Skip to Content

Photoshop

Topaz Labs DeNoise AI Review

The goal of Topaz Labs DeNoise AI is to reduce digital image noise while preserving detail and increasing image sharpness. If you’re no stranger to astrophotography image processing, that almost sounds too good to be true.

I was skeptical of the application myself but now find myself using it in some capacity for every astrophoto I process. I believe that Topaz Labs DeNoise AI is a powerful tool that amateur astrophotography enthusiasts should consider adding to their bag of processing tricks.

In fact, I even added a new section about Topaz DeNoise AI in my premium astrophotography image processing guide. I have no interest in spending time learning and writing about tools with a short shelf-life, and I don’t see myself shying away from DeNoise AI anytime soon. 

Until discovering this photo noise reduction software, I typically used the built-in noise reduction features of Adobe Camera Raw. This works well enough but is not nearly as reliably effective as the one-click, automatic noise reduction function of DeNoise Ai.

Whether you use Topaz Labs DeNoise AI to batch process a handful of night sky images in Lightroom or use it to subtly control the noise in your deep-sky images in Photoshop, this advanced learning software feels like it was built for astrophotography.

If you have experience using other noise reduction tools such as Skylum Luminar, Noise Ninja, DXO Prime, or any other of the many options available, please let me know how my results with DeNoise AI compare in the comments. 

Photo noise reduction software

Download: Topaz DeNoise AI 30-Day Trial

Topaz Labs DeNoise AI Review

As you may already know, noise can be a big problem for amateur astrophotography, especially when using a DSLR or mirrorless camera with a high ISO setting. I have personally been battling noise in my astrophotography images for nearly a decade. 

Even with great image acquisition best practices, lots of data, and a cool astronomy camera, you will likely still need to minimize the noise in your image at some point in the image processing stages of astrophotography.

In general, the tricky part of applying a noise reduction filter to your image is the possibility of detail loss, which can create a blurry looking image.

Noise Reduction

I originally shrugged off anything to do with this software due to the fact that I am a diehard Adobe Photoshop fan. I realize that many people are not willing to fork out the dough for a Creative Cloud subscription, but I certainly am. I’ve been using Adobe Software for nearly 20 years, and get full value out of my subscriptions to Photoshop and Premiere. 

I’ve used many third-party plugins with Photoshop in the past (see the resources page for a few that stand out), but the Topaz Labs DeNoise AI plugin is different. I don’t believe any of the software I have installed in the past actually tapped into the benefits of artificial intelligence that continues to get better over time. 

Topaz Labs DeNoise AI allows you to download a full-featured 30-day trial, so you’ve got nothing to lose. I took advantage of this opportunity, fell in love with the tool, and gladly forked out the cash for a lifetime license of the software. 

In this post, I’ll explain how to incorporate Topaz DeNoise into your astrophotography image processing workflow, and why I think it’s a no brainer for anyone that spends as much time photographing space as I do.  

Is it a Photoshop Plugin?

First things first, you’ll need to install the software on your computer. Topaz DeNoise is available for both Mac and PC operating systems. For better or for worse, I am a diehard PC user, and I installed the lightweight software on my Windows 10 desktop.

I am using Topaz Labs DeNoise AI as a Photoshop 2020 plugin exclusively. Users have the option of using the DeNoise AI as a standalone program, or a plugin in Adobe Lightroom or, of course, Topaz studio.

I don’t typically utilize the batch processing feature of Topaz DeNoise AI and prefer to invoke the tool on a per-image basis when noise reduction is needed.

results

A before/after example of Topaz DeNoise AI on the Orion Nebula

As far as astrophotography goes, I am sure that most users will use the tool as a Photoshop plugin as I do, during the post-processing stages of their workflow. However, daytime photographers may find the tool handy when editing photos in Lightroom.

This is one of the biggest reasons why I was able to introduce the Topaz Labs DeNoise AI into my existing workflow so effortlessly. I simply run the tool on a new layer in Photoshop just as I would with any other Photoshop filter or external plugin.

Topaz Labs DeNoise is available as a plugin for the following applications:

  • Adobe Photoshop
  • Adobe Lightroom
  • Corel PaintShop Pro
  • Serif PhotoPlus

I absolutely love the simplicity of the user interface (UI). At its core, the software has to be good at one thing, and that’s effective noise reduction. The minimalist interface, large buttons, and massive preview window make it very easy to see what’s taking place. 

