DSS processing - vertical lines?

[GarethA]

I would also think 5min subs with that scope might be on the low side.

I once had this on M101 even though I had 4hours of data, the crud appeared on hard stretching.

I had matching darks but they did'nt do any good in DSS or Imagesplus.

In the end I accepted the subs were not long enough and only stretched so it did'nt appear.

Maybe 5mins was a it short... though m51 isnt exactly dim, certainly brighter than m101.

I can see the lines on the stacked image even before I stretch it...

[GarethA]

and I had the focal reducer in so was running at F7.2 which isnt the slowest ever! was hoping 5 min subs would be a good length...

[Thalestris24]

Hiya

Looking at your image a bit closer it looks like streaks - like the sensor has been left after a rinse or something! Um, you haven't recently cleaned it, have you? What does a single sub look like?

Louise

[wxsatuser]

and I had the focal reducer in so was running at F7.2 which isnt the slowest ever! was hoping 5 min subs would be a good length...

The only way to find out if they were long enough for your setup and sky conditions, is to

check the camera histogram. The peak wants to be at least 20% as this will get you out of the noise floor.

If the peak is between 20>40% then you should be ok with your lights.

[GarethA]

Hiya

Looking at your image a bit closer it looks like streaks - like the sensor has been left after a rinse or something! Um, you haven't recently cleaned it, have you? What does a single sub look like?

Louise

i cant see the issue on single subs - looks like dss is introducing it.

The camera was modded professionally and I haven't cleaned it so I dont think it's a camera issue...

[Thalestris24]

i cant see the issue on single subs - looks like dss is introducing it.

The camera was modded professionally and I haven't cleaned it so I dont think it's a camera issue...

How many lights do you have? Is the problem apparent on the stacked image prior to post processing?

What is the outcome if you just stack lights (no calibration frames)?

[GarethA]

I have 25 decent lights. I can't see the issue on individual subs (even super-stretched). I'll post one of the subs when i get home but i cant detect the issue.

When i stack the lights (with no other calib frames) then the problem is visible on the stacked image before any stretching or post processing.

[Psychobilly]

You can extract the sensor temp from the raw files using an app like exiftool

http://www.sno.phy.queensu.ca/~phil/exiftool/

Peter...

[glowingturnip]

think I've seen those streaks from other Canon cameras before too, though mercifully not my own, maybe a feature of some of the older sensors ?  There's even a script in Pixinsight to remove 'Canon banding' though I've not used it myself.

Best cure would seem to be to take lots of subs, with dithering in between each one, and stack with some kind of sigma-clipping routine.

Cracking image though, even with the streaks

[GarethA]

You can extract the sensor temp from the raw files using an app like exiftool

http://www.sno.phy.queensu.ca/~phil/exiftool/

Peter...

Cheers Peter, yep, I have an exif tool and will pull the temp data and have a look - hadn't really considered it before.

[GarethA]

think I've seen those streaks from other Canon cameras before too, though mercifully not my own, maybe a feature of some of the older sensors ?  There's even a script in Pixinsight to remove 'Canon banding' though I've not used it myself.

Best cure would seem to be to take lots of subs, with dithering in between each one, and stack with some kind of sigma-clipping routine.

Cracking image though, even with the streaks

Cheers Stuart - I hope it's not the camera or I'll have to break open the 'CCD Piggybank' earlier than I was planning to

Am home tomorrow so will try the various DSS clipping and dithering techniques that have been described.

Appreciate the compliment re the image itself - just wish the stripes would go away

[Thalestris24]

Hi Gareth

You have to do dithering when you take the images - it ensures there is a slight movement between subs and helps deal with hot pixels. Not sure what capture software you're using but APT supports auto dithering of PHD2. It would be interesting to see whether your master bias and/or master flat also show the banding.

Louise

[GarethA]

Thanks folks - ok so I now know what dithering is

I don't use capture software yet - just have the intervalometer clicking away.

After a bit of investigation:

It's not the camera. Here is a hyper stretched sub which does show some faint HORIZONTAL banding but my bias frames seem to deal that pretty well.

I pulled the temp data and there's hardly any difference between subs.

Sub length could have done with being a little longer.

The various DSS methods don't seem to have any impact - (though now I've discovered kappa sigma by previously discarded subs with plane/satellite trails are now back in the stack!)

After a bit of processing I've got an image I'm happy with

I'm still not totally sure what caused the issue but onwards and upwards!!

[GarethA]

that last process looked a bit extreme so here's one last one

I'm done

Next step - more data!

[wxsatuser]

Thanks folks - ok so I now know what dithering is

I don't use capture software yet - just have the intervalometer clicking away.

After a bit of investigation:

It's not the camera. Here is a hyper stretched sub which does show some faint HORIZONTAL banding but my bias frames seem to deal that pretty well.

I pulled the temp data and there's hardly any difference between subs.

Sub length could have done with being a little longer.

The various DSS methods don't seem to have any impact - (though now I've discovered kappa sigma by previously discarded subs with plane/satellite trails are now back in the stack!)

After a bit of processing I've got an image I'm happy with

I'm still not totally sure what caused the issue but onwards and upwards!!

Think you answered your own question, longer and possibly more subs.

Hard stretching of less than perfect subs will reveal all the nasties that are hidden.

Thats one reason why I went for fast f ratio scope and lenses, coupled with a decent LP filter,

makes life a little easier in these light polluted skies.