Thanks for posting this! APP is usually my goto for stitching mosaics. But sometimes it just refuses to work very well. This technique cracked a problem mosaic I've had sitting around for almost three years. Thanks again!
Thank you for this excellent primer. Tried my first PI mosaic this evening and it worked perfectly the first time! Had some fun with the North America nebula.
Russell, I tried your method on a 6 pane stitch of the Heart and Soul nebula. It's the first time I tried a mosaic and it turned out great!. There are absolutely no signs of seams or artifacts! I did have to play with the shrink and feather radius a little, but nothing to write home about. Thanks so much for posting the video! Dale
I have been wanting to start processing mosaics with PI and have been wondering what the workflow would be. This video was just what I needed, and the workflow worked really well for me. Thanks so much for sharing it!
I just did it with the pleiades. Same adjustments and it worked out great. Before I tried endless with PI, APP and DSS. Your method seems to be the perfect solution! Thanks man ;-)
Great tutorial Russell! I'm planning a couple mosaics if I can get the weather to cooperate before my targets disappear for the year. I'll try this method if so. Thanks for sharing!
Hi Russell. I’m thinking of upgrading from my z73 to a fra400 which is what you did I believe. Did you ever use the f3.9 reducer and was it any good? Less exposure time and a wider fov both sound nice to me. Would you say the upgrade is worth while or are the scopes too similar and it’d make more sense to go for something bigger? Maybe a fra500 or 80phq….
@@benjiwhittle Hi Ben, yes I upgraded from the z73 to the askar fra400. I absolutely love the Askar, its a fantastic scope. I haven't used the reducer with that scope as I also have a redcat 51, but I did use the reducer with the PHQ80 and I was very impressed. I made a video about that which you might find useful. here's the link if you are interested ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-TzmEwSwiUTk.html Regarding the other scopes, I've not used the FRA500, but I have heard very good things. I absolutely love the Askar 80phq, it is a brilliant telescope. If I had to pick one between the FRA400 or the 80phq (it would be a very difficult decision as they are both fantastic) I would probably just lean towards the 80phq. The stars it produces are fantastic and have a slight edge over the FRA400. The FRA400 is cheaper though and faster at f5.5 vs f7.5. I hope this helps? Any other questions, just shout.
@@RussellsAstrophotography thanks for your opinions, they are really appreciated. My other consideration is a wo gt81. The z73’s big brother. Have you had any dealings with that one!
@@benjiwhittle I’ve not no, but I know a few people who do love that scope. I think Joe at Joes astrophoto uses the GT81 and he loves his. I’m sure he’d be happy to answer any questions you might have 👍
Thanks! I have been putting off attempting a mosaic for years now but think I'm ready to go for it. Something easy for my first attempt like Andromeda. Would love to see your final mosaic image of the Orion/Horsehead area.
Thanks, yeah it was really simple. I tried lots of other tutorials including Amy’s Astros and for some reason it didn’t work for my image. This method worked well though
Another great instructive and helpful video, Russell. Mosaics are on my agenda sometime before Christmas, so I'll come back and study your workflow. Just like working up your final image, this process requires a good measure of trial and error. Your findings/settings will be a good starting point for me. Thanks a lot for sharing your knowledge and experience. Paul.
Thank you very much, glad you found it useful. I hope you get some good data on a mosaic soon. Clear skies 👍 Ps sorry for the slow response, I’ve been away on my holidays these past few weeks.
Very helpful tuttorial Russell, thank you very much. In my case I use a mono camera and have to apply the workflow to each individual filter as is appropriate. You end up with a lot of files but they all come togeher OK in the end and then you just combine in PI and process as is appropriate as usual. One small but seems to be important point to add FYI - in step 1 it's important to check Distortion Correction at the bottom of the page to minmize possible star distortion along the overlaps. Thanks again, will be following more of your videoa in the future,
Great workflow video Russell. This looks a lot easier than the method I used once in Pixinsight a while back. You mentioned about APP - I use that for mosaics and it works brilliantly and can be done all in one hit when you are calibrating and stacking the images and will spit out a calibrated, stacked and mosaic’d image at the end for each filter (it’s in my Vela Supernova remnant video). The downside is it is an extra cost to buy APP.
I own APP, it's what I first used in this hobby. Tomorrow or the next day, I am going to run my "difficult" data through it (ha, if I can remember how) and check that result. A couple of other people have also suggested APP for mosaics, which suggests to me that PixInSight needs to make the mosaic process more user friendly.
Good tutorial. We will try it. What about acquisition of these 4 frames? Are they identical? Did you crop them before stitching? Did you do DBE and color calibration before stitching?
Wow, I had tried the "Star Alignment" method from another person who posted a comment on your present one. That process I found very tedious, and it was easy to get mixed up with where you were, especially with 4 panels! By contrast, this worked the first time! I can see 1 or 2 star artifacts, so I am running it again with a higher "feathering number". (Result: still a star artifact right at the meeting point of the 4 corners) This method took minutes compared to several hours for the other process I tried. You know, the "problem" with these mosaics is the camera angles. With Telescopius/ASIAIR/PixInsight combo, it is VERY difficult to get the angles and overlaps right. So I think I should just install the camera on my RASA, shoot a preview at the target area, note the camera angle, and then put that into Telescopius and make that the reference instead of trying to match the camera to Telescopius. Even then, Telescopius shows slight variations between the panes of 1 to 2 degrees. That's too much to ask, rotating the camera out in the dark by tiny increments, what do you think? Anyway, I am amazed that you sat and tested increments on the Gradient Merger Process. This video basically saved my project, and I had tremendous difficulty getting PixInsight WBPP to be able to successfully generate my 4 panels when they were all put in together. Later: was happy with photo but I could see the lines of all 4 panels. And when I would use masks, they clearly showed the bullseye target on my image. My camera angles were tilted down top and bottom on the left side, so getting that part right is really hard. Much cropping required.
So glad you found it useful and it helped with your data. Like you I tried multiple methods and thought I wasn’t going to get it to work, then I found this approach. Yeah I’ve heard of a few reports of the telescopius coordinates being slightly out, I’ve not had that issue, however I have only done a couple of mosaics. I think the biggest challenge is getting the rotation of the camera spot on. Making a note of your camera rotation first seems like a very good idea, this makes way more sense that making those small changes trying to get it spot on…which I what I did 😂 you’ll have to let me know how that approach works. Thanks very much for watching, clear skies 👍
@@RussellsAstrophotography Yes, to match the camera to Telescopius is tedious in the dark, and after every tiny camera twist you have to reshoot a preview and then recheck the angle. It's a pain. And then, every panel needs to be about 1 degree different? Oh please! If I could afford a Planewave then perhaps I could also afford a camera rotator. As I said my left side panels were canted compared to the right side and I was lucky I could still get something after significant cropping. BTW, I believe that the "Lazy Geek" suggested to take the camera angle to Telescopius instead of vice versa, and when I first watched his video, he lost me with that. Now I see why.
This is one of the best videos for a mosaic in Pixinsight that I've found, I'm going to use this method when I process my heart nebula! Much appreciated. One question though, you did the mosaic BEFORE stretching the image and doing all your final processing? So all you did was stack your images before the mosaic?
Thank you very much, so glad you found the video useful. Yes that is correct, I did the mosaic before stretching. All I did was stack each panel. I did try it the other way as well, i.e., stretching first, but found this worked better.
@@RussellsAstrophotography Well I ended up trying both ways and you're right, before doing any stretching is the best way to go. I got much better results. With the new mosaic mode in the latest update I've opened up many more possibilities for objects bigger than my setups framing, interested to see what I can come up with! Here is my first go at a mosaic with Pixinsight, two panels... i.ibb.co/jTf7sTN/heartnebula-Merge-Mosaic-DBE4.png
Thank you very much for such a clear explanation video! However, once I use the "MosaicByCoordinates" script, the resulting registered images come totally white. I cannot get through this. I can perfectly the STF streched panels once I run "ImageSolver", but not after when they are registered. No errors appear in the process. Maybe I am missing something? Thank you!
Thanks! This is super helpful. However, I'm having an issue. I created a nice mosaic of the Heart and Soul Nebula but when I try to run SPCC on it I'm getting error messages. This happens even if I run Image Solver before going to SPCC. Any advice appreciated!!
Hi, watched some of your videos and they are very good, this one seems to be the best I have seen re mosaics however, not one to assume but in this video where is your starting point? It looks to me that you have calibrated the separate images and created a master light for each one, is that correct? Also, what process has each image undergone to get to this point, stretched, denoise, star reduction, curves etc? A small critique, this information would be helpful to a newbie like me. Thanks.