DeNoise AI Photoshop Plugin

The Problem with Noise in Astrophotography

Many factors come into play when assessing the reasons why your astrophoto is so noisy. Sensor design, size, and the camera settings used when the photo was taken are the most obvious. 

Noise is the unwanted, randomly placed “grain” caused by your digital camera’s sensor. Noise often increases when using higher ISO settings, and appears as uneven color and grainy pixels distributed throughout your image.

It is most noticeable in solid areas of color, especially darker areas such as a black night sky. This can be a major problem for astrophotography, as our images usually have plenty of dark sky areas in them, surrounding the subject. 

When you attempt to brighten your astrophoto to reveal faint structures of nebulae, stars, and galaxies, you also risk increasing the presence of noise. 

When you zoom in on your image at 100% magnification, don’t be surprised to find grainy, random color specks, and overall inconsistency in the darker areas

The two ugly sides of the noise equation are Luminance Noise and Chrominance Noise. This article will help you understand what causes noise in digital photography.

The Signal-to-Noise Ratio

The best way to reduce noise in your astrophotography images is to improve the signal-to-noise ratio (SNR). Meaning, capture as many sub-exposures as possible and integrate the data using software such as DeepSkyStacker, Astro Pixel Processor, or PixInsight.

By nature, long-exposure photography in dark situations is just asking for noise. But amateur astrophotographers have found ways to overcome this issue to a large extent through image stacking, and calibration frames such as dark frames

Yes, this practice is essential for creating a quality astrophoto, but sometimes it just isn’t enough. I’ve found myself in situations where over 5 hours of data were collected with my DSLR camera, and the healthy SNR still wasn’t enough to achieve an image with a smooth, sharp background. 

A neutral, smooth background sky is one of the toughest challenges in the world of astrophotography image processing. Finding ways to isolate this element of the image, and tame the noise and unsightly artifacts found within it is essential. And this is exactly where Topaz Labs DeNoise AI shines. 

The chroma noise reduction feature is especially handy in these situations, but monitoring your image as a whole during this step is a must. I can’t stress enough the fact that masking (see my Select and Mask tutorial), isolating, and defining each element of your astrophotography image is paramount for success.

How to Use Topaz Labs DeNoise AI

The software is simple to download and install, and automatically integrates itself into Adobe Photoshop. There is no need to drag-and-drop specific files into the program folders as you do with other third-party plugins. 

At its core, DeNoise AI is a photo noise reduction software. More specifically, the tool was designed to enhance sharpness, remove chroma noise, and eliminate noise without losing details. The official Topaz DeNoise user guide includes a complete list of primary functions.

It’s safe to say that astrophotography was not the primary intended use for Topaz DeNoise AI, but it happens to benefit from it a great deal. The horizontal and vertical banding lines, black level correction, and high ISO image recover features are the most utilized aspects.

To run the plugin in Photoshop, you simply need to navigate to the Filter drop-down menu. At the bottom of the list, you should see Topaz Labs > Topaz DeNoise AI. If this option is greyed out, make sure your image is in 16-bit or 8-bit mode (it will not work in 32-bit mode). 

Upon activating the plugin, you’ll be delivered with a large preview window of the image with the noise reduction effects applied, next to your original (split view). 

preview image

You’ll notice that the software interface presents you with two processing models: DeNoise AI and AI Clear. You’ll want to use the DeNoise AI option, as it is the newer feature and tends to work a lot better than the previous AI Clear mode (which was a feature of the original Topaz Studio).

With the DeNoise AI model selected, you can now select your noise reduction mode. I recommend trying out the “auto” mode on your image, as applying this filter to your picture often does a fantastic job (you can thank millions of AI training images for this).

DeNoise AI user interface

The Manual Mode gives you much more control over the noise reduction filter. You can independently adjust the following sliders to your taste:

  • Remove Noise
  • Sharpen
  • Recover Original Details

Best Practices for Astrophotography Images

Deep-Sky Astrophotography

I primarily shoot deep-sky astrophotography images through a telescope, of distant nebulae and galaxies. For these types of photos, Topaz DeNoise AI is an incredible tool to help remove an uneven background sky. 

