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 am testing out the software on a 9 panel mosaic but came across a problem I cannot resolve. I tried a number of methods which did not work finally I found a method which did work. Here is what I did:
1. Stack each mosaic panel in the normal way using APP.
2. Align the individual panels using settings like this: Analyze stars (kappa 3 auto min and auto max with slider at min 1000, max 2600), Register (pentagons, scale start 2, scale stop 15, use dynamic distortion correction, uncheck same camera, tried various distortion margin settings but none were much different then default), Integrate (4th degree LNC, 3 iterations, enabled MBB true at 5%, no rejection, the rest defaults).
3. Integration succeeds with above settings but there is a bad seam between panel 4 and panel 7. The stars do not line up properly though they do on every other seam. I can send some images of this and screen shots of whatever is needed but I do not see a way to do this under add topic.
Has anyone else seen this happen where registration succeeds but some of the panels have alignment issues? I have tried tweaking every setting I can think of and the result is always the same.
I must admit that I haven't come across this problem myself but I did have great success stitching a 20 panel Ha mosaic taken with a 135mm lens. This in spite that the fact that I had some severe distortion in one corner. You might want to try increasing the minimum amount of stars, the tool tip recommends 2,500 for Mosaics. The rest of your settings seem similar to mine but I found that I had better results with LNC set to the 2nd degree with 10 iterations. How much do your frames overlap? If it's more than 5% you should try increasing it to 10%. It that still doesn't work I remember Mabula recommending to use 1 degree LNC with 3 iterations and MBB set at 5% when you integrate the individual panels.
I hope this helps,
Marc
