›

Newbie Question - from Fasttrack Training

Matt Frepp
in PixInsight
Hi,
the new Fasttrack Training videos use FITS files. The old ones used RAW files. Does that mean that I cannot use RAW files anymore? Or is there a possibility to convert them? Obviously the RAW files from my DSLR camera don't have any astrometric data. Is there a way to get it? I would know the object I was aiming at, and the date and time and location.

thanks
Matt

Comments

  • BlockHeadBlockHead
    edited February 2024
    Matt,

    You should set PixInsight to load DSLR files (e.g. Canon CR3 files) as Pure RAW.
    Then you have the same situation as demonstrated in FastTrack Training.

    Please confirm you agree with the above sentence.

    As to the astrometric data, your statement is not entirely true. It is possible with some software to use a DSLR camera and save FITs files with the coordinates and plate scale values needed for a plate solution by ImageSolver in PixInsight. Maxim Dl, for example, does this. However, if you are not *acquiring* your data using an astronomy software program and just downloading images from the camera- then yes you are correct. However in FastTrack Training I show the manual method of plate solution since I have viewers use WBPP in a calibrate only mode.  This means that at the end of StarAlignment none of the images have been plate solved. The difference between FastTrack now and your DSLR data is that the FastTrack data came with coordinates and a pixel size/focal length that is used by ImageSolver. In you case you would input the coordinates and focal length/pixel size by hand since you do not have the information in the files. As you mention.. you do know the information that is needed. 

    The issue isn't the fact the images are DSLR... the issue is that you are (presumably) not using a capture software that writes the metadata to a FITs file.

    Please let me know if this is accurate information.
    -the Blockhead
  • Matt Frepp
    thanks Adam. Yes it's entirely correct.

    My Nikon does not capture astrometric data. So are you saying that Maxim DI could figure out the astrometric data of my raw images by "looking" at the image? Plus some input I would be able to provide, like shooting-location, date/time, target I was selecting? Which I could then enter into ImageSolver?

    Not entirely clear on that point. Would appreciate if my assumption is correct. Much appreciated.
    Matt

  • BlockHeadBlockHead
    Sorry Matt, I think are you missing my point. 

    An image capture software will put information into the header that allows ImageSolver to do its job.

    1. Coordinates
    2. Pixel Size
    3. Focal Length

    This is the most common way astronomical data is acquired. In and of itself, this is not astrometric data- though it is used to compute it so that you can assign a coordinate to each pixel in your image.

    I believe you are telling me you do not use a program to connect to a telescope and acquire the images. Thus you do not have 1-3 in your images. So ImageSolver cannot do its job automatically. You can manually provide ImageSolver 1-3.  Above you regard this information as "plus" .. but we disagree. 1-3 is the important information. A "plate solution" is what ImageSolver spits out... 1-3 are the input information.

    So you have the information you need and you just need to input into ImageSolver at the time you need a plate solution. By the way, this is the meaning of the WBPP setting (in Lights under Astrometric Solution) of "Interactive in case of Failure" ...the FAILURE is not providing items 1-3 and the interactive is manually inputting the information so that WBPP can proceed. 

    -the Blockhead
  • BlockHeadBlockHead
    Let me add that I explain many aspects of this in different videos.

    1. I explain how to calculate your plate scale. 
    2. I explain how ImageSolver works 
    3. In WBPP I explain these settings and behavior.
    4. I believe in the WORKFLOW example of DSLR images (exactly what you are doing) I demonstrate there as well. 

    -the Blockhead
  • Matt Frepp
    sorry for any confusion. Referring to your 1-3, I have 2-3, but not 1. And yes, I load my RAW files directly into PI. I shoot them with my 135mm and 500mm lenses and my Vixen S103 telescope, but it does not provide astrometric data either. I'm missing something here that could capture it.

    I watched the videos 1-6 and 10 in Fasttrack but could not find any reference to raw-files with missing astrometric data. If that is contained in another video, could you please let me know which one(s)? I have bought PI Fundamentals, hopefully it's somewhere there.

    thanks again!
    Matt

  • BlockHeadBlockHead
    But you do have coordinates since you know what you are taking a picture of right?

    Please watch the Cassiopeia example int he WORKFLOW section:

    I have not looked at it in a while.. but I think it will be helpful.
    It was made before SPCC...so it likely shows PCC...but the idea is exactly the same.

    More deeply...I explain more in the first part of the SPCC videos:

    Please watch and let me know if these are on target. The one above is indeed on the Fundamentals path.

    -the Blockhead
  • Matt Frepp
    thanks! Will do and let you know.
    Matt

  • Matt Frepp
    I watched the video in the second link, and it is really helpful, but again you start out with FITS files that already have the coordinates of the image-center. My Nikon does not provide that. So I guess the only way is to ensure that the object I'm aiming at is more or less in the center, and then later use Stellarium to get the coordinates; as I know the place and date-time when the images were taken, On a 2 hours exposure, I'd take the coordinates that applied after 1 hour of shooting. Just not sure if this is accurate enough, but experience will show.

    thanks again
    Matt

  • BlockHeadBlockHead
    Sorry, concerning the links..it is the FIRST link that is exactly your situation. Did you watch it??