The chroma noise reduction slider is especially handy when dealing with an astrophoto with a noisy, unevenly colored background sky.

I’d still mask the nebula or galaxy (and in most cases, the brighter stars) in the image before running Topaz DeNoise, but you can count on the DeNoise filter to improve the ugly luminance and chroma noise in the darker regions.

The power of DeNoise Ai in these situations should not be understated. Astrophotographers (particularly ones shooting with a DSLR camera) have been dealing with this problem for many years, and this software handles it better than any other tool I’ve ever used. 

nebula example

Nightscape Photography

I’ve stated how impressive the tool works for images of nebulae and galaxies, but what about a wide-angle shot of the Milky Way? Well, I tested that scenario on an image of the Milky Way I captured under dark skies with a Rokinon 14mm F/2.8 lens. 

As you can see, Topaz Labs DeNoise AI did a fantastic job of preserving important details of this demanding image, while making a noticeable improvement to the overall aesthetics by smoothing out grainy areas of the picture.

DeNoise AI example on the Milky Way

I don’t think that this post-processing tool can totally make up for an image shot using an ISO setting that was too high, but it may be able to save some of the images you originally thought were just too noisy to share.

Usage Tips

The best part about Topaz Labs DeNoise AI is how simple it is to use. Most often, I utilize the “auto” feature of the plugin and apply the default AI noise reduction amount to my image. It’s best to apply this action to a new layer on top of the image, so you can adjust the opacity and overall impact of the effect.

The “Auto” Noise Reduction setting works surprisingly well most of the time. 

Usually, I will apply the DeNoise AI layer at approximately 65% over the entire image. Then, it is wise to monitor any negative changes to the look of the stars in your image. I have found that sometimes the DeNoise AI plugin will change the shape of small-to-medium-sized stars, and even create “holes” in them.

To avoid this scenario, you have a few options. You can create a layer mask in Photoshop to protect all of the stars in the image before running Topaz Labs DeNoise AI.

You could also (this is my favorite method), apply the DeNoise action to the entire image globally, and then use the eraser brush (set to the opacity, softness, and size of your choice) and carefully reveal each element in the image from beneath the DeNoise layer.

This may seem like a painstaking process, but I have found this way to offer the most control.

california nebula

DeNoise AI has helped me revive some of my noisier images from the archives.

AI-Powered Noise Reduction

I know that most of you are no stranger to the concept of machine learning. Essentially, the team at Topaz Labs trained an AI model using specific filters to produce remarkable results.

This process is well-beyond my capabilities and understanding as an amateur astrophotographer, but that doesn’t mean I can’t leverage this AI power to create better images.

This article explains how Topaz Labs used millions of clean and noisy images to train the software to understand what to remove (chroma and luminance noise), and what not too (important details, colors, and sharpness). 

To witness this AI technology in action, all you need to do is appreciate the automatic results of the DeNoise AI plugin on one of your noisy astrophotos. 

landscape images

System Requirements

A number of people have reminded me that you should consider the system requirements of this software before purchasing the plugin.

I am currently using Topaz DeNoise (version 2) on my Windows 10 Desktop with an Intel Core i7-9700K processor @ 3.60 GHz, and 16 GB of RAM. The Graphics Card is an NVIDIA GeForce RTX 2060.  

On my machine, the DeNoise AI preview window updates really fast and applying the filter takes about 1-2 seconds. However, slower machines will inevitably bring this process to a crawl, so I looked into the system requirements Topaz recommends.

CPU Minimum

  • Intel i5 or equivalent (3.0GHz and above)
  • AMD Ryzen 5 or equivalent (3.0GHz and above)

CPU Recommended

  • Intel i7 or greater (4GHz and above)
  • AMD Ryzen 7 or greater (4GHz and above)

Graphics Card Minimum

  • NVIDIA 2GB of dedicated VRAM (GT 740 or greater)
  • AMD 2GB of dedicated VRAM (Radeon 5870 or greater)

Graphics Card Recommended

  • NVIDIA 4GB of dedicated VRAM (GTX 970 or greater)
  • AMD 4GB of dedicated VRAM (Radeon RX 460 or greater)

RAM

  • Minimum: 8GB
  • Recommended: 16GB
  • Optimal: 32GB

If your computer meets the recommended requirements for the software, Topaz Labs states that ‘Users should experience no performance issues, though slowness may occur with large files”. I have found this to be true in my experience using the software on a fast Windows 10 PC. 

