Flat grey backgrouds on newbie's image (aka please help!!!)

[e.ventura]

Hi there.

First of all let me salute everyone of you guys. I'm new to theese forums as well as astrophotography (at least deep sky stuff). My hobbies for the past few years have been astronomy and photography so I thought of combining those two. I've had the chance to say to Ivo that he is quite a miracle worker because through StarTools I've been able to see interesting objects in the night sky through astrophotography that I just couldn't in the past (I don't have a telescope). But enough of that.

The first picture that I processed was this one http://www.astrobin.com/147818/... Now don't get me wrong I'm OVER THE MOON with this one, even though I've spotted some big mistakes like making a dodgy mask when using Life/Isolate rendering the nebulae's imediate surrondings rather different from the rest of the background (unfortunatelly I was only able to spot that on another screen witch had different calibration settings). But still I could get a hint of The Running Man (very excited) with only 63 x 1.3s exposures. I'm still a bit restricted as far as exposure times go because I don't have any tracking equipment... Am going to build a barndoor tracker soon (hopefully...)!

But that is THAT image (beginners luck I guess). I'm struggling with this Lovejoy image I took:

https://5y9w6p.s.cld.pt

I don't get it... I've stacked it with DSS (I know that there is not much data (I just had a bit of time so I rushed to my balcony and shot this)). There are 45 exposures 4 seconds long, with 30 darks. I didn't do flats because of time restraints and I guess I didn't do Bias because of the excited state I was on just wanting to see the resulting image. Now, the problem is, after croping the image (to remove the balcony (top left) and home in on the comet) I AutoDev'd it and has expected a yucky orange background appears, I ran Wipe (on Color and Brightness) and the image then presents a flat grey background ALL around. When I go for the AutoDev again (redoing global stretch) I get a whole array of extremely saturated, noise like pixels that I can't do anything with...

If someone could point me out some kind of direction I'd be much appreciated... Some people tell me I'm a fast learner... Well I just want to learn...

[admin]

Hi Eduardo,

The first picture that I processed was this one http://www.astrobin.com/147818/... Now don't get me wrong I'm OVER THE MOON with this one, even though I've spotted some big mistakes like making a dodgy mask when using Life/Isolate rendering the nebulae's imediate surrondings rather different from the rest of the background (unfortunatelly I was only able to spot that on another screen witch had different calibration settings). But still I could get a hint of The Running Man (very excited) with only 63 x 1.3s exposures. I'm still a bit restricted as far as exposure times go because I don't have any tracking equipment... Am going to build a barndoor tracker soon (hopefully...)!

Wow, that's pretty impressive with 1.3s exposures!

But that is THAT image (beginners luck I guess). I'm struggling with this Lovejoy image I took:

https://5y9w6p.s.cld.pt

I don't get it... I've stacked it with DSS (I know that there is not much data (I just had a bit of time so I rushed to my balcony and shot this)). There are 45 exposures 4 seconds long, with 30 darks. I didn't do flats because of time restraints and I guess I didn't do Bias because of the excited state I was on just wanting to see the resulting image. Now, the problem is, after croping the image (to remove the balcony (top left) and home in on the comet) I AutoDev'd it and has expected a yucky orange background appears, I ran Wipe (on Color and Brightness) and the image then presents a flat grey background ALL around. When I go for the AutoDev again (redoing global stretch) I get a whole array of extremely saturated, noise like pixels that I can't do anything with...

You're doing everything right!. When this happens, AutoDev is having a hard time detecting any 'real' signal in the region that you give it (by default this is the whole image!), so it starts to try digging out detail from the noise. The simple solution is to give it a Region Of Interest by clicking & dragging an area over the image that you feel is a good representation of the signal that needs to be stretched. If you have an object on an otherwise noisy background such as a galaxy or (in this case) a comet, the ROI is the galaxy or comet. Do play around with the ROI. Sometimes just a slice of the object is more appropriate.
The other thing that you can/should tweak if a lot of noise is visible, is the 'Ignore detail <'. This tells AutoDev not to optimize for small dark detail so much (which typically describes fine noise), instead concentrating on bigger structures and objects.

I'm currently downloading your data set. Will report back when I've had a look.

EDIT: Had a look and it's pretty much what I said above; define an ROI! Also be sure to bump up the Dark Anomaly Filter in Wipe as the background is quite noisy which means that some background pixels are 'abnormally dark'.

20150111_Lovejoy_ISO1600_f3.5_50mm_3min_Autosave.jpg (28.9 KiB) Viewed 3839 times

Workflow;

--- Auto Develop
To see what we got.
--- Crop
Framing comet.
Parameter [X1] set to [1600 pixels]
Parameter [Y1] set to [1646 pixels]
Parameter [X2] set to [2404 pixels (-2217)]
Parameter [Y2] set to [2129 pixels (-935)]
--- Wipe
Parameter [Dark Anomaly Filter] set to [6 pixels]
--- Auto Develop
Final stretch. ROI (click & drag) over comet.
Parameter [Ignore Fine Detail <] set to [4.2 pixels]
--- Color
Color module got colors wrong quite badly (can't blame it as there is not much signal to go by). Looking for reasonable distribution of star color temperatures instead (red->orange->yellow->white->blue).
Parameter [Dark Saturation] set to [4.20]
Parameter [Blue Bias Reduce] set to [1.00]
Parameter [Green Bias Reduce] set to [1.00]
Parameter [Red Bias Reduce] set to [1.00]
--- Wavelet De-Noise
Parameter [Scale 1] set to [61 %]
Parameter [Scale Correlation] set to [1]
Parameter [Color Detail Loss] set to [28 %]
Parameter [Grain Size] set to [9.8 pixels]

[e.ventura]

Hi Ivo, thanks for your input.

I had actually fiddled a bit with a ROI, but got that horrible redish noise on pixels the size of bloody alpacas, and thought that StarTools couldn't cope with it... What a fool I was... It actually turned out better than I was hoping... That just goes to prove that I should rely more on the de-noise abilities of ST...

This was what I came up with, just messed a bit with Life/Less=More...

http://www.astrobin.com/148811/

It still has some blurry alpacas roaming free on the background, but I guess I must compromise with it... After all it was only 3 min integration time with 4 seconds exposure time (and no flats)...

Cheers, thanks for the help. I just needed to know I was on the right path...