    The coordinates just need to be in the field of view and the focal length needs to be within a factor of two. (However, I would say you really want to be within 20% of the real value).

    -the Blockhead
  • Matt Frepp
    Yeah I watched Cassiopeia Part 1+2. I see it's back to the previous PCC, not SPCC. That's fine, thanks.
    Good idea to shoot this with a 35mm lens. I never thought of that. Will try it.
    Clear skies!
  • BlockHeadBlockHead
    Yeah the PCC isn't material... the issue remains concerning ImageSolver..and the solution is the same.
    -the Blockhead
  • Matt Frepp
    Hi Adam, I think I managed to enter astrometric data through ImageSolver into my DSLR Integrated Image, but running SPCC now gives me the error below. Any idea? Much appreciated. thanks, Matt

    *** Error: Invalid Gaia DR3/SP XPSD server execution: Couldn't retrieve a valid spectrum wavelength table.

    Reading swap files...

    1793.532 MiB/s

    <* failed *>



  • Matt Frepp
    Hi Adam, never mind above, was some issues on my computer, it's fixed now. I could run SPCC but the result is rather disappointing, IC405 looks like a mono-image. Does PI "eats" colors somewhere in the calibration, registration or integration process, or could I have forgotten to tick an important field? I did get debayered images though. Any feedback appreciated. Thanks, Matt
  • BlockHeadBlockHead
    Hi Matt,

    Can you provide some screenshots and describe the issue? 
    Just so we are on the same page, you are using the data provided correct?

    I don't want to guess until there is a bit more information.
    Thanks,
    -the Blockhead
  • Matt Frepp
    thanks for your quick response Adam, as always, much appreciated. I might have messed up, I'll start over with the DSLR images I have. I'll post the integrated image before and after SPCC, hopefully that gives you an idea.

    Was kind of hoping SPCC is something like a "color-me" function, that it would color even a mono-image based on the astrometric data provided, but I guess there's a bit more involved at this stage.

    I'll do my best to explain my issue though. Give me a day or two. Again, thanks a lot for all your help.
    Matt
  • Matt Frepp
    I think the main problem is that on my OSC light frames the colors are too faint, probably insufficient exposure. SPCC won't or can't do  much about it. So my only question is about the ImageSolver. I opened it, clicked Search, entered IC405, and it came back with RA and DEC values, which I don't understand. In Stellarium they are different when I look up IC405 on that date, time and location I was shooting from.

    Ignoring this discrepancy, I entered longitude, latitude, time (UTC I assume) from where I was, plus focal length and pixel size of my camera. Hoping that PI has all info it needs I clicked ok. It took a while but it finally completed and updated the FITS header.

    Question - have I done this correctly? IC405 is quite in the center of the image, confirmed by astrometry.net.

    thanks
    Matt

    Image Solver.jpg
    694 x 764 - 115K
  • BlockHeadBlockHead
    Just to be clear.. SPCC will not make colors "appear" or brighter. It will only make certain your data is calibrated and not having a bias in color because of the sensor/telescope itself. That is the point. You are right though, if you have not detected much color... than not much color will be present. This actually means SPCC is doing its job.

    Regarding IC 405, the date, time and location (this last has no effect) will not matter in any significant way when it comes to large astronomical objects like galaxies and nebulae. Precession does affect the coordinates...but you need to be off by hundreds of years to start to see anything that is going to even somewhat catch your attention.

    The fact that the coordinates are different is because it depends on the catalogue that is being used between the two programs. Large extended objects do not have well defined coordinates...STARS do. If you want to be precise you can look up a star in the frame instead of the large object. This is a more advanced method... and usually totally unnecessary.

    The coordinates I found on-line differ from the ones you have populated in your screenshot by less than 3 arcminutes. This is good enough...because  I know your field of view is much much greater than 3 arcminutes. I would put good money on that. lol

    Yes, you did everything correctly. Please give some thought to the ideas I mention above.

    -the Blockhead




  • Matt Frepp
    thanks a lot for your detailed response Adam. Glad to hear that I'm slowly, very slowly, getting ahead. I have a more realistic view of SPCC now, and what you explain probably applies to PCC and standard color calibration as well.

    So in order to bring out more colors in nebulae is longer exposure? I think - but could you confirm? - that 50 light frames shot at 5 minutes each is better than 250 light frames shot at 1 minute each. Both are exposed for 250 minutes, but I'd think that very faint signals are simply not caught in 1 minute, but maybe in 5?? But ok, this is way off-topic to PI, but you can probably tell from top of your head.

    Glad I purchased the fundamentals. I just wish I had more time, both to go out at night and for post-processing. Looking forward to my retirement in 5 months. :-)

    thanks again, Matt
Sign In or Register to comment.