Final Thoughts

Processing astrophotography images can be challenging, especially when you haven’t collected the amount of data you hoped to. It would be nice to acquire 10+ hours of exposure time on each target, but that isn’t realistic for a lot of people.

Topaz Labs DeNoise AI allows you to make the most of your data integration. If you’re sitting on a handful of images that aren’t quite “post-worthy”, run them through DeNoise AI and give them a second look. 

One of the most exciting parts of this experience, for me, was going through some old astrophotos and seeing which ones I could potentially recover. Tools like DeNoise AI have the ability to breathe new life into old projects.

To justify the cost of a new software tool, you need to receive a lot of value from it. I truly think that you will enjoy using Topaz Labs DeNoise AI on your astrophotos now, and in the future. 

M82 Galaxy

Helpful Resources:

Related Posts:

Related Tags

Remove Gradients in Your Astrophotos with Photoshop

|Tutorials|8 Comments

Adobe Photoshop is the preferred weapon of choice for many astrophotographers of varying levels of experience.  The intuitive user interface and limitless image processing capabilities make it a real contender in the astrophotography world.

The seamless integration with the .RAW image files produced by a Digital Camera makes Photoshop an attractive choice for photographers using popular Canon and Nikon DSLR’s.

It continues to be my personal favorite tool for processing astrophotography images.

The M78 nebula in OrionWhether you are brand new to astrophotography image processing, or a seasoned veteran, an uneven field in your image is something that every astrophotographer will deal with at some point.

The steps I will discuss below can be done in Photoshop without using any additional plugins. However, I strongly recommend investing in the Astronomy Tools Action Set, and Gradient Xterminator.  They are well worth the expense and can make a monumental difference to your images.

This Photoshop tutorial involves the following:

  •   Assessing your uneven field
  •   Removing the DSO from your image
  •   Creating a synthetic flat frame
  •   Subtracting the flat frame from your image

 

An Effective Photoshop Technique for Removing Gradients

One of the most time consuming and frustrating stages of your image processing workflow can be dealing with gradients.  Your background sky goes from a dark blue to pink as the encroaching glow of city light pollution stains your image.  Luckily, there is an extremely useful and effective method for removing gradients using Photoshop.

This method involves creating a synthetic flat frame and subtracting it from your original image.

Quickly correct your uneven field

The method you’ll see me use in the video below is a very popular way to remove gradients using Photoshop.  Variations of this technique have been used by amateur astrophotographers for years.  I do not take credit for this method.  Like almost everything else I have learned about this hobby, I picked this up by watching and reading countless image processing tutorials shared by others.

Video: How to Remove Gradients in Photoshop:

This technique works better on some deep-sky images better than others.  Large targets such as nebulae that fill the entire frame will be difficult to tackle using this process.  In my example, the Leo Triplet of galaxies worked very well, as they are surrounded by a large area of surrounding space.

Assessing the Data

  1. Start by opening up your final stacked image.  I use DeepSkyStacker to register and stack all of my image frames.
  2. Crop your image to remove the stacking artifacts and overlapping frames.
  3. Convert the image from 32bit to 16 bit, to open up further editing options in Photoshop.
  4. Perform a quick levels adjustment, bringing the left-hand slider up against the data on the histogram.
  5. Make a curves adjustment, pulling the details contained in your deep-sky object forward.
  6. By now, you should have a good idea of how bad the vignetting and color gradients are in your image.

 

Gradient issues in a astrophotography image

A curves adjustment will show the uneven field

Removing the DSO from the image

Now comes the fun part.  This is where you either have the option of running a third-party plugin such as Gradient XTerminator or tackling the issue yourself.  It’s beneficial to learn this method of removing gradients in photoshop for all types of astrophotography including wide field Milky Way shots.

  1. First, copy your original image layer and paste it on top.  Name it “Gradients”
  2. Copy this layer to a new image. Select All > Copy > File > New > Paste.
  3. On the new image that was just pasted, remove the deep sky objects from the field of view.

