IC1396 and the Elephant Trunk Nebula in the constellation Cepheus - Processed in the Hubble Color Pallet. This rather large region in the northern sky is hoome to a wide variety of both ionized gasses emitting light as well as dark dust clouds forming interesting globule shapes throughout the nebulosity. The long dark structure in visible seems to resemble an elephant's trunk, coining it's common name. Most of my favorite parts of this nebula are the dark globules through the mid-upper and lower regions of the image.

This has been a summer-long project as I was testing many different exposure times, gains, offsets when I first started to dive into narrowband imaging with the 135mm f2 lens, and getting a good focus has proven to be more and more elusive each time out, so I've got an auto-focus solution on route (and it sure can't get here soon enough, I'm quite excited!!)

A quick calculation of all of the data taken on this target shows over 17 hours of data in both Ha and Oiii, and almost 8 hours of Sii. This image represents the best 31 of that total 42 hours of data. I've ended up with a consensus of shooting 5 minute subs anytime the moon is up from my Bortle 4 backyard with this setup, and 10 minute exposures when the moon is away at half-unity gain (53 offset 10.)

Equipment:

- Rokinon 135mm F/2 at F/2

- ZWO 8-position EFW

- Astrodon Narrowband Filters (5nm Ha, 3nm Oiii, 5nm Sii)

- ZWO ASI183mm Pro

- Celestron CGEM

- Intel Compute Stick (STK1AW32SC): Reformatted to 64-bit

- Unguided

- Acquired with Nighttime Imaging N Astronomy (N.I.N.A)

- Thousand Oaks Optical Dew Controller and Dew-Not Dew Strips

For each filter's set of data all images were calibrated with corresponding master darks master flats and with an output pedestal (value 150) to prevent black-pixel clipping. They were weighted using the equation below using the Subframe Selector process:

```

20*(1-(Eccentricity-EccentricityMin)/(EccentricityMax-EccentricityMin))

+ 30*(SNRWeight-SNRWeightMin)/(SNRWeightMax-SNRWeightMin)

+ 40*(Stars-StarsMin)/(StarsMax-StarsMin))

+ 10

```

Each set of images were integrated using the *Average* combination, Normalized using *Additive with Scaling*, used the *Linear Fit* rejection algorithm with Sigma High and Low of 10.0.

Ha Processing:

```

let ha =

crop edges

|> linear fit to Sii

let decon=

ha

|> Sharpen

|> Deconvolution iterations:60 deringing:none regularization:5 layers of decreasing amounts/thresholds

|> replace stars using a star mask and a copy of ha

|> with the star mask above MMT Sharpening to bringing out features

let nr =

ha

|> TGVDenoise mask: inverted low contrast copy of ha

|> MMT mask:very high protection luminance mask

let blendMask =

nr

|> stretch using screen transfer -> histogram tool

|> stretch more clipping white point and black point

let l =

blendMask*decon+~blendMask*nr

// adjust the blend mask using the midpoint slider on the histogram tool

// until the noise reduction targets a blend of the low-signal areas and

// the sharpening is applied to the high-signal areas.

|> Stretch with Histogram Transformation



```

Oiii and Sii were processed with a similar yet simplified workflow:

```

let result =

(Oiii,Sii)

|> TGVDenoise mask: inverted low contrast copy of the image

|> MMT mask:very high protection luminance mask

|> Stretch with Histogram Transformation

|> Morphological Transform using Morphological Selection to shrink stars

|> Star Alignment using Distortion Correction and 2-d surface splines to attempt to fix flaring stars around the outside of the image



```

Color Combination:

```

let sho =

Red<-Sii

Green<-Ha

Blue<-Oiii



```

The following series of steps were then taken to produce the resulting image.

- A set of color masks were created as outlined in the [SHO Processing tutorial](https://www.youtube.com/watch?v=3kZPm8wRYgQ) by Christopher Gomez. To this day I've had real trouble incorporating Ha into the color of my SHO data, but this technique has really helped me include more variety of colors that I haven't been able to be happy with until now. (Thanks Christopher!)

- After getting the color the way I like it, I applied the stretched Ha image as luminance

- A number of iterations to reduce the larger red circles around the brighter stars in the outer edges of the images were done including using the color saturation with a star mask (generated using starnet++)

- Additional contrast was added by using Local Histogram Equalization and HDR Multiscale Transformation with masks

- The stars were reduced using a contours based mask and morphological transformation

- The final crop and brightness shift was done in Affinity photo.

A rough timelapse each step since combining to SHO has been uploaded [here](https://www.youtube.com/watch?v=UlZcw0eAqoI)

Total Acquisition (including data rejected/omitted):

Ha

- 60: 594 files = 09:54:00

- 120: 3 files = 00:06:00

- 240: 97 files = 06:28:00

- 300: 2 files = 00:10:00

- 600: 5 files = 00:50:00

- **Total: 701 files = 17:28:00**

Oiii

- 120: 180 files = 06:00:00

- 240: 119 files = 07:56:00

- 300: 43 files = 03:35:00

- **Total: 342 files = 17:31:00**

Sii

- 240: 73 files = 04:52:00

- 300: 6 files = 00:30:00

- 600: 15 files = 02:30:00

- **Total: 94 files = 07:52:00**

Subtotal Acquisition Time: 42 hours 51 minutes