Advice on different subs, HDRComposition vs ImageIntegration

Jon Pither

Hi,

I made a mistake / clanger with the rotation of the image train between consecutive nights, by a whopping 90 degrees difference!

You can see in the attached image that I have 5 different exposures here for L: 60, 30, 15, 8, 4 seconds.

On the first night I took the 15 seconds - the top right image - and you can see how registration has rotated it, but I've lost a substantial chunk in the process (the black areas).

The 30 exposure was completed over both nights, and you can see an imperfect blending of the two framings, in the middle left image, happening during registration.

This is my first attempt at HDRComposition. If I run HDRComposition against the five images, PI gives an error, which is fair enough: "Error: Inconsistent HDR composition detected (bad linear fit, channel 0)".

I'd appreciate any advice on a route to take, to go from here. I took different exposure lengths, to take advantage of HDRComposition, such that I can avoid blowing out the core, but now I'm a little stuck.

My data is here, just in case: https://drive.google.com/drive/folders/1M-DOO2ma1fYrxtY1KHRnaB80V0O8YFHp

Thanks!

BlockHead

Hi Jon,

It would seem to me that all of the problems would be solved if you choose to integrate everything (and let local normalization take care of the overlaps). Are you 100% certain the data in the center of the nebula is useless for the 60 second exposures?

-the Blockhead

Jon Pither

Hi Adam,

Thank you. I HDR'd the RGBs, as my problem is only with L. I rejected the mal-rotated images from the first night. I'm happy enough with the attached image! (though would love any feedback!),

Jon.