This can be done various ways, but I prefer to use the healing brush.  The important part to remember is that we are only interested in the color information of the background sky.  We don’t want to change the data found in the deep-sky objects themselves.  See this in action in the video above.

 

Using the healing brush in Photoshop

Remove the DSO using the healing brush

Creating a Synthetic Flat Frame

Now that we have a version of our image without our deep-sky object(s), we can correct the uneven field in the background sky.  At this point, you may also want to remove any bright stars that may negatively affect the resulting synthetic flat frame.

Richard Hum had this to say on YouTube:

What I usually find helpful is to use Select > Color Range > Highlights to select the stars, and then do a content-aware fill. I find it works better than not removing the stars and just doing dust and scratches. You can use the Select and Mask tool to refine your selection mask.

  1. Now, we need to blur the details of our copied DSOless image. Choose Filter > Noise > Dust and Scratches.
  2. For my camera’s resolution in the example, a Radius value of 80 pixels was used, and a Threshold of O.
  3. You should now see a blurred version of the background sky, with an evident uneven field.

 

creating a synthetic flat frame

Our synthetic flat frame

Applying the Flat frame to your Image

  1. Now, go back to your original image, and make sure you have the “Gradients” layer we created selected
  2. Next, choose Image > Apply Image.
  3. From the Source drop-down menu, select the copied, blurred image we just created. (Untitled-1)
  4. From the Blending Mode dropdown, select Subtract.
  5. Leave the Opacity at 100%, and set the Offset to 30 and hit, OK.

 

Deep sky astrophotography

Your new and improved image

Your new image with the gradients layer on top should look much better.  The “Gradients” layer we created can be scaled back by using the Opacity slider on the layer.  You may not need to use this layer at 100% to completely correct your gradient issues, but expect to have it set to between 80%-100% in most cases.

This layer can be toggled on and off to review and inspect the improvements to your image.  If necessary, you can go back and test some of the variables including changing the Radius value, and/or removing the stars before blurring the frame.

From this point, you can go about your image processing as you normally would, with a much improved, even background sky.

Wide field images captured with my camera lens suffer from horrible vignetting in my backyard.  The gradient removal technique above was used on this image of the Orion constellation to correct the background sky:

Orion constellation

The Orion constellation from my backyard

Try this method on some of your existing widefield images that suffer from a gradient in the background sky.  An uneven field is a common problem in almost all astrophotos, so mastering this technique will come in handy in your future endeavors.

Did you know you can sell your astrophotos as stock photography?  I have sold several of my images on Shutterstock over the past 3 years.  View my portfolio.

You can stay up to date with the latest images and information on the AstroBackyard Facebook page, or by following me on Twitter and Instagram.

Until next time, clear skies!

Related Posts

Astrophotography Tutorials – AstroBackyard

Astrophotography Tutorial – Deep Sky Image Processing in Photoshop

Galaxy Season Target – The Leo Triplet of Galaxies

My Complete Deep Sky Astrophotography Equipment Setup

Beginner Astrophotography Telescopes – My Top Picks

Resources:

Astrophotography Tutorials – PhotographingSpace.com

Gradient Xterminator – Photoshop Plugin

Astronomy Tools Action Set – Pro Digital Software

Related Tags

Deep Sky Image Processing in Photoshop

Well, this is it.  In this deep sky image processing tutorial, I’ll be combining all of the data I was able to collect on the Orion Nebula this winter.  As we transition into Spring, a new array of deep-sky imaging targets will present themselves.  The winter astrophotography targets in the Orion constellation will have to wait another year to get photographed.

The camera used for this image was a Canon EOS Rebel T3i (600D), an excellent choice for beginners looking to dive into deep sky astrophotography.

Deep Sky Image Processing

Processing Walkthrough – Orion Nebula with a DSLR

Canon DSLR for astrophotography

The total amount of detail I was able to capture on M42 this winter was 3 Hours and 8 minutes of color RGB data.  I will be incorporating 2 hours and 40 minutes of Ha data into the final image using the HaRGB processing technique.  In this post, I’ll show you exactly how I process my image of the Orion Nebula using Adobe Photoshop.  I’ll start with the Autosave.tif file produced by DSS.

