0.00
...
·
·
1
like
|
|---|
Please critique my image of LBN 437 https://www.astrobin.com/full/03c4yf/0/ This is about 16 hours of OSC data from Bortle 3-4 rural skies. I used a Vixen R200SS with QHY268C and 10 minute subs at -10 Stacking was done with Astro pixel Processor followed by Graxpert and then in PI I did SPCC Blurx GHS Star reduction Noise reduction  |
7.22
...
·
·
2
likes
|
|---|
By the looks of it, I think you used to much noise reduction. If it is too noisy with less noise reduction, that means you need to cool down on the stretching. Also, to me, I think the stars need a little more increased highlights in the histogram and less stretching.
|
8.34
...
·
|
|---|
The diffraction spikes do not look right, especially the small ones, probably from the stretch but could also be from star reduction in BXT. Try removing the stars before stretching. I think the noise reduction is OK as is although could maybe use some tweaking.
|
10.40
...
·
·
1
like
|
|---|
My thoughts… definitely some wierd things going on with the stretch here. Did you use starless processing? Also, there is some very noticeable artifacting around the bright stars. The spikes also don't seem to draw out from the core, theres a disconnect there. Not sure how that one happened. The dust has lost alot of color for some reason. Personally, the denoise may be too much, but I think HAD the data been stretched a little better, it wouldn't stand out as much as it currently does. This is something even I'm guilty of, trying to stretch the data too hard. Relax your stretch, process your image starless, and try again. I think this image has alot of potential and with some effort, you might have a banger here. |
|
0.00
...
·
|
|---|
thanks, so I should create my starless image, and then simply stretch on that, and then add back the stars. So BlurX should be done on starless image?
|
18.64
...
·
|
|---|
thanks, so I should create my starless image, and then simply stretch on that, and then add back the stars. So BlurX should be done on starless image? That’s how I do it. I clone the image after DBE and CC and process the clone for the stars—stretching it separately and then creating the stars-only image from that non-linear image. That gives me fine control over the impact of the stars in the final image. I personally loathe starless images. But I’m fine with trying to balance their contribution. As for the original as-yet-unstretched image, I use the PSFImage script to calculate the PSF for a preview of about the central third of the image as per Russ Croman’s recommendation. Then I remove the stars, apply BXT, then NXT, then stretch. |
5.10
...
·
·
1
like
|
|---|
There are several approaches to processing the image, and mine is to use SPCC to get the best color possible based on your sensor and filters. If necessary I do DBE and then as Tim suggested use PSFImage or DynamicPSF on the unstretched image with stars and use BXT on a small preview to inspect what the stars and details look like before applying the full BXT. I then stretch very carefully in small steps and separate stars for processing nebulosity and stars separately on the RGB. I do not use GHS alone but combine it with other stretches and iterate in small steps, ArcSignStretch helps intensify color but has to be used sparingly in combination with HT and GHS or AutoHistogram. Once I am happy with the color I create a separate Lum either by extracting Lum from the RGB or combining the channels to bring out best detail, and work on sharpening that using PS or PI tools and use LRGBCombination when I am ready to combine. All this is done starless. I finally get the stars balanced with the nebula and this is a trial and error process too and combine and do some final cleanup. |
|
0.00
...
·
|
|---|
Okay I guess I need to up my processing. Will watch some videos since its all french to me  My current process is 1. Stack in Astro Pixel processor - I have finally gotten to some fine tuned settings which give me better results than WBPP 2. Run Graxpert - I verify my images to be gradient free 3. Open in pixinsight, and run the solver 4. Do SPCC 5. Remove green 6. BlurX 7. I use GHS for streching 8. Run Starnet II to get starless image and then use pixelmath (Saw Luke's video) 9. Run NoiseX Starnet II and NoiseX require stretching I think as per Luke's videos. |
