Binning How Much

[OhNo]

Trying to get educaated on the finer points of AP I found myself looking for solutions to what was right, wrong and preventable. With the program I am using in post processing (PixInSight) there are tools for just about everything. One such tool is the Deconvolution and Convolution tools. To the best of my understanding you can manipulate your images Stars with these tools.

Searching for answers I found an article written by ZWO on Binning. As I read I couldn't help but think about past images that I looked at with the deconvolution tool. I then get to 3 images that defined what Under Sampled, Suitable Sampled, and Over Sampled imageslook like. From playing with the convolution tools in PI I am convinced that every image I have taken with my ZWO ASI533MC Pro at 1x1 binning are indeed Over Sampled.

By increasing the binning to 2x2 this may (other factors to be considered) help with this issue. If you change the capture settings rather than fix in post processing.

Opinions on this topic are welcome. Here is the link to one article I read: https://astronomy-imaging-camera.com/tu ... ntals.html

[Star Dad]

Well, I do not routinely use Binning (I have the ASI294MC-cooled). Particularly on Nebulae where I prefer as much detail as possible. On some open clusters I use binning mostly because I do not care if some of the resolution is lost and the increase in speed of post processing helps a lot. Though with my drought of clear skies for the past 6 weeks (I think) speed is not a good reason. It does help enhance the stars, I think, without diminishing the result. Take this in the "For what it's worth" department.

Edit: I just realized that my avatar is actually a binned image.

[OhNo]

I was surfing AstroBin, and noticed a lot of images taken with 0x0 binning. Which I didn't realize was a option. I am finding with me new camera (ZWO ASI533MS Pro) that I am getting a lot more data from faint Stars than I did with the Canon 1100D. The sheer number of photons from the extra photons from those tiny Stars id almost to the point of being distracting.

I see your point about gather nebula details though.

So here is a question, how do I go about reducing the over sampling? Lower exposure time, Lower Gain, change off-set, oh and so many more!!!

[STEVE333]

Hi guys - I read the article on binning in the original post for this thread. It mentioned that any binning for CMOS cameras should be done in software (processing) not during acquisition. Since we are all using ASI version cameras (which I'm pretty sure all use CMOS sensors) I would suggest that we do acquisition at 1x1 binning (i.e., no binning) and then do any binning when processing. That should produce the highest quality images.

Just my thoughts.

Cheers,

Steve

[OhNo]

So 0x0 is the same as 1x1? What is the cause of over or under sampling I guess is the real question.

[yobbo89]

ahh yeah, using bin mode on a cmos chip usualy drops it's bit depth crippling it's dynamic range . i would do any down sampling in processing .

oversampling realy is the limit on what you are imaging and your sky . how to overcome it, better tracking ,better sky , down sample it in processing ,buy an appropriate camera based on it's pixel size, use bin mode as a last resort . being really oversampled helps to image fine detail in planetary imaging especialy when you're at 10-12 meters fl with 2.4 um pixel size , "lucky imaging" can counteract that bluryness from the atmosphere which is a pro for a camera running very high frame rates from 30- to 200 + FPS a second .

all i can say is, why would one need resolution on a star if all the detail it is going to give you is that it is round .

[Star Dad]

I agree that binning should be done via software with CMOS cameras. And yes, the ASI294xxx is CMOS based.

Gain introduces noise, but prudent use also enhances the signal - I cannot recall what mine is set to, but I think it is around 10% or so.
Longer exposures generally give less noise and more signal, but if there is light pollution there is a break even point. In my Bortle 4.5 skies it is about 2 minutes. Any more time will result in LP noise increasing along with the signal - so it is pointless to go for more time. More exposures will help reduce noise.

I too found a HUGE difference when I switched from my Canon 70D to the 294. Even a crappy night with the 294 is as good as a great night with bortle 2 skies and the 70D. It is amazing how many stars there are!

Edit: Another reason I keep my exposure (even in low LP areas) at 3 minutes is the ever increasing satellite and aircraft trails. I would HATE to toss a 20 minute exposure. I'm ok with a 1-3 minute loss - but 20 minutes would have me screaming.

[OhNo]

As an intro to the world of new age imaging cameras a buddy lent me his QHY163C. He sent along instructions for Gain and Offset. Didn't take long to realize with that camera "Less-is-More" worked best.Shorter exposures more of 'em.

I have a great video I found of a gyuy imaging from a rooftop in a major City in Japan. He went through the "Unity" thing is a understandable presentation. I have to re-visit it to see what he had to say about binning!

Nothing worse than only getting a portion of the information. ZWO could do better! IMO QHY has a long way to catch up to ZWO.

I'm glad to I finally have my PHD2 guiding issue sorted out. The sequence of events that led to the failure of my guiding had as much to do with the problem as the solution did! Kind of like image processing with PixInSight!!!