Some of the images used in my final photo were shot during the AstroBackyard YouTube video: Let’s Photograph the Orion Nebula.

DeepSkyStacker

The screenshot below shows the results of registering and stacking 4 nights worth of imaging from my backyard.  This winter has been plagued with numerous cloudy nights, so I had to capture photons here and there, under varying sky conditions.

Yes, it is very white!  That’s light pollution for you.

DeepSkyStacker

Orion Nebula stacked .TIF file in DeepSkyStacker

The photo sets from each imaging session were loaded into the group tabs of DeepSkyStacker.  My modified Canon T3i camera was set to ISO 800 for each imaging session, but I bumped the exposure time up to 3.5 minutes for the fourth and final set.

Using the group tabs in DSS

  • Dec 22, 2016 – 23 frames – 180″ @ ISO 800
  • Feb 2, 2017 – 24 frames – 180″ @ ISO 800
  • Feb 3, 2017 – 10 frames – 180″ @ ISO 800
  • Feb 27, 2017 – 11 frames – 210″ @ ISO 800

Image sets 1-3 were stacked using darks, bias and flat calibration/support frames. The final and fourth set did not use flat frames as I was not able to shoot them the morning after the imaging session.

I do not make any adjustments to the stacked image in DeepSkyStacker.  I bring the 32-bit Autosave.tif file into Adobe Photoshop for all post-processing.

Processing in Adobe Photoshop

I use two Photoshop Plugins in this tutorial, Astronomy Tools Action Set, and Gradient Xterminator. See all of the astrophotography software I use here.

Cropping/Rotating the file in Photoshop

The first thing I like to do is to rotate and crop the image.  A temporary levels adjustment was made to get a better look at the edges of the frame.  As you can see, my frames rotated and shifted slightly between the imaging sessions.  This creates an unusable sky at the edges of the image, so I will crop the image to about 85%.  In the future, I plan to incorporate a plate-solving software such as AstroTortilla to help line up my images over multiple nights.

deep sky image processing

Rotating and cropping the image in Photoshop

To save some of the outer regions around the nebula, I will have to repair some of the outer background sky using the healing brush, and the content-aware fill tool in Photoshop.  Ideally, you would want to keep as much of your original frame as possible.  Once I have cropped the image, I will adjust the black point of the image.

Levels Adjustment / Setting Blackpoint

As you can see in the image below, the histogram shows that the majority of the image data is contained in the mid-level tones.  I will move the slider to the left of the histogram over until it touches the information contained within the image.  This will darken the background sky and increase the contrast of the original image.

Levels adjustment

the first levels adjustment creates much more contrast in the image

The slider to the right of the data was moved inwards as well.  It’s important that you do not clip the data and lose any pixel information.  You may notice that the core of the Orion Nebula is completely white and “blown out”, I will correct this issue later on.




Before setting the initial black-point, I will give the image a semi-aggressive curve stretch to reveal more of the outer nebulosity.  This will also discern where the nebula ends and the background sky begins. Before Photoshop will let us make this adjustment, we will need to convert the image from a 32-bit file to a 16-bit file.

Image > Mode > 16 Bits/Channel

An HDR Toning window will open up.  Avoid choosing the tempting default preset of Local Adaptation, and instead, select Exposure and Gamma from the Method selection area. Leave the default exposure and gamma settings.  As this tutorial moves on, we will be creating our own HDR (High Dynamic Range) version of the Orion Nebula using very specific actions and settings.

At this point, you can adjust the levels once more, as there is likely empty space to the left of the data in the histogram again.  You may also choose to create a copy of your original layer, or create a new adjustment layer to work from.  Having snapshots of your image at each stage of the processing workflow will help you go back and fine-tune your edits.  Personally, I like to use a mixture of new layer copies using the History feature of Adobe Photoshop.

Here is what my initial curve stretch looks like:

curves adjustment in Photoshop

 

The curves stretch I applied brought forward the fainter details of the outer nebulosity.

Here is a little trick I like to use: With the curves window open, hold down CTRL, and click an area of the nebula you want to bring forward.  This will plot a point on the histogram you can pull from to stretch that particular tonal range.  You can also plot an additional point of a neutral area of background sky, and know that you are pulling data forward from only the nebula itself, and not the space around it.

