Show Posts

This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.


Messages - pfile

Pages: 1 [2] 3 4 ... 277
16
Bug Reports / Re: Registered Image Anomaly
« on: 2019 August 11 11:23:44 »
I know what dithering does. it also moves your bright star off of one set of pixels and onto another set of pixels, which might react differently to the charge overflow, thus changing the exact nature and width of the bloom.

just blink your registered frames and see if you see the blooms jumping around a tad. if not, my theory is wrong. if so, my theory may be right.

nothing has changed in ImageIntegration...

rob

17
General / Re: Issue saving in TIFF or JPEG
« on: 2019 August 11 02:27:07 »
PI doesn't consider saving an image in any other format than XISF as "really" saving the image. this is because no other file format can capture the entire state of the image - for instance no other format can remember the STF setting because that concept does not exist in any other file format. also some file formats are lossy, like JPEG, so they do not capture the true image data.

so this is normal behavior.

rob

18
Bug Reports / Re: Registered Image Anomaly
« on: 2019 August 10 14:05:53 »
I'm just thinking that the dither puts the bright star in-between different pixels, thus making them slightly different from one another in each subframe... and then when you register and integrate, the blooms grow a bit since you're seeing the union of all the slightly different blooms.

rob

19
you can just copy the license file over... this is in the FAQ section 2.7

http://pixinsight.com/faq/

rob

20
Bug Reports / Re: Registered Image Anomaly
« on: 2019 August 09 12:08:46 »
wouldn't this just be because of your dither?

rob

21
i found that using axial symmetry in the Ha image helped extract the gradient a little better.

i also did an HOO mapping without stars (using starnet++).

there's a zipfile attached with a dynamic crop icon and a dbe icon for the Ha. however, it is on a workspace named "tulip" and i'm not sure what PI will do when you load that since most likely you don't have a workspace with that name.

rob



22
well i was trying to say that (i assume) you have done DBE to the masters for those jpegs, but i can still see the gradient hanging around. so i think that limits what you can do with the data color-wise, since reds that aren't really there start becoming dominant when you push the saturation, etc.

maybe its worth a try with ABE to see how it solves the gradient.

rob

23
yeah bicolor is sometimes rough. it's true that SHO images are way more interesting color-wise and look much better.

i think in your processed images the gradient is still messing things up. i suppose you could use ColorMask to get a green mask and then tweak the green a little bit so it looks more blue. but processing this data is really difficult...

rob

24
Wish List / Re: High-Level Process Step History
« on: 2019 July 31 11:31:07 »
OK... yes you have to load the history explorer from whatever image you are interested in, but on the left are just the names of the processes that have been run on the image. i think that's what you want?

rob

25
yes, the idea is that where there is Ha at strength 1, there is Hb at strength 0.2-0.3; Hb is deep blue. however of course atmospheric extinction means that you wouldn't necessarily see that exact proportion if you had an Hb filter... so it's kind of fudged. the 0.8*OIII is because @ 496um, the OIII signal is 100% green and about 75% blue. you might experiment with that proportion as well.

you will get different results depending on whether or not you have linear fit the OIII to the Ha. while there is no physical basis for linear fitting them, the combination may look better as usually the Ha signal totally dominates the OIII and in the combination you don't really see the OIII at all if they are not linear fit. since it's a narrowband image, you might as well do what you think looks best...

rob






26
ok, well even with DBE it's hard since there is so much signal where you're trying to extract the background.

have you tried straight HOO or maybe HO (0.2*H+0.8*O) ?

rob

27
i think the Ha image suffers from bad flattening; there is a strong bright circular large-scale artifact there. you should probably resolve that before trying to process the image.

rob

28
Wish List / Re: High-Level Process Step History
« on: 2019 July 31 10:23:59 »
what constitutes high level processes?

rob

29
Wish List / Re: High-Level Process Step History
« on: 2019 July 31 09:18:34 »
meaning a summary of what is in the history explorer? cause... all of that stuff is in the history explorer.

rob

30
frames from SGP? it's a bug in how SGP writes the sitelat/sitelong fits keywords.

this never happened in PI before because it never payed attention to those keywords, however, with all the astrometry built in to the latest version of PI, the site lat/long becomes pretty important.

rob

Pages: 1 [2] 3 4 ... 277