›

Mono Luminance + OSC RGB Dual Narrowband

Stickfarm8
in General
Adam,

So I've gotten thru most of the Fundamentals videos - they were so good. I started on the Narrowband series last week. I wish I had watched them when I first got back into the hobby last Spring - I might have made different decisions on the camera. Since I had been imaging with various DLSR's when I made the decision to get a true astrophotography camera, it seemed like a logical choice. I was aware of the issues between the two camps - mono vs OSC - but thought the OSC was the best choice given budget constraints, what I was used to dealing with, and so went with the ASI2600 MC Pro. I like everything about the camera in the 9 months I have had it. Especially the low noise and the overall image quality.

I had been looking into the mono option from what I had learned in the Fundamentals. But after watching the first three videos in the Narrowband series, I now better have a understand some of the drawbacks to OSC if one wants to really get into narrowband imaging. I realize its all doable with OSC but the processing is more complicated, and long term there are many advantages to transitioning to a mono system.

After talking with my CFO here at home, the budget won't allow for the full transition to camera + filter wheel + filters - and least not the higher end Baeder or Astrodon filters I have been considering if I transitioned. The seven filters add up to quite a bit coupled with ZWO seven unit filter wheel. But I could get the ASI2600MM (mono version) of the OSC camera I have, and then luminance channel to the images.

My question is - will getting the mono luminance images with the ASI2600 MM camera, and processing them into the OSC dual narrowband images (I have a Ha+OIII and SII+OIII filters already) add anything significant to the overall image or would a better path be to wait until I can make the complete transition? I understand I will would still have to deal with the color image processing issues for the time being.

I thought it was interesting in one of the Cloudy Night threads, Alex McConahay  said he seen a lot of people go from OSC to mono but hasn't seen many go the other way.

Your thoughts would be appreciated sir.

Rich


Comments

  • BlockHeadBlockHead
    Hi Rich,

    Unfiltered mono is such a different kind of data in terms of brightness and wavelength- that my intuition says it doesn't make sense to combine unfiltered (or broadband)  data with NB data. 

    It is kinda of like the following. You take a picture with your DSLR camera and the image you get is complete...but no colors- just gray scale. Then, using some other camera, you take images of the same scene in RED and TEAL. And that is it. Finally you will want to color the scene. 

    Your scene has a picture of a yellow ball and a purple dinosaur. Your NB data is only RED and TEAL. 
    You cannot color the scene to capture the yellow ball or the purple dinosaur... you do not have these wavelengths. But you ask... can't you add the greyscale to your NB data? They will just lose contrast. The entire idea behind NB images is that the SKY is D A R K compared to the object. By adding in the greyscale image with the bright sky... you defeat the NB data. You might as well just have taken a regular color image.

    Now, if you can find some examples of people doing this I would be interested to learn what the benefits are. I cannot think of any at the moment.

    -the Blockhead 
  • Stickfarm8
    Adam,

    Your explanation certainly makes sense. 

    There are several threads in the PI forum and Cloudy Nights from folks that claimed it was not an issue. 

    I'm going to talk with "my CFO" and see if we can make this transition fit without the filter wheel. That alone was $400 for the ZWO 7 element one at Highpoint. I have a ZWO filter drawer now that I use with the OSC ASI2600MC Pro, and could do manual changes. At least for a while. Certainly not ideal.

    Thanks Adam for the response.

    Rich
  • Donald Meyer
    Hello Adam and Rich,

    I have a followup question. I notice quite a few people are posting nice images on Astrobin created from a combination of NB images (HaOiii and SiiOiii) plus UVIR images, all using an OSC camera. As I've worked my way through Adam's Fundamentals and NB Fastrack videos, I've started processing some of my own NB + UVIR OSC data obtained in a similar fashion. I like Adam's technique of creating a pseudo-LRGB image from the NB data. I just don't know how people are incorporating the OSC UVIR data into this process. I've tried creating 2 separate images: one psuedo-LRGB from the NB data, and one more standard processing of the UVIR data. I have then played with combining the two using ImageBlend and/or Photoshop. Is there a better, more systematic approach for this type of OSC processing? My apologies ahead of time Adam if you've already addressed this in one of your videos. 

