Hyades and Pleiades

[ralph1]

Thought I'd have a go at a Hyades and Pleiades wide field. I'm somewhat happy with the result but there is a vast amount of noise and an annoying glow that I couldn't get rid of down the bottom. I suspect it was because it was a warm night and the camera was hot but other opinions are welcomed as this plagues all of my wide field images.
Olympus PEN E-P3, 14-42 mm lense(I think it was set to 32mm, but not sure.)
ISO 12800, 60X10 second subs.

[LightningNZ (Cam)]

I'm not really familiar with the camera, but could it be light pollution instead?

One thing, you might want to knock back the ISO on such bright targets. It's reducing your total dynamic range and burning out the cores of the stars while not gaining you any extra faint detail. That's not what's causing the brightening in your image but it's not helping you either. Probably around ISO 400 (check out DXOmark.com for their sensor ratings) would be good for you. You can always make things brighter by stretching the image in GIMP or Photoshop.

[ralph1]

Thanks for the advice Cam. I don't think it's light pollution as there is only one light dome and it is to the southeast while I was looking north. It tends to appear on the side of the sky closest to the horizon but that occurs all the way around the sky. Somehow I always image things low in the sky so that probably doesn't help.
I like to play around with settings a bit - I use ISO 200, the cameras minimum, for the moon and planets and usually push it up to between 800 and 3200 for DSO's. One of the reasons I was using it was the camera is on a simple non-tracking tripod and at 15 seconds the stars started to trail. For the next target that night, Orion, I did one at 400, one at 800 and a few at 6400 just to see which worked best. The 400 and 800 produced much darker backgrounds but were about 2 magnitudes short of 6400. The 800 looked best overall but silly me used 6400 anyway

[raymo]

Why on earth are you using ISO 12800? Stick with the 10sec subs, but at
ISO 400 as Cam said, or maybe 800, trial and error.
raymo

[ralph1]

Cam
Wouldn't stretching require me to take more subs to fight the noise? I have no shutter cables or computer control so to take 60 exposures 10 seconds each I stand around pressing the shutter button every 10 seconds, 60 times. 60 frames is a major improvement over 19 for orion in terms of noise, even with orion mostly taken at half the ISO(does ISO affect noise?). If anyone else has the camera it would be great if they knew a simple way to take multiple exposures automatically. I think the high ISO did burn out the bright stars, most problematically aldebaran, it made it difficult to keep some remnant of colour. I'm not too worried about the faint detail as I see no nebulas there and stars aren't the most detailed of subjects. DXOmark.com seemed to say ISO 536 but I was confused by all the numbers and I may not have found the right one Is it possible to take subs of many different exposures and ISO's to get both faint detail and bright star colours?

Raymo
Like you said, trial and error. I was playing around with a few different settings on the camera to see if they made a difference in image quality. From memory high ISO degraded faint extended objects like the milky way but I hadn't tried its effect on stars. It did let me see fainter stars but it made the background too light so I probably won't use a setting quite so high for the next wide field. What areas of the sky are interesting but tend to not get photographed much?

Could it help to to take darks, flats etc to get a smoother image?

[LightningNZ (Cam)]

Originally Posted by ralph1

Thanks for the advice Cam. I don't think it's light pollution as there is only one light dome and it is to the southeast while I was looking north. It tends to appear on the side of the sky closest to the horizon but that occurs all the way around the sky. Somehow I always image things low in the sky so that probably doesn't help.

It probably goes without saying, but try to wait for these objects to get as high as possible before imaging them. When you are stacking/stretching faint images it doesn't take much haze or fog to get lit up by any terrestrial lights that are around and ruin an otherwise good image.

It might be worth stretching a few subs and seeing if there is some cloud that only appears in a few of them. You can then biff those out and reprocess the stack.

[ralph1]

Sadly the pleiades never get very high from Australia but that excuse does't work on orion. There aren't any lights around me within 500metres that aren't blocked by trees but there is often both haze and fog in the early morning mainly. This was captured on the one clear night in a 2 week run of cloudy weather so there could have been residual cloud around that I didn't see with the naked eye. I used kappa-sigma clipping on DSS to remove satellite trails - would that remove stray clouds as well?
Should have said in the original post: processed in PIPP(the wrong use for this program but it's just so helpful), Registax 6, DeepSkyStacker and apple iPhoto.

