Rob Hawley's Pages Almaden Observatory Main Page

Messier 27 in Narrowband Using Drizzle

June 2017

Messier 27 (Dumbbell Nebula) is one of the most popular Summer Objects.  Using filters its bright O III emission is easy to show even with a first quarter moon.  I have seen pictures in Narrowband of this object and was curious to see what I could do on my own.

Since this object is rather small (8.0' × 5.6') for my wide field setup this was an opportunity to try the Drizzle Algorithm that was included in PixInsight last year. I will let you follow the links for a complete explanation, but here is the Cliff's Notes.  For large pixel cameras like mine I lose resolution.  Since I also dither (moving the camera between images) as a noise reduction technique, some of the addition resolution can be recovered using Drizzle.  Normally expanding an image 2X will pixelate an image.  Using Drizzle I can mostly avoid this.  Thus instead of a 3.5" pixel I have closer to 1.75" resolution which allows me to see finer detail.

I was originally going to do this project as a classical SHO (Hubble Palette).  The Sulfur signal was rather weak though.  Instead of over stretching it I decided to gather Nitrogen and then combine the Nitrogen with the Sulfur multiplied by 2 instead of by 5 as I was doing with pure Sulfur. I also mapped Oxygen to Red rather than Blue since I liked the resulting colors better. I decided that the resulting image was balanced and did not do further channel balancing.

Note the jet of Hydrogen/Nitrogen.  Note also the distinct ring around the edge of the bright nebula and the structures inside.  With this image M27 looks more like a classical planetary nebula.

Drizzle x2

M27  Red=Oxygen  Green= Hydrogen  Blue= Nitrogen+2*Sulfur

Non Drizzled Image

non Drizzled Image

Red= Oxygen Green = Hydrogen Blue = N II + 2 * S

There are a couple of things that I notice from the processed image.  First is the halo of Oxygen.  This is not visible when viewing with your eyes likely because it is too dim.

The second is the classical Dumbbell shape is not visible.  That is likely a processing artifact as I demonstrate below.  Our eyes can detect many levels of brightness.  With a camera I can only display a limited range of brightnesses.  The tools that I use to select that range have a preference for edges.  Thus they tend to detect and enhance structures (like the fine detail seen where the dumbbell is usually seen).  Only the Sulfur image showed something like what is seen visually.  These are the original images for each filter stretched to make the nebula visible.

M27 Hydrogen

M27 Oxygen

M27 Nitrogen

M27 Sulfur


Recovering the Dumbbell

As I said above the enhancement techniques I usually use bring out edges.  This is important to show fine structure detail like the turbulence of an expanding gas cloud.  To get something closer to how our eyes worked I took the Hydrogen image and applied LocalHistogramEqualization.  PixInsight provides this description of the algorithm.

Histogram equalization takes the histogram and computes a transfer curve, which grants more brightness range to higher histogram peaks and less brightness range to histogram valleys. In other words, large areas of similar brightness get more contrast. Local histogram equalization works on individual pixels and computes a transfer curve from the histogram of a pixel neighborhood.

LocalHistogramEqualization brings out the classical shape

M27 Hydrogen processed like visual image

Nitrogen vs Hydrogen

On first glance the Nitrogen and Hydrogen images are similar.  To prove I was actually capturing something different I created this image.  Red in this case are areas were there is more Nitrogen.  Green are areas where there is more Hydrogen

M 27 Nitrogen vs Hydrogen

Clearly the distributions are different which validates that the filters are recording something different.

Processing Details

Data Collection was fairly routine.  My guide camera failed so I had to capture this project using my AP1100 AE only using APCC Pro tracking.  Few images had to be tossed due to Eccentricity which shows how good the tracking is.  The resulting image was rather noisy and would have benefited from another 5 good images of each filter.  Unfortunately I ran out of time.

 # Images
Hydrogen (3nm)
All images were processed by Pixinsight. 

Most of the processing was for noise reduction.  To tease the structure out of the bright central block I used "Gaussian (5)" HDRMultiscaleTransform. This is a method I have not used before.  The other transforms resulted in an image that looked overprocessed.

I built nebula masks using MLT residual layer and then clonestamped out the large stars.

TGVDenoise was used in both the linear and non linear phases to manage noise.

Copyrights For Photos

Creative Commons License
(c) 2017 Robert J Hawley.
Except as noted,all work on this site by Robert J. Hawley is copyrighted under a Creative Commons Attribution-Noncommercial-Share Alike 3.0 United States License. This permits the non commercial use of the material on this site, either in whole or in part, in other works provided that I am credited for the work.