    Thank you,

    Joe
  • BlockHeadBlockHead
    Joe,
    That sounds like the sensible approach to me. 
    Of course the nuance is all in exactly how you blend the images together.
    Simply screening them might get you partway there- but likely isn't the best you can do.

    -the Blockhead
  • Donald Meyer
    Thanks Adam. That's reassuring to hear. I'll keep working through your videos and fiddling with my data. It's making more and more sense as I go.

    Joe
  • Stickfarm8
    Hey Adam and Joe,

    Very interesting to read Joe's question and your response. 

    After working thru most of the NB videos, your response made even more sense. I still keep digging thru some posts in the other forums (mainly CN) on the luminance to OSC process - which I know is a different process than what Joe is working with. I've used your pseudo-LRGB on my M51 image and it really helped the image. 

    I reactivated my Cloudy Nights account so I could post some of the images for community feedback. I had joined CN back in 2019 but never posted any images because my product thru PS was just not good (i.e., awful). Your videos have really opened the hobby up for me - they have just been wonderful. The feedback (so far) on the two images has been nice confidence boosters - and your training videos get all the credit.

    I'm still looking for a way to make the change to mono in steps, if at all possible. 

    One of the more experience gentleman posted an image of M51 that he had captured with as ASI2600MC-Pro - same as I use. He reposted after shooting with a mono camera and a different scope. He added the luminance data to the OSC and it sure added to the image.


    Unfortunately he uses PS to do this - I'd hate to open that box again. However I would really like to hear your opinion on the approach and what you thought of the second image he posted.

    BTW - the new website looks fantastic. I can't wait until you get us transitioned to the new site. 

    Thanks Adam!
  • BlockHeadBlockHead
    Using OSC data to color any luminance (from any source) is a perfectly fine thing to do.
    So the principle is solid.
    Some caveats are that you need to be certain that the color information is really good S/N- (not noisy) or..you will just make a mess.

    Part of the process will be to register the OSC data with the other luminance images you have. There are a number of ways to do this- but likely registering the masterlights with the luminance as the reference is good.

    Finally when you get to the point of blending the Luminance and the color data- you can use ImageBlend in the way I show in the M83 Workflow example. In addition, I am currently making videos for a small ImageBlend course and I will demonstrate another LRGB blending example there. ImageBlend *IS* how Photoshop is brought into PixInsight in a simple way.

    -the Blockhead
  • BlockHeadBlockHead
    Oh... and regarding the image.

    At first glance I would say that:

    1. The color information is actually more "powerful" than the luminance. What happens is if you boost the color- you will get blotches of solid color that overwhelm the subtle differences in brightness. There is a balance and this image went a bit to far tipping that balance.
    2. The contrast is very high and the background is very dark/black/clipped. l
    3. One of the important ways to measure the quality of an image is to look at the image at 200%. If it looks *reasonable* at this magnification..then it will look fine at every other. However...in today's world... many people process for the small screen from the beginning. They process the image to look good only at the size they intend to display it. However... artifacts are easy to see and become distracting when doing this. That is the case for this image. There are lots of artifacts in terms of detail because of some of the processing. If you just look at it superficially it is fine. But a more critical look reveals some processing stuff that I think could be mitigated. I think what I am talking about would be more than evident if you zoom in a bit... not even to 100% the intrinsic scale.

    -the Blockhead
  • Stickfarm8
    Adam,

    As usual - thanks for the reply and all the detail. 

    The image blend course would be fantastic sir !!

    I meant to include the links to the two images I posted in CN.  

    I know you are busy with the new website - but if you have time your feedback would be most welcomed. I know I've still got a LOT to learn - but it is coming faster..

    Hope the student didn't embarrass the teacher!

    Rich
Sign In or Register to comment.