Levels Once the curve stretch has been applied there are two ways to set the black point of the image.  The Set Gray Point eyedropper in the levels window is great for a quick overall adjustment.  Although some astrophotographers will argue that this method results in a loss of overall range of data.   You can also manually set the color of your background sky by plotting a Color Sampler eye dropper in a neutral area of space.

Using the Info window, adjust the left-hand slider on each RGB level until the values are balanced.  A background sky with Red/Green/Blue values of about 30/30/30 is a good starting point.

Creating a star mask

If you don’t want to risk the chance of brightening the stars in your image and blowing them out, try using a layer mask to protect them from growing in size and intensity.  The art of stretching the deep-sky object, but not the stars is a constant challenge when processing astrophotography images.

You can create this mask by using the Color Range tool. Select > Color Range.

Select Color RangeThen, use the eyedropper to select a medium-sized star within the frame.  Adjusting the Fuzziness slider will affect how much of the color range (and stars) will be selected.

You will have to experiment with the fuzziness slider to select your intended amount of stars.  In my example, I used a value of 140.  After the stars have been selected, I suggest softening the selection for a more natural blend in the mask.  To do this:

Select > Modify > Expand (2 Pixels)

Select > Modify > Feather (3 Pixels)

Again, these values will vary based on your image scale. If you are shooting wide field through a Canon T3i or similar model, these settings should work well.

Like many tasks in Photoshop, there are numerous ways to accomplish a layer mask adjustment.  For this step, I prefer to invert the selection of stars (Select > Inverse) and make my curve adjustment to all areas of the image except the star mask I created.

Here is what my image of the Orion Nebula looks like at this stage:

Image Processing - Orion Nebula

I cropped the image in a little more and used Gradient Xterminator around the edges of the DSO to balance the background sky.  Again, the core is still blown out at this stage.  I will add 2 additional stacks of 15 and 30-second images of the bright core to reveal the full range of detail in the Orion Nebula.

Astronomy Tools Action Set

At this stage of my image processing workflow, I will use my first action from Noel Carboni’s action set.   The action is called Local Contrast Enhancement.

This action does a great job at sharpening details and increasing the contrast of the deep-sky object.  It is wise to create a new layer with this action applied, so you can toggle the effect on and off.  For my image, I am going to apply a layer with this action at 75% opacity.  I have also created a mask on this layer so that it does not affect the areas of space where I do not want to increase the contrast.

Directly after this action, I prefer to run Enhance DSO and Reduce Stars.  This action can takes up to a minute or more to complete, depending on your image and the computer you are using.  Again, a new layer using this action is recommended, as this action can dramatically change the look of your image.

Here is a before/after look at my image after running Local Contrast Enhancement and Enhance DSO and Reduce Stars:

Photoshop actions before - after

Before and After applying actions in Photoshop

To make a new adjustment layer with all previous actions and adjustments made, use the keyboard shortcut: CTRL + ALT + SHIFT + N + E. This is a very helpful technique to use as your continue to add adjustment layers to your image.

Applying the “Tamed Core” Layer

At this stage, I will apply a pre-processed stack of shorter exposures to the image.  To capture these images I shot a series of 15-second and 30-second exposures with the goal of collecting detail in the brightest areas of the Orion Nebula.  A good indicator of this dynamic range in values is the ability to discern the individual stars in the Trapezium.

The short exposures were stacked in DeepSkyStacker using dark, bias and flat frames just as the primary image was.

Orion Nebula Core ExposureThese layers were processed in the exact same fashion as the primary image.  This means that similar adjustments were made to the levels, curves, and actions – but in an isolated area.

Blending the two images will be a lot easier if they have been pre-processed in the same manner.  Some may argue that combining the core should have taken place much earlier in the process.  However, this timing of this workflow works best for my personal taste.  With so many opinions about how to properly process a deep sky image, I prefer to lean towards the workflow that I enjoy most.  This way, I can enjoy the hobby for years to come.

Here’s where it gets fun

Select the image of the detailed core, and paste it onto your original image as a new layer.  Rather than using a traditional mask method, I like to use a feathered eraser brush at an opacity of 15%.  This allows me to subtly remove the unwanted data on the top layer (the core), one brush stroke at a time.




