Canon 6D Super Bias Question

[Andrew J]

I created a Master Bias from a combination of 250 images from my modified Canon 6D at ISO 1600. When I converted this Master Bias into a Super Bias I am getting some strange results. I am expecting a nice smooth Super Bias, but instead getting something with a bunch of streaks (see attached). It looks even worse if I apply the STF Boost stretch. Any ideas why I am getting this result? Is there something wrong with my camera’s sensor? Should I even use the Super Bias or just stick with the Master Bias? Is it fine and I am just over reacting?

My Bias Image Integration Settings: Pixel Combination = Average, Output normalization = none, Weight mode = don’t care, Scale estimator = iterative k-sigma / BWMV, Pixel rejections: Winsorized Sigma Clipping, Rejection normalization = none, Rejection clippings low/high = yes, rejection parameters = 6 sigma.

Super Bias Settings: Orientation = Columns, Multiscale layers = 7, Median Transform = checked, Exclude large-scale structures = checked.

I appreciate the feedback.

Andrew J

[msmythers]

The vertical streaks are correct for super bias.

Here is a link to the posting from Juan about the SuperBias tool. http://pixinsight.com/forum/index.php?topic=7266.0

Here is another link to a similar discussion last month. http://pixinsight.com/forum/index.php?topic=9889


Mike

[Andrew J]

Thanks Mike for the links. I had read Juan's posting but had not seen the other. I may have come in with a false expectation. I saw a presentation David Ault did for the Astro Imaging Channel where he showed a nice smooth Super Bias with no lines. I guess that maybe just the difference between a DSLR and a CCD.

So I take it that the vertical banding is normal. What about the dark horizontal band I see across the bottom? Is that normal as well? 

Also, does it matter for a DSLR if I choose 6 or 7 Multiscale layers? I have seen recommendations both ways. I have run it both ways and there seems to be more banding with 7 layers vs 6 layers. Not sure if one is better than the other. Is there a way to tell?

Thanks again.

Andrew J

[msmythers]

Andrew

The only thing I can tell you is for my images I see an improvement in very faint data when using super bias. It's simple to use which is great. As far as 6 or 7 layers I can't tell any difference for my images so I leave it at default.

The horizontal bands are present in your master bias so I'm not surprised they are there in your super bias. They are not random in their nature.


Mike

[Alejandro Tombolini]

Hi Andrew, this post could also be useful.

Saludos, Alejandro.