Need Help on Processing Comet

[twade]

To all,

After a couple of iterations of DBE using lots of sample points on each channel, several iterations of SCNR to remove a red cast, and a curves transform using the red channel to remove too much red in the tail, I finally have a decent color image to blend with the luminance data.  I also painted most of the remaining stars out of the image.  Here are the imaging details:

Camera:  FLI PL-16803
Lens: Pentax 67 300mm f/4.0 EDIF (used wide-open)

Data:

Red (3x3): 5 - 90s
Green (3x3): 5 - 105s
Blue (3x3): 5 - 135s



I'm hoping to add the luminance data either later tonight or Friday, depending on how much effort it takes to remove the gradients in the luminance channel.  I also plan to do some noise reduction.

Wade

[twade]

To all,

I believe it's ready for the star-field.  I have added the luminance data.  It's still a little noisy, but given what I had to work with I believe it turned out quite nicely.  I appreciate all your help.  Here's the final image details:

Camera:  FLI PL-16803
Lens: Pentax 67 300mm f/4.0 EDIF (used wide-open)

Data:

Luminance (1x1): 180s
Red (3x3): 5 - 90s
Green (3x3): 5 - 105s
Blue (3x3): 5 - 135s

Calibration was done in Maxim DL.  All other processing was in PixInsight.

I plan on taking the background star-field in August when this area is higher in the sky.  Hopefully, it will be clear and smoke free.



Wade

[mmirot]

Well done Wade.

Tricky processing. I have 6 x 45 seconds image in RGB only for this object. I am not sure the S/N will be worth it but I will give this a try.

Max

[RBA]

Hi Wade,

Thank you for uploading the images. Here is my solution to this nice image processing problem.

Juan, where did the screen-shots that came with your post go??

[Juan Conejero]

Hi Rogelio,

Thanks for pointing this out. These files were in the wrong directory after some restructuring of our server. They are back in place now.

Unfortunately these aren't the only forum images that are now inaccessible. I'll try to fix all missing images; if you see other threads where this happens, please let me know.

[FunTomas]

Hi boys. I cannot reproduce Juan's tutorial. Images of comet and comet+stars are so different, that after substraction I got near black image. Could someone help me with this problem? I uploaded set of images to http://endor.uv.es//data/public/5972fad22f61f26123a53a9501f2c6e5.php?lang=en
Many thanks for any suggest.

[astroedo]

Hi boys. I cannot reproduce Juan's tutorial. Images of comet and comet+stars are so different, that after substraction I got near black image. Could someone help me with this problem? I uploaded set of images to http://endor.uv.es//data/public/5972fad22f61f26123a53a9501f2c6e5.php?lang=en
Many thanks for any suggest.

Hi! I have the same problem because the background in Full and Comet images is the same so many pixels are clipped to 0 when I perform
Full-Comet operation to get the stars image.

I've solved changing Juan expressions as follows

for the stars image:

Full-Comet+med(Comet) this is the new stars image

Star removal from full image

x = k*med( stars );
$T - iif( stars > x, stars, stars*(1 - (x - stars)/x) )+med( stars)

I really don't know if it is the right solution, but it works!

bye Edoardo

[FunTomas]

Thank You Edoardo. I try your suggestion as soon as possible. Maybe with data of the comet Lovejoy taken this night.

[astroedo]

I'm trying right now with the same comet 

[cdavid]

Hi everyone, I also am trying this pixel math process on Lovejoy and I'm getting nothing but black.  When I add back the +med(stars) variable I end up with a bizarre "negative" like image of some of the comet.

Any ideas?

Thanks
Carlos

[NKV]

x = k*med( stars );
$T - iif( stars > x, stars, stars*(1 - (x - stars)/x) )+med( stars)

my solution:

x, k=1.1, m
x=k*m; m=med(stars); $T - iif( stars > x, (stars-m), (stars-m)*(1-(x-stars)/x) )