my camera hows I wanted and take the files in and combine them then process (stretch) them and give me a single nicer image at the end. As I read and learnt and experimented I listed the steps all along the way so that I could follow them again later and "get a picture" as I referred to it. I grouped some times together as they were done in sequence together and I could workflow the list. Over time I could slot in more groups into the workflow as my gear changed (giving me more capture options) and my understanding changed. This workflow is a living document for me and continually changes. Its not perfect or even the correct way of doing things, but it is a repeatable process that allows me to "get an image" that I'm not ashamed of. Having a workflow I find gives me something to polish too. As software improvements and new features become available I can adjust a bubble in my flow chart, making that step slightly better than it was. Over time it all adds up to a happy final image. I know I go on about APP a lot but its made a lot of my workflow irrelevant and is such an idiotproof program I dont see any point anyone ever touching DSS really, its just so dead easy to get a picture with almost no effort. Its not in the same class as PI or other established programs, its reallly more akin to DSS and has more things to play with if you want and I would say its the first program everyone should have when starting astrophotography and ditch DSS totally. Of course one is free, the other isnt.

Andy, I recommend you break everything down in the workflow of getting "an image". Eg step 1: capture, step 2 process.
Now those steps get broken down into many smaller steps, And those in turn more smaller steps. Keep going eventually every little thing listed is something that can be improved. Never consider any of them as being perfect. For me My workflow now exists in two versions: one for solar system targets and the other for widefield starry views. I originally started trying to process nebula differently to clusters differently to galaxies etc But Over time I realised the same process can be used for them all and depending on the view parts of my workflow have little or no effect to the image but the same workflow works still nd I didnt need to branch steps for various targets. So start with your capture process step, you can do things like attach hanging weights under your tripod to help dampen vibrations that may result in bad wiggly frames. With DSLR the shutter and mirror flipping inside the camera can induce a wobble, so look for a way to lock the mirror up in the camera menu. Thats two small things in your whole workflow you can polish a little. Trust each little improvement will result in an improvement at the other end, which Is why I write down and modify my workflow, I can trust it as long as I follow it. Do you take darks and flats? How can they be improved? Are your camera settings ideal or using the camera sensor to its best? You can test and work out ideal/maximum settings for your gear...You can NOT get that information online, you have to test YOUR specific gear and prcoess the captures to determine where the sweet spot i to your eye in the final image.

Like I said, break it all down to every little tiny step and look for ways to improve those. Start with a high level broad workflow of what you do and over time break them down. Look online for articles on astrophotography and see if you are missing any broad steps and then look at how the smaller stepps are broken down, are they the same or different to yours? And can you try what they do and see what difference it makes to your final image. Every change you make I think must be considered in terms of improvement to the final image you end up with. That helps you decide practical concerns such as money. Better cameras are coming out all the time and yes you can upgrade and you can be confident of better results but how much better? I am confident it will never solve all the issues you (all of you) are having. Money wont solve all the problems just help you improve some areas (not always readily apparent) but not all improvements involve money either. When you understand why you are doing certain steps you can then start research that step for ways to improve it.

Wanna join my cult? lol

If you have the willingness to learn and try new things and have some patience you will find ways to move forward in anything you do. I consider myself fairly average with astrophotography, I have limitations most of you will hopefully never have so you should all be able to do far better with little effort. Andy's post I think shows a willingness to put aside ego and explore a path to improvement in his imaging. I hope I've helped, I dont have much patience for those how expect a silver bullet solution and think they know everything already so I wouldn't put the effort in to help them.
Reply With Quote
Reply

Bookmarks

Thread Tools
Rate This Thread
Rate This Thread:

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump


All times are GMT +10. The time is now 06:18 PM.

Powered by vBulletin Version 3.8.7 | Copyright ©2000 - 2019, Jelsoft Enterprises Ltd.
Advertisement
Bintel
Advertisement
Lunatico Astronomical
Advertisement
SkyWatcher Australia
Advertisement
OzScopes Authorised Dealer
Advertisement
Meade Australia
Advertisement
Nitecore
Advertisement
NexDome Observatories
Advertisement
Astronomy and Electronics Centre
Advertisement