When I need to see the faint details of the edges of the core layer, I simply create a 100% white layer and place it as the layer below.  The amount of brightness of the core is a matter of taste. This aspect of the image has varying points of view as to how an HDR Orion Nebula is “supposed to look”.

I personally think that the Orion Nebula should have a bright core!  With the right amount of blending it is possible to show the full range of detail and keep the core as the brightest area of the image.  Flattening core to a lower brightness than the outer nebulosity can give the nebula a plastic look.

Blending the core

Layering in the core can take a long time if you are particular about the overall look of your image.  I used several copies of both stacks of shorter exposures to gradually work the new core into my existing image.

Final Processing Steps

With the full dynamic range captured in the image (depending on who you ask), I can now go ahead and make my final image processing steps to further increase the color and detail of the image.

Color sampler toolAt this point, I like to double check the levels of color in the background sky.  Using the Color Sampler Tool in 2 areas of the background sky indicates that the image is rather well balanced at the moment.

Increase Vibrance and Saturation

To increase the saturation of the Nebula without bringing noise and unwanted color from the background sky, I’ll use the Select Color Range tool again.  This time, use the eyedropper to select the color from the nebula you wish to intensify.  I choose the mid-level pink areas of Orion.

You may also want to run some actions on your image such as Increase Star Color, and Make Stars Smaller. As always, apply these actions to a new layer so that you can control the amount of the adjustment using the opacity slider.  I will often use both of these actions, in small amounts.

Adding a layer of H-Alpha

This is where the image really starts to “pop”.  I shot over 2 hours worth of data through a 12nm Astronomik clip filter with my Canon T3i.  I will combine this data with the RGB image we just processed using the HaRGB processing technique outlined in this tutorial:

Deep Sky Image Processing in HaRGB – Tutorial

Orion nebula in Ha

The Orion Nebula in Ha

The image above is 32 X 5-minute subs @ ISO 1600

If you are interested learning how to shoot H-Alpha with your DSLR camera, read my post on how a DSLR Ha Filter can improve your astrophotography.

Without explaining every detail in the HaRGB tutorial I linked above, the premise is basically to add the Ha as a luminosity layer at about 75% over your original color image.

Because the data in the core of the H-Alpha version of Orion was blown out, it is important to note that I removed this area of the Ha luminosity layer, so that I did not lose any detail in the final composite image.  By turning the Ha layer off and on, you can determine which areas of the nebula are being improved, and which areas are losing detail and/or color.  I prefer to create another layer mask using the Ha layer, leaving only the key improvement areas at the full 75% opacity.

Below is my final image of the Orion Nebula using the processing methods outlined above:

Orion Nebula - AstroBackyard

Final Image Details:

Hardware:

Mount: Sky-Watcher HEQ5 Pro Synscan
Telescope: Explore Scientific ED102 CF
Imaging Camera: Canon T3i (600D) Modified
Filters: Hutech IDAS LPS, Astronomik 12nm Ha
Flattener/Reducer: William Optics FF III
Guide Scope: Orion Mini 50mm, Starwave 50mm
Guide Camera: Meade DSI, Altair Astro GPCAM2 AR0130

Software:

Image Aquisition: BackyardEOS
Autoguiding: PHD2 Guiding
Registering/Stacking: DeepSkyStacker
Image Processing: Adobe Photoshop CC

Exposure Details:

RGB: 3 Hours, 8 Minutes (55 frames)
Ha: 2 Hours, 40 Minutes (32 frames)
Total Integrated Exposure: 5 Hours, 48 Minutes

Photoshop TutorialI am always looking to improve my deep sky image processing techniques.  For a video presentation of these techniques in action, please visit the AstroBackyard YouTube Channel.  If you like to see more of my deep sky astrophotography images, please have a look at the Photo Gallery.

This winter was a memorable one for me.  By sharing my experiences in the backyard on this blog and on YouTube, I was able to connect with fellow backyard astronomers on a deeper level.  There may not have been many clear nights, but the ones that were felt extra special.  Until next time, clear skies!

Deep Sky Image Processing Help:

Settings for DeepSkyStacker

Video Tutorial: Deep Sky Image Processing in Photoshop

Astrophotