PixInsight: Batch Fits Header Editor (Script)

[@marshawilcox]

Thank you, Adam. This will be wonderful for adding Filter information to files in which it is not written. The ASIAir only writes filter information when the filter wheel is used; there is no easy way to indicate filters used with a filter drawer. I’ve been using the folder name (… Filter_X) and am grateful to have a way to imbed this information in the files themselves.

[@AdamBlock]

A perfect usage!

[@JeffFishman]

The other way to fix my NB filter issue. Thank You 🙏

[@jims9982]

Thank you sir! Astrobin would not plate solve my final image. I looked back at it in Pixinsight with Fitsheader and the Fitsheader data was empty. Thank you for the great video.

[@davewilton6021]

Thanks! I run into bad FITS metadata from time to time, and this is great solution.

[@carolsantulis2874]

Thanks, I really needed this info.

[@eugennaiman1195]

Great tip Adam - I had no idea this script existed and I was wondering myself about how to do bulk edits. Problem solved now :)

[@Mr77pro]

A great script indeed! I had an issue of files that should have been Bin2 but were Bin1. I re-sampled them to Bin2 in Pixinsight but WBPP still saw them as Bin1.

[@AdamBlock]

Yes! This is another very good example. The resample process does not update the header. (and at least with integer resample... perhaps it should...)

[@jeffdutton1910]

I had this exact problem, though only with one or two images in my dataset. I was able to do the required edit with the fits header process, then manually save the file back to disk. This script is a slick way to capture it all into one smooth operation! Thanks Adam, and thanks Mike Cranfield

[@andreya108]

Thank you, Adam. It's a miracle, your video was published the same day I recognized the fact that I've incorrectly set filter during recent session! 😀

[@AdamBlock]

Serendipity!!

[@AnomalousAstro]

Thanks Adam. I've edited the filter names one at a time in PI, but +-200 images is a non-trivial exercise. Huge thanks to Mike C for creating the script (and all the others) and you for bringing it to our attention!

[@Orionshield1948]

It's interesting that Pixinsight can extract the Fits Header info but the script cannot do likewise with the same .xisf file. Which will mean that If the occasion arrives I will still have to edit such manually.

[@guyonamotorcycle1]

Great stuff, this will help lots of people, including me. Thanks Adam!

[@AdamBlock]

Thank you for watching!

[@gregerianne3880]

Thanks so much, Adam. Not sure if you’d really want to do it this way, but this would allow processing in PI of files with different filters that WBPP might not be able to combine more easily, correct? Something similar to what APP can do with two different dual narrowband filters, for example.

[@AdamBlock]

Sure... you could force a filter name- renaming one filter to the same as another. In this way when using WBPP the data will be integrated together.

[@gregerianne3880]

@@AdamBlock Tried this and it works perfectly, Adam! WBPP processes the data from both filters and produces a single image. I did separate the data from the two different filters into sub folders (Filter_1, Filter_2) so the flats would be processed with the appropriate data. I haven't actually processed the image yet, so I'll have to see if there's a difference between combining the dual narrowband data 'after the fact' vs. having WBPP generate a single image using both filters. What a super utility. Many thanks to Mike Cranfield!

[@MichaelLeslie-rs2qp]

ls there a app or script that you know of that will rename files using metadata from the header

[@AdamBlock]

I do not know one...I know how to do it in Python with AstroPy methods. It is pretty trivial. Though since you are opening the file to read it...it would be tricky to rename the same file. Making a new copy would be easy. Anyway, it is all possible.

[@normanhey8555]

Looks great for solving a problem I have in filter names--except the script just isn't working as advertised. I loaded files--that took about 8 seconds to load 10 files. Selecting Edit from the Action to Perform panel took another 8 seconds, and the Target Keyword field remains blank. I deleted the script from PI's src folder, deleted the repository and restarted. Reloaded the repository and script--same result. Reset all repositories, reloaded all scripts--same problem with BatchFITSKeywordEdit. All the other scripts I have all work fine. The files I would like to edit all open fine and show the appropriate FITS headers when checked with the FITS Viewer in the File menu. Any help would be appreciated. My next step will be to do a re-install of PI and all of the repository information to see if that solves the problem. I realize that this isn't a problem in your wheelhouse, but since there is no way I can see to contact Mike directly, I thought I would see if you had any ideas of how to repair this. No help from the PI Forum so far...

[@brinkoo7]

nice! i have been using a nasty little python script to bulk modify haha

[@AdamBlock]

Well with astropy it is a 10 line script or so. I actually like the python method... entirely flexible... but for convenience in PI... this script is perfect.

[@malcolqwe2]

as per usual in my world, this doesnt work for me. After selecting edit, the target keyword list is empty. and I know there are values to be listed. In my situation, I am trying to plate solve a bunch of files in a mosaic. But the focal length is wrong in the fits header. I thought if I could edit the header to the right values solving would work. Oh well. maybe theres ANOTHER way

[@AdamBlock]

In your world... is it s FITS or an XISF file? Only works on FITS. Check the header . You know where to find me so you can post screenshots.

[@malcolqwe2]

@@AdamBlock of course - its been processed six ways to sunday - its an xisf :( thanks!

[@donaldwaddel4600]

Ok, so my acquisition software is screwed up and put a 60 in the " header. Now I have to spend 20 minutes, assuming I actually notice the one file that breaks WPP, and change the 60 to 00 and carry the one. The programmer/script writer could have simply done the following: If " = 60, " = 00, minutes = minutes +1. duh????? lol

[@AdamBlock]

You don't blame the acquisition software? WHY? Why are you not calling them out? Regarding WBPP, you will know instantly. It will not load. No time is wasted.

[@petekrz]

@@AdamBlock Postel's Law: "Be liberal in what you accept, and conservative in what you send"

[@AdamBlock]

@@petekrz THis is interesting..I did not know this law and learned something. Thanks.However.. in this case I cannot determine which side the law critiques (maybe it is both). In my mind "60" is literally an invalid number being produced by another software. It is like writing code that recognizes the digits 0-9 .. .but then requiring the code to "accept" something not from the set. "60" is simply not in the set of possible values for a sexagesimal representation. That position of the representation should never be a "6". Should we also accept "7" or "8"? And then increment ?? I think you can see which side I fall in for this case.