Light Pollution Subtraction

Got an idea for something that SharpCap should do? Share it here.
Forum rules
'+1' posts are welcome in this area of the forums to indicate your support for a particular feature suggestion. Suggestions that get the most +1's will be seriously considered for inclusion in future versions of SharpCap.
Post Reply
VCTyree
Posts: 3
Joined: Thu Jan 09, 2020 7:55 pm

Light Pollution Subtraction

Post by VCTyree » Thu Jan 09, 2020 8:09 pm

I have been using SharpCap 3.2 Pro for a couple of years to do imaging in a Bortle 8-9 urban sky. I typically use live stack with real-time dark frame and flat field calibration. The results have been rather surprising given that the sky is very bright with light pollution.

Over the past couple of years I have been involved with using SharpCap during public star parties in the urban areas around Los Angeles, California. I usually live stack images while members of the public view the objects that I am imaging through a parallel telescope that is aimed at the same object. By using the live histogram stretching feature in SharpCap, I can show the object in color on the computer screen in which I subtract light pollution by setting the black point up scale. Unfortunately, this tactic leads to a severe color bias in the deep sky object image because the light pollution has a strong blue-green color, so the object does not look to be its true color. This is because the black point setting clips the red component more than the blue and green components of the light pollution.

If I take the live-stacked 32-bit FITS image into PixInsight and use the DynamicBackgroundExtraction (DBE) tool to remove the light pollution then I get a fairly nice looking true color image of the object. I post these images for my astronomy club on my Instagram account: @vctyree The problem is that these images are not immediately available and most of the visiting public never bother to look at the processed images the next day after the public star party.

Since SharpCap 3.2 has live Preprocessing to do dark frame and flat frame calibration, I started to think that perhaps I could also remove light pollution in each sub-exposure by subtracting a synthetic light pollution image from the PixInsight DBE tool. I can take a number of live stacked images from a region near where I will be showing the deep sky objects, generate a light pollution image with DBE and perform live stacking or real-time imaging along with light pollution removal to show people the true color object without light pollution. I can also imagine that this technique can be used to show wide field images of the sky to show meteor showers in deeply light polluted urban areas. Would it be possible to include a third preprocessing function that takes a saved light pollution image and subtracts it from the live image after dark frame and flat field calibration?

User avatar
admin
Site Admin
Posts: 4274
Joined: Sat Feb 11, 2017 3:52 pm
Location: Vale of the White Horse, UK
Contact:

Re: Light Pollution Subtraction

Post by admin » Fri Jan 10, 2020 7:46 pm

Thanks for the suggestion. I've been thinking about this a bit over the last day or so and as far as I can work out you are creating a dark and flat corrected background light pollution frame and want to be able to subtract this from the image frames after normal dark and flat correction have taken place.

Right now I'm planning a slightly simpler approach to the same problem in which is to have an option to automatically try to detect and remove background glow and gradient from the image – see this thread : viewtopic.php?f=17&t=529&p=11446#p11446 . The developer of AstroLive has been kind enough to contribute some of the ideas and code that he used to use in that application for this purpose, so hopefully the feature will work well.

It would be interesting to see how it compares to the more sophisticated approach that you are taking – if it looks like the more advanced approach has distinct benefits and there are more people interested in using it then I will consider that as a future feature.

Cheers, Robin

VCTyree
Posts: 3
Joined: Thu Jan 09, 2020 7:55 pm

Re: Light Pollution Subtraction

Post by VCTyree » Fri Jan 10, 2020 10:51 pm

The method would to be to take either live images or live stacked images in which the SharpCap preprocessing applies darks and flats to each image and offer the option of subtracting a light pollution image that is saved in a folder. I currently am using SharpCap to live stack (dark and flat corrected on each sub-exposure) to create a live stacked image with the light pollution included. I later process the stacked image using PixInsight DBE to create a synthesized light pollution image which is then subtracted from the stacked original image. I have a short write-up on the club (LAAS) web site that describes two light pollution removal methods. The LAAS article:

http://www.laas.org/articles/A_Techniqu ... lution.pdf

This article primarily describes a low cost light pollution removal method using free tools (a number of LAAS members did not want to spend 230 Euros on PixInsight), but I describe the more accurate PixInsight-DBE method for comparison.

Since the light pollution brightness is slowly varying over several hours, I thought that a nice feature for our public sessions at the Garvey Observatory to be able to view live images with the light pollution removed. This would entail taking a short live stack of the sky, use DBE to generate the synthetic light pollution frame, and then save it into a folder. A new feature in SharpCap preprocessing would do the current calibrations (flats and darks) followed by subtracting the synthetic light pollution frame saved in a folder. I am not suggesting that this method is as accurate as what I currently do with SharpCap, but it would significantly enhance the public experience for visitors to the Garvey Observatory. We will still have the option of disabling the light pollution function (by not selecting a light pollution image) when doing imaging as before with only flats and darks. My, probably naive, view is that adding an image subtraction function to preprocessing function is not any more complicated than the dark frame subtraction that is already there.

Regards,

Vance Tyree

User avatar
admin
Site Admin
Posts: 4274
Joined: Sat Feb 11, 2017 3:52 pm
Location: Vale of the White Horse, UK
Contact:

Re: Light Pollution Subtraction

Post by admin » Sat Jan 11, 2020 8:49 pm

Hi,

Thanks for the link to your document – an interesting read. Is it really just a simple as adding a second option to subtract a frame where the subtraction occurs after flat correction rather than before it?



Cheers, Robin

VCTyree
Posts: 3
Joined: Thu Jan 09, 2020 7:55 pm

Re: Light Pollution Subtraction

Post by VCTyree » Sat Jan 11, 2020 10:44 pm

The dark and flat correction is absolutely essential before doing the light pollution frame subtraction. This is because there will be a more complex light pollution gradient without the flat calibration. I have found that, even though you can use DBE without flat field calibration, the removal of light pollution may not be as accurate. Take a look at some of the images that I have taken at Garvey Observatory on my Instagram account @vctyree. The older images (like M8 with some residual blue gradient) were done without flat field calibration. The light pollution removal is not as good as the later images, where I live stacked using SharpCap with darks and flats applied to each sub-exposure. I have found that SharpCap is an absolutely essential tool at Garvey Observatory as well as in my back yard, which has only slightly lower light pollution (Bortle 8 versus Bortle 9+ at Garvey).

If we have the ability to subtract a light pollution image taken the same night from each sub-exposure in live stack then there will be a major improvement in the results, even though it is more accurate to do the DBE on the final stacked image. Even if we under correct the light pollution, the result will look much better than what we are currently getting. Also, I can take the under corrected light pollution image and apply DBE again to get rid of the residual light pollution in the final result with the live stacked image. In fact, there is some advantage to having an option to purposely under correcting the light pollution (by multiplying by a factor like 0.95) to avoid an over correction. The under corrected image will at least have the advantage of not having the color error that typically occurs (as described in the article).

Regards,

Vance Tyree

Post Reply