New to PI - Basic Questions

[shammo]

Hello,
I have just recently began to process some images in PI. To this point I have not tried calibrating and combining any subs, but have been doing this process in CCDStack and then opening the individual LRGB combined fits in PI. When I open unstretched combines, I get a range of brightness on the image, from black to light gray and anything in between. Can anyone explain why some images are black and some are gray? There must be some problem between CCDStack and PI and it clearly affects my end result. I've tried cropping away any excess after the combine, but this doesn't seem to yield consistent results.

Any help would be appreciated.

Thanks,
Scott Hammonds

[Carlos Milovic]

Hi Scott

From what you said, it is possible that CCDStack is writing using a signed format. This means, instead of mapping the pixel values in the 0-65535 range, they are using -32768-+32767. Also, they may not be using the whole range, but only the positive values. That's the reason that the RAW image is displayed as grey.
PixInsight uses the normalized range as convention for every image, no matter the bitdeph. This means, internally everything is rescaled so 0 is black and 1 is white. Greys are fractional values in between.

One way to fix that would be to use PixelMath. Just type the following expression: "$T-0.5". This should also work, and be displayed a bit better: "($T-0.5)*2". Do not enable the rescale option.
Also, try searching in the forums. Maybe there is a little script somewhere to fix several files (a batch script). If not, it is not hard to do that with the PixelMath instance described above, and an ImageContainer.

Oh, also try searching for the "format hints". They are instructions that may be used to read the raw files by the calibration process, and a few others. It may also solve your problem.

[shammo]

Good morning Carlos or perhaps good afternoon,
Thank you for the reply. I will also determine if it is possible to save the file in a different format. Although it is supposed to be saving it as a 32-bit floating point file. I have an option to save as 16-bit unsigned integer, but that seems counter to the idea of processing in the full 32-bit mode.

The PixelMath option worked perfectly. One additional question, should I run the PixelMath option on all of the files I am processing, even though they appear black? For example, I am trying to process an RGB image of M92 and the blue file was the only one with a problem, the other two appeared black.

Thanks,
Scott

[Carlos Milovic]

Morning I'm in Chile, not Spain

Just fix the files that are shown middle grey when loaded. The others should be fine.