[raymo]

Ralph, about $15 will get you a cheap intervalometer on ebay, which will
let you plan and execute as many subs of any length as you like. It plugs
into the same socket as a conventional shutter release cable.
Higher ISO equates to higher noise, so the lowest possible ISO in any given situation is always best. This is even more important in the warmer months, as noise increases dramatically with higher ambient temps.
Many serious imagers have cooled cameras to tackle the problem.
Darks are essential for any night sky imaging, but flats can come a little further down the track. If you are taking short subs [up to 30 secs or so],
enable the camera's high ISO and long exposure features, and the camera will automatically take a dark after each sub. A dark has to be of the same duration as the normal sub, so it does mean that the number of subs that you can get in a given time will be halved. If taking longer subs
too much imaging time would be wasted taking darks; better to take some darks separately yourself, so you would disable the noise reduction features in this case.
All the interesting things are frequently imaged, so initially I suggest
concentrating on getting your imaging technique up to the standard you require before worrying too much about what targets to select.
raymo

[cometcatcher (Kevin)]

That orangy stuff down the bottom might be cloud. On the other hand that camera has a very low ISO mark so it could be amp glow.

[ralph1]

If it helps here's some of the subs. After stacking in DSS, the image was more of a washed-out orange. I examined the subs, stretching a few, but I couldn't see anything looking like cloud. I noticed the first subs had a glow at the bottom which transitioned to fairly even about 4 subs in and changed to a glow on the right side about 10 subs in. In general they got darker as it got later in the night. I looked at the subs from orion where the camera was turned on its side and the effect moved with the camera.

[LightningNZ (Cam)]

Well it moves around in the first three images (which all have the same orientation), so I'm going with light cloud.

[Tony_ (Tony)]

Hello Ralph,

Your image can be improved with further processing.
Here's a quick go I had using photoshop, not perfect but it gives an example of what you can do. With more time it could still be better.

Regards,
Tony.

[ralph1]

Cam
Remember those are 3 of 60 subs at different times in the sequence and the 4th image is a different target. Next clear night I'll go out and try again and see if it happens in the same way. If it doesn't then it must be cloud. If it does it could be glow from the sensor as Kevin suggested.

Tony
What adjustments did you do in photoshop? I don't have it but if it can improve an image that much it could be worth getting.

[cometcatcher (Kevin)]

You should see any amp glow present on dark frames. You can take these with the lens cap on.

[Tony_ (Tony)]

Hello Ralph,

The overall process I used is fairly detailed and unless you have photoshop it won't mean much. I have forgotten exactly what I did.
Basically I first used an add on called gradient exterminator.
There was a fair bit of mottling/uneven background due to excessive noise. I used a background copy with the brighter stars erased, then median noise added. This makes a blank gradient which can them be subtracted from the main image (an offset is needed to prevent the background becoming too black).
Then some noise reduction using masks (https://www.youtube.com/watch?v=32Mj2Ia-tC0). This is a good technique to remove noise from the background. The guy who did this one has some other good tutorials too.
Then some offset/gamma adjustment (not too much or you lose data and "clip" the black).
Then star size reduction using add on "starshrink".
Also levels/curves adjustments and some colour saturation of the stars using masks. If you aren't happy with the final result you can start again.

It takes a while to get used to this. You will learn more with practice and online tutorials + trial and error. Getting good data is the first step - but good processing is also important. You want your final image to look like stars not dots on a black background. I don't take many images of open clusters because of that.

I think the gradient in your image is from stray light/light pollution. Wide angle shots are more prone to light interference especially low in the sky where moisture or low level light clouds have more effect.

To improve your images stick with 10 sec subs (if you don't have a drive mount). Reduce the ISO. You will need some sort of processing program. I like photoshop but there is a program called GIMP which you can download for free (legally). It has similar features to photoshop but not as good. It is okay to start with though. Another popular program is pixinsight. I don't have it and I suspect it takes a while to learn it.

Whatever you use you will get better with experience.

Regards,
Tony.

[ralph1]

Tony
Thanks, I downloaded GIMP and it's much more flexible than registax for post-processing.
Raymo
I took darks after imaging the Jupiter-moon-regulus conjunction but I have yet to process it and see if it made a big difference.