New File Format: XISF - Extensible Image Serialization Format

[Buzz]

For info, I'm currently finishing off my third book, this one on Astrophotography, to be published by Focal Press in the new year. The initial reviewers of the book proposal all asked for PI content.

It includes over 100 pages on PI processing, including an introduction, alternative PI workflows and how to use it on a number of example images and comparing it with other applications' tools. To achieve this I have had to evaluate and make sense of dozens of resources on web and forum and then consolidate this into a structured approach. The research and writing has propelled my learning and I'm glad they demanded it.

100 pages is not enough though - it deserves many more to fully do PI justice.   

Maybe there is a collaborative project out there - between the developers and the users, to put something back to the community?

regards
Chris Woodhouse

[MortenBalling]

Since I've upgraded today, I get this strange error message:



I was trying to Blink my Lum and my Ha stacks, and I'm pretty sure that worked yesterday with fits?

So who am I to push, and where? 

Best regards

Morten

(And yeah, I know my paths are a mess)

[Juan Conejero]

Hi Morten,

This is a limitation of the Blink tool, but it has nothing to do with XISF. This happens because you are selecting a master image created by the BatchPreprocessing script with the "generate rejection maps" enabled (as it is by default). The file contains three images: the master frame and the high and low rejection map images. Note that this happens exactly the same with the FIST format.

This said, I've taken a look at Blink's source code and it would be very easy to change it to support multiple images, or at least to allow loading a multi-image file by simply ignoring the additional images. I'll try to make this change in the next version. You can also write a PM to Nikolay.

[MortenBalling]

Hi Juan

Thanks! I was thinking about something similar earlier today. I guess some of the things I do with PI isn't your normal out of the box routines. I use the software a lot, for combining a lot of different images, (typically nonlinear creative common licensed images), so I often run into problems, that might be relatively easy to fix, but I'm also aware, that it might not be the most urgent stuff for other users.

I actually have a few other suggestions (masks integrated as alpha etc.), but I'll try to sum that up, and make another thread.

BR

Morten

[Juan Conejero]

Just to keep the people informed, I have released a first draft XISF 1.0 specification document.