New User DBE Question

[DaveLeeNC]

Very new user here (new to AP and PI). I am processing a simple star field image that came from a dozen shortish subs from a DSLR/50mm lens. This is a learning exercise primarily.

I have a HUGE gradient from that starts in the lower right (very bright there) and fades out as it spreads across about 3/4's of the image (if I am going to image from my back yard I will be needing a lens hood here). I just cannot (either automatically or by manual placement) put sample points where the gradient is the brightest. The auto place just doesn't put anything there. Manually they all become bad samples.

And as you might guess DBE does a decent job of getting rid of this gradient except where it is the worst in the lower RH corner.

What are my options here?

Thanks.

dave

[Nocturnal]

Hi,

why are all your samples bad? Questions like this benefit tremendously from at least a screenshot that shows the image you're working on. Is your image flattened?

[pfile]

you need to increase the tolerance factor. if the gradient is really bright DBE thinks that the sample is placed on the foreground and so it rejects the sample.

you can also mess with the shadows relaxation value, but you should be able to get it with just the tolerance value.

finally, sometimes DBE can get confused if the image has some kind of a border with not-quite-black values. this is a common occurence after integrating several registered images. in this case you should crop the image first before applying DBE.

rob

[DaveLeeNC]

Sander, here is a link

https://www.dropbox.com/s/19qf74kldkezat3/f3.2LinearReject.fit

to the image processed with 25 darks/bias/flats applied. 11 30 second subs with a sensor temp of 30*C (learned the hard way about using LiveView). f3.2 and I used Linear Rejection in this run rather than Windsorized Sigma (mostly because I was experimenting). I believe (but can't yet show) that the gradient is from a window in my home that was low/right relative to the camera and there was no hood on the lens. FWIW, this was my first astro image. There was no guiding and the balance on my G11 might well have been funky (only load was a camera and dovetail). More experiments to be done (lots more and I think that I know what they are).

I don't really have a strong interest in the image, but do have an interest in how to use DBE and this image seemed a good place to start.

Thanks.

dave

[DaveLeeNC]

pfile - thanks!

dave

[Nocturnal]

Crikey that's a gradient! Rob is right about the tolerance, try increasing that until the samples no longer show as red.