Improved Results from Small GraXpert Processing Change

My DSO images are processed using Autostakkert!3 for stack averaging, GraXpert for gradient removal and Corel Paintshop Pro 2021 (PSP21) for aesthetic touchups. This posting addresses how GraXpert processing. 

GraXpert numbered steps are 1. Load Image 2. Crop, 3. Background Extraction (gradient removal) 4. Denoising 5. Saving (Stretched Processed file). This sequence suggests one should click each step setting any parameters if needed. 

Having assumed the latest denoising version would be the best and fastest I was surprised that version 3.0.1 could take 5 minutes to denoise one image adding incentive for noise reduction in PSP21. 

It turns out Denoise version 2.00 delivers  satisfactory noise reduction in 20 to 30s. Unfortunately the sequence often resulted in excessively stretched images and histograms when imported into PSP21 if the normal sequence was followed.

A simple change to the GraXpert sequence fixes this issue giving me incentive to reprocess many past DSO images. Instead of using PSP21 for noise reduction the GraXpert process is changed as follows. After step #3, Background Extraction the image is saved to a stretched processed tiff file. Stretch is set to No Stretch and the stretched processed tiff file is reloaded. Then denoising is applied. 

Now, when loaded into PSP21 the processed image shows no excessive stretching and the noise is greatly reduced along with improved signal to noise. More nebulosity can be pulled out of the otherwise noisy background. The three examples below show the advantage of the the modified sequence. Older DSO images are being reprocessed using this modification.

Update: Images from original posts were replaced by current images.

Example 1: Cropped image of Messier 94. Processing: GraXpert 30% stretch, AI gradient removal, denoise. Image shows excessive stretching indicated by brightness and noise in dark sky areas.

Example 2: Processing is the same as Example 1 except for the following change. After A.I. gradient removal image is saved and stretching is set to None(0%). Then image is reloaded and denoise is applied.

Example 3 - Processing: Stretched 30%, A.I. gradient removal, GraXpert denoising skipped. Default digital noise removal in PSP21.  Background sky shows more mottling and stars are slightly less sharp than in example 2. 

New Starizona Night Owl Focal Reducer -- Updated

For some time I have used the combo 0.63x focal reducer(FR) and diagonal to achieve about an f/5 focal ratio with some rather distorted stars out to the edge of the ASI 294MC Pro with its 16x13mm sensor. Recently I procured a Starizona Nightowl FR which promises .. "to provide excellent image quality over a 16mm image circle." This is a bit of an over promise. Even near the 13mm dimension edge of the ASI294MC star distortion is quite evident; albeit less objectionable than the aforementioned 0.63x FR optical configuration. 

A bigger concern is uneven background vignetting. This is expected in any optical system which is why flat field correction is essential. However, in just my short experience with the Night Owl there is a small residual greenish glow  evident in the center of the Horsehead Nebula and NGC1981 images even after applying several flat fields acquired at different exposures. Imaging was from  a quite light polluted area, Frederick, Maryland and it's hoped a Orion SkyGlow Astrophotography Filter will help with this. Meanwhile here are a few early results. 

Update: Since this writeup almost all NightOwl images have been acquired at 3104x2116 or 3104x2844 pixel formats. At these formats flat frames are effective eliminating the residual greenish glow in the center of DSO images. I now do all NightOwl imaging at the 3104x2844 format. 

Barnard 33(Horse Head Nebula), NGC2023, 

NGC 2024(Flame Nebula)

Sharpless 279, NGC1981 

It's been a good two years since I posted here. I'm ashamed to  admit I forgot I created the blog. That's not to say I have abandoned the DSI-II. For some reason I no longer remember I stopped posting. Time to give it a renewed try. For the near future posts will catch up on past activities.

Meade no longer produces the DSI-II. There was a DSI-III which has been discontinued along with the full DSI line. In fact I was "lucky" to grab mine for a substantial discount as Meade was closing out the IIs. I put lucky in quotes because the DSI-II is a mixed blessing. It was cheap $290. It does work as you can see from the earlier images.

With its automatic dark frame subtraction it is easy to use. On the other hand the product is really poorly documented and the software interface is poorly designed. Meade left it to an enthusiast to create a video tutorial that is not done under the dark but none the less is more informative than the included documentation. The documentation does little to explain how to use the various settings or what many of them mean. This has been a recent pattern with Meade. The documentation for its Autostar

Whoever coded the software interface did not account for varying screen resolution. Thus the user must futz around with screen resolution settings on a high resolution laptop just to get a usable screen. The lack of temperature regulation is another source of frustration on nights when temperatures constantly change. One is forced to keep generating new sets of dark frames as the device cools along with the night air. A few users have modded their DSI-II with Peltier cooling. I'm not yet brave enough to try that.

Under a full moon

June 16 was just one day past full moon. Between 10 and 11PM the moon was still low to the southeast which offered a test for using the DSI-2 under fairly adverse conditions. Randolph, MA where I live is already very light polluted with poorly shielded sodium vapor lights. Adding to the challenge, the  local elementary school in whose parking lot I do my local observing has both broadband mercury vapor along with sodium vapor lighting.

Messier 3 just past full moon

After setting up the scope it took longer than usual to iterative align the 127ED because I forgot about the problems posed by using Arcturus or Spica as alignment stars. I need to recompile a list of compatible alignment stars which which differ between 3 and 9 hr away from Polaris in R.A.


I chose M3, which rides high in the sky this time of night to image. The globular cluster was obvious at 1s exposures in focus mode. After  creating a fresh set of darks for the night and merging with my old set of darks I acquired two sets of exposures at 2.8 and 4.0s. The 4.0s set were averaged to produce the final image after some histogram, brightness and contrast tweaking.