Hi Russell Great vid many thanks.I also had all sorts of trouble with mosaics if ICE didn't work. Question though I see your image is taken with a color camera.I was using an ASI2600MC .Do you have to process each pane to at least de bayer them.Just to try your method I put the unprocessed images in Pix and it did a good job but trashed the colour data.It would be great to be able to join it all together and then process it but by the look, this doesn't happen. How do you de bayer to make sure each pane matches?Be easier if you could do it at the end Many thanks Stu
Hi Stu, sorry for the slow response. If I remember correctly, I think I stacked and de-bayered all of the four panels in APP (this will work in PI too) and then combined them using the above method before stretching the final image. I hope this helps 👍
Thank you very much Dave. Yes, I usually edit them separately, I create the RGB image and edit that, then edit the Luminance data before combining. I think this is the best approach for LRGB. I always take the stars out first as well. I hope this helps 👍
Sounds awesome. I made a 16 images panorama of M31 with my 150/750 2 days ago. Just bought the license for Pixinsight because this tool is THE tool. I shoot with a ZWO 533MC Pro (Color imager). Do you first work your color astrometry on each pictures ? Or do you do after merging the mosaic ?
@@RussellsAstrophotography Hi Russ, I use a simple shade matching to eliminate the darker corners by using shade matching… ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-BFIw24tUv0c.html
hmm... I was using this and now I can not find the Mosiac sections anymore - since I purchased and upgraded to the latest version. I can not find the Scripts Image Plate Solver and Mosiac by Coordinates
@@RussellsAstrophotography I had to redownload all of it and then apply and then shutdown and reload the program. Took me way longer then utvsgoyjd have to figure that out.
Hi Russell so I tried this last night a 2x2 and although I have merged everything its not the same as it is from Telescopius. I suspect its to do with the camera rotation. When running the plan I checked the new AAP rotation function which I think is the same as plate solving and checking the degrees and then matched that within a degree of Telescopius. I am bit confused as in Telescopius there is a position angle and a equivalent to in my case 239.3 position and -120.7 equivalent. I used 238 on AAP and its miles out. Any ideas? Thanks
Have you tried using MosaicByCoordinates with drizzled data? I cannot get it to work with a two drizzled panel mosaic, because the script runs and then just exits without completing. It gets about 50% and then just stops. My output directory is set accurately and the settings are exactly like in the video here. Just curious. I also posted this question in the Pixinsight forum
@@RussellsAstrophotography No worries. Yes, the issue isn't drizzling the data, but MBC has an image size limitation. I do not know what that limit is, but I'm thinking it is approximately 10,000 pixels by 10,000 pixels. I haven't fully test it yet, and there isn't much documentation about it.
Sorry to hear this. I’m not sure why this wouldn’t be showing for you. Have you managed to solve the issue? Ps sorry for the slow response, I’ve been away on my holidays these past few weeks.
Mr Russell, I only just now noticed that in Telescopius, a default setting for camera rotation for cameras with a rectangular lenses (like our ASI2600s) requires a rotator. Well who has one of those. Without a rotator, the panels cannot be "perfectly matched". But in a mosaic, the camera rotation is critical. To get the same camera angle for each panel, did you turn that feature off? I didn't realize it was there. But then, how tolerant of not precisely matched panels is WBPP in PixInsight for mosaically registering and normalizing the frames to each other? For me, WBPP can't even get that far. Next time I will at least turn off the camera rotator option. I think my panels were off perhaps as much as 2 degrees.
Hi Paul, I didn't have that box checked no, I just tried to match it as best as I could with the rotation of my camera. I have a manual rotator built into my Askar telescope, so just eyeballed it to begin with and they had a few trial and error attempts until I got the camera rotation as close to the telescopius preview as I could. The key thing is to make sure the overlap tolerance is high enough, then you should be ok when it comes to processing the images. I hope this helps?
@@RussellsAstrophotography Yeah, as I had remarked, I plan to bump it up from 20% to 25% overlap. But today at least, I am thinking it had more to do with the couple degrees of camera rotation, which was evident when the panels were connected. It's really hard to adjust to 1 degree by hand on the RASA.
