Howdy, Stranger!
It looks like you're new here. If you want to get involved, click one of these buttons!
In this Discussion
M42 and HDRMT - a winning combo?
Adam,
I was struggling a bit on how to approach my current M42 project since I wanted to preserve/display the dimmer cloud details as well as the Trapezium which would normally be blown out. I've gone through some hoops in the past with my older OSC data, including 'staged' stretches, ranged masks combined with 'stepped' stretching all in attempts to preserve the Trapezium to some degree.
This time around, as I've gone to the dark side (monochrome), I had L, Ha, Sii, and Oiii datasets. I also started out copying some prep that I'd seen on Masters of PixInsight, by having a mix of exposures for each filter from "short" at 30s (10s for the L) to 180s for the "long" exposures.
However, after watching your video on HDRMT, I chose a little different path. I setup WBPP to combine the different exposure for each filter into a single master for each filter rather than masters for each filter/exposure combination. I also had all the images registered to a single frame for the entire set, the four masters were aligned and autocropped together (and 2x drizzle).
After DBE and BXT, and creating a starless and stars version of each, I set aside the stars to create an RGB stars frame. The starless frames I combined using the foraxX dynamic SHO pixel math algorithm. I then stretched the combined image to bring up the detail I was after in the dimmer cloud and used the HDRMT tool to 'compress' the image and it did indeed bring up the stars in the core of the nebula but without the artificial (to my eye) darkening of the area.
For the HDRMT, I experimented with a preview, but ultimately found the best results (to my eye) with 2 iterations of the bSpline (3) on 10 layers. I did use Bill Blanshan's modified SCNR pixel math tool (AnotherAstroChannel) and the curves tool to tweak the final image. I think the gradient/visibility of the core stars looks more natural that what I've seen on a lot images, that is user preference obviously. You guidance on the HDRMT tool was what made this work for me, so just wanted to let you know!!!
You can see the final result here: https://astrob.in/6zlrwh/0/
I was struggling a bit on how to approach my current M42 project since I wanted to preserve/display the dimmer cloud details as well as the Trapezium which would normally be blown out. I've gone through some hoops in the past with my older OSC data, including 'staged' stretches, ranged masks combined with 'stepped' stretching all in attempts to preserve the Trapezium to some degree.
This time around, as I've gone to the dark side (monochrome), I had L, Ha, Sii, and Oiii datasets. I also started out copying some prep that I'd seen on Masters of PixInsight, by having a mix of exposures for each filter from "short" at 30s (10s for the L) to 180s for the "long" exposures.
However, after watching your video on HDRMT, I chose a little different path. I setup WBPP to combine the different exposure for each filter into a single master for each filter rather than masters for each filter/exposure combination. I also had all the images registered to a single frame for the entire set, the four masters were aligned and autocropped together (and 2x drizzle).
After DBE and BXT, and creating a starless and stars version of each, I set aside the stars to create an RGB stars frame. The starless frames I combined using the foraxX dynamic SHO pixel math algorithm. I then stretched the combined image to bring up the detail I was after in the dimmer cloud and used the HDRMT tool to 'compress' the image and it did indeed bring up the stars in the core of the nebula but without the artificial (to my eye) darkening of the area.
For the HDRMT, I experimented with a preview, but ultimately found the best results (to my eye) with 2 iterations of the bSpline (3) on 10 layers. I did use Bill Blanshan's modified SCNR pixel math tool (AnotherAstroChannel) and the curves tool to tweak the final image. I think the gradient/visibility of the core stars looks more natural that what I've seen on a lot images, that is user preference obviously. You guidance on the HDRMT tool was what made this work for me, so just wanted to let you know!!!
You can see the final result here: https://astrob.in/6zlrwh/0/
Comments
I have another question that I will post separately since it is a different topic/issue.
But if you look at my M42 with a lot of zoom, I believe you will see it.