haze on image

[alacant]

Hi Fighting against the lack of darkness, I came up against this light patch in between the galaxies. The best I can do is crank up the contrast to try and get rid of it but then I lose a lot of the galaxies. Does anyone have any idea what may have caused it? I looked at other examples of the same target and no light patch. This is around 12 exposures stacked in Deep Sky Stacker. If I know the problem, I'll have another go tonight... TIA.

[vlaiv]

You might be hitting the noise floor there. Take a look at your darks with appropriate stretch to see if you have any amp glow in that region (that is a bit odd as amp glow is usually somewhere around the edges).

Excessive dark current at a part of image will generate additional noise in that region after dark subtraction, so this could be region where noise levels are higher, and while you adjust curves for noise at other parts of image to fade into black this might still be visible.

It certainly looks grainy enough to be noise related.

[alacant]

Thanks. Yeah, no darks. Maybe that's what I need for this one. I also remember the camera sensor: 34º, the highest I'd seen. Related?

[vlaiv]

Could be, most sensors experience something that is called dark current doubling - meaning that for raise in temperature for certain number of degrees C - dark current doubles in intensity.

This value is order of 5C-10C - so for example if your sensor has dark current doubling of 6C this means that you will have x2 dark current at 34C as opposed to 28C and x4 more than at 22C

If you did not take darks and did not do dark frame calibration of subs, then the haze might be related to uneven dark signal itself. If this is the case, then you still have a bit headroom to get to noise floor, and this haze can be removed with dark frame calibration.

[Chris Murray]

i had similar problems, i think condensation just starting to form, basic but some times its the little things that make you scratch your head

 

[alacant]

Yeah. Annoying. Dark frames clear it but they make it noisy. The dither usually knocks out most of the noise. But this is me being minimalist (read lazy!). I'll read about darks frames.

[vlaiv]

Dark frames will make it noisier, but by how much, that depends on number of darks you take and stack. It also depends on ratio of noise levels between sources of noise (read, sky, dark, and shot noise) and also on properties of master dark itself.

There are two types of noise (unwanted signal) associated with dark current - random component and constant component (this would be amp glow, column/row artefacts, hot/cold pixels).

When you dither / sigma stack you remove hot/cold pixels, when you dark calibrate you remove amp glow and column/row artefacts but inject back some random component. Depending on your dark frame profile you might end up with more or less noise when dark calibrating. If your dark frame is uniform - then don't dark calibrate (you'll remove DC offset in postprocessing), but if there is difference across the master dark then you should.

I've attached three images to help you understand what might be going on and relations of noise to dark current.

First one is my master dark frame made of 16 subs and stretched to show uneven distribution of dark current. I have pretty big amp glow. This is uncooled ASI185 camera where sensor temp is around 38C.

Second one is simulation of what would image background look like if I were to process it without dark calibration. Yes I would be able to make background black but if I push curves to much I would end up having haze on the corners where the amp glow is the strongest.

Third one is calibrated/simulated background. It is stretched more then the second one just to show the impact of random part of dark signal to noise distribution. It clearly shows there is more fine grained noise in areas where dark current was stronger but the image itself has less noise (here it maybe looks like there is more due to different stretching) and noise is of different type.

[glowingturnip]

i think that's vignetting that's not been corrected by flat fields - the middle of your image is lighter than the edges, and when you've lowered the background (be careful not to black-clip by the way), it's removed most of the vignetting but just left the brightest bit in the middle.

Did you take flats ?  Gradient xterminator or Pixinsight DBE would get rid of it