processing help

[Anthonyexmouth]

can i do that in DSS ?

[Marci]

Possibly, but I don’t let DSS do anything other than stack. Once you’ve got the stacked TIFF you’d usually open it in Photoshop (or your image editor of choice) and convert it to 16bit to do your processing, downsample it to original resolution, then resave it out as a 16bit TIFF for astrobin etc, and export out as JPEG for sharing on Social Media etc

If not using convolution / excessive unsharp-mask filters etc, can leave drizzle off in DSS. Can’t speak for all, but I tend to rarely use it unless I’m shooting with my low-res planetary camera (ASI034MC - shoots at 728x512).

[Marci]

(NB: Lots of edits to last two posts... sorry!)

[Anthonyexmouth]

just noticed i can save my tiff from DSS as 16bit, should i do that or keep 32bit. 

[Marci]

You can, but memory says there’s a reason why most don’t (I can’t recall what it is off hand)... as I say, I only let DSS stack and then usually take the 32bit autostack file it produces rather than saving it out. _Everything_ else I do in Photoshop. Either way, if you do, ensure you choose "embedded adjustments saved but not applied", then take it into your image editing software to process further.

EDIT:  See https://expertphotography.com/post-processing-astrophotography-all-you-need-to-know/ - and scroll down to "Case Study IV: Deep Sky Objects" as a for-instance of a typical editing workflow.

[Marci]

(More edits, again... soz - I’m a nightmare for it)

[Anthonyexmouth]

thanks, ill have a read of that. 

[MarsG76]

9 hours ago, AKB said:

I may be wrong, but I think that this is M27?

You're right, well spotted.... definitely the Dumbell nebula..M27.

[wimvb]

There's really no need to drizzle images taken with dslr cameras. If you drizzle x 2, the image will become 4 times as large (2 x horizontally, and 2 x vertically)

A 14 bit raw file that is 6000 x 4000 pixels has an uncompressed size of 2 x 6000 x 4000 x 8 bit or 48 Mbytes

At the raw stage, each pixel is only gray scale, since the colour information is in the x, y position of the pixel together with the Bayer matrix (RGGB or similar).

After deBayering, the same uncompressed image will be 3 times as large, because now, each pixel contains red, green and blue data: 144 MBytes. If you then drizzle x 2, the image will become 4 x 144 =576 MBytes. Conversion to 32 bit will increase the image to 1152 MBytes or 1.15 GB. Stacking doesn't alter this.

In DSS I would keep the image 32 bit, but remove the drizzle settings. The Autosave image should then be only 288 MB.

Btw, your image contains strong vignetting, which isn't a surprise, as you use a large sensor with a scope that can't quite illuminate that sensor. Flats will remove (most of) the vignetting.

There is also a stacking artefact, and I suspect that a number of frames were offset and rotated. If there aren't too many of those, it's easier to remove them from the stack. Trying to get rid of these artefacts usually means cropping anyway.

If you image over several nights and have to remove your camera between sessions, it's better to align the sensor with RA or DEC. Take a 10 - 15 second exposure while slewing your mount at 1 x sidereal rate in RA. This will give star trails in RA. Rotate the camera and do a new exposure, until the trail lines up with either the long or the short edge of the image. This ensures that the only offset you have between sessions is in RA and DEC, without any frame rotation. You will be able to use a larger portion of the sensor this way.

[Markkus]

I had to use RawTherapee to open the 32 bit tif.  Did a little black levels, lightness and contrast processing in there, then exported to Paintshop Pro for some further levels processing to darken the background, then exported to jpeg.  My very limited experience of star field processing suggests that it's a judgement call between achieving a nice dark background and losing the dimmer stars.  Anyway, FWIW...