15th Feb 2024: Astro Pixel Processor 2.0.0-beta29 released - macOS native File Chooser, macOS CMD-Q fixed, read-only Fits on network fixed and other bug fixes
7th December 2023: added payment option Alipay to purchase Astro Pixel Processor from China, Hong Kong, Macau, Taiwan, Korea, Japan and other countries where Alipay is used.
I have been interested in others' images of duo-band filters and OSC cameras used to mimic the HOO or HUB palettes. I watched a couple of tutorials online where APP was used to extract Ha and OIII. After making many mistakes and eventually correcting them ( I think), I still ended up with an error when I went to combine RBG channels.
The error said:
CombineThreeChannelFloatBufferWorker
Error Message: java.lang.ArrayIndexOutOfBoundsException: 29736366
Cause: 29726266 dB.a (Unknown Source) dB.doInBackground (Unknown Source)
Of course, I don't know what any of that means.
I carefully selected Ha/OIII extract Ha, then OIII, then mono for the 3 stacks, renormalizing each time. Each stack looked fine and looked like the kind of results (b&w) that others were getting.
I tried it with the multi-channel/multi-session boxes checked, then with them not checked. Im guessing they shouldn't be checked. I remembered to select the desired palette when I went to combine the 3 channel files. This stuff seems to work the first time for many folks. I can't figure out where I am missing a step or perhaps doing some step that I shouldn't do?
Did you save each normalized stack? So you have to use those norm stacked file...
@pswilson1, apart from what @ippiu wrote, if you go to tab 9 and select Combine RGB then you can select different formulas to combine the data. One of them is HOO 1 and another HOO 2. You can select your Ha and OIII frame (no need to use the mono frame) and, when they load, select the corresponding filter. Then click calculate and you should get the combined frame. Note that when you select another formula, you need to click the New formula button first otherwise those settings will not be used. Do you get the Java exception in one of these steps?
Wouter
@pswilson1D
Did you check "Force bayer/x-trans CFA"? This would make APP recognize these files were not monochrome.
Paul, it would really help if you give more info. Can you describe the steps that you took to get to that error for example? Did you use all three frames or only two. Please give us as much info as you can otherwise it will be impossible to help you solve this.
Thanks, Wouter
Sure.
RASA 8, ASI533 (OSC), L-extreme duo band (NB). 135 subs/120sec/unity gain/-10 cooling.
Stacked: 0 Tab: RGGB, Ha-OIII/extract Ha, extract OIII, mono. Force bayer. 3 integrations.
Tab 1: UNCHECKED Multi-Filter/Multi-session (tried with these boxes checked, led to sizing mismatch) Loaded lights, flats, dark flats, darks.
Tab 5: Advanced Normalize, checked neutralize background.
Tab 6: avg/quality, sigma rejection, 1st deg LNC/2 iterations MBB 15%
After each integration of lights/flats, etc. Tab 5 Normalization (still set to advanced). I didn't select "save normalized frames, this is the only possible error I can think of? I didn't reload the lights/flats etc, just re-selected Integrate. This at least avoided the sizing mismatches.
Tab 9 Combine RGB
Normalize "ADD" but tried all of the choices, made no difference, New formula: SHO (Hubble 2)
Added Ha, OIII, and mono as Sulpher II.
Calculate: ERROR
Any suggestions appreciated, I tried about 10 times (embarrassed to say), but I know other people are doing this successfully and not reporting any problems at all....sigh.
@pswilson1 Have you tried the suggestion I gave in my first reply of yesterday? In other words, only use the Ha and OIII frames and use the HOO 1 and/or HOO 2 formulae?
I have yet to try that. Both tutorial videos I watched specifically highlighted use of the mono extraction. Both used APP to do it or I wouldn't have even tried it. Why would they be able to make that work but I can't? I may give this a try, so thank you anyway. I suspect that I am making some basic mistake, I don't think there is anything wrong with APP.
these two I watched. One uses APP for all, which is what I tried. The other uses PixInsight for post-processing.
@pswilson1 Thanks for sharing those. Very interesting. Please let me know how the HOO formulae work for you. If that goes OK then I think that the mono frame may be the cause of the error.
Ha-Oiii mono, even Ha-Oiii color are not problem as long as we assign in luminance or RGB channels as appropriate. I combined an LRGBHOO with channel files from all 4 Ha-OIII algorithms in 5 600s images taken with STC duo narrowband filter without error. (Sorry, for using un-calibrated data)
Thanks for chiming in Kijja. I, too, have been able to combine data from several filters in many different ways in APP. I have never, however, integrated tri-band data into a mono image and tried to use that as SII.
@pswilson1 Paul, I just did exactly what you did. I took raw data taken with an L-Extreme filter and integrated them while using Ha-OIII Mono, Ha-OIII Extract Ha and Ha-OIII Extract OIII sequentially. All three integrations were created with exactly the same data and settings apart from the algorithm in tab 0. Then I went to tab 9 and chose Combine RGB. I loaded the three integrations choosing SII for the mono sub and successfully created an SHO (Hubble) 1 combination.
What version of APP are you using exactly? I did this with APP 1.083 beta but it should also with with APP 1.082.
Wouter
Well, I didn't have 1.83 downloaded, so I just did that.
Yes, I know, it works! For other people. Even though the steps I outlined work, somehow I am tripping over myself. I wish I knew which step I am messing up so bad that the Combine RGB won't work.
I will restart my computer and try with the new version, though as was pointed out, the older version should have worked.
I'll admit that I only loaded the lights, flats, darks and darkflats in tab 1 and selected the appropriate algorithm in tab 0. Everything else was left at default. Usually that's only necessary under special conditions where APP can't figure out how to process the files by itself. After loading the files in tab 1 all you need to do is go straight to tab 6 and scroll down to the bottom to click on Integrate. No need to save the intermediate files.
Is there a reason why you made the changes you did in tabs 5 and 6?
Well, a couple of times, I did get satellite trails when "only" using automatic. Other than selecting Sigma rejection, anything else was just me trying to follow what the video guys did.
Is there any way to glean what this error msg. is saying? Other than, "this doesn't work"? I will eventually find out what dumb input I am making into something that works for most users.
I'm afraid it is not that simple unfortunately. Could you please upload the three frames to
https://upload.astropixelprocessor.com/
using user and password upload2 and create a directory paul_wilson_hoo so we can have a look at the them?
@wvreeven which 3 frames? Do you mean the 3 stack files?
I appreciate that you have tried to help. It's user error on my part.
I encountered with the same error message this morning when I tried with my old QHY10 lights. I do not understand why the extract Ha, OIII, and mono integrations were read as RGB again.The different version of acquisition software may be the cause. However; after clear and reboot APP, loading these integrated files as lights, register, normalize and save normalized files, they can be combine with HOO algorithm by select only one channel per each normalize integrated file.
@wvreeven which 3 frames? Do you mean the 3 stack files?
I appreciate that you have tried to help. It's user error on my part.
Well, as you want. You can upload the three stacks if you want so we can have a look at them and see why they cause the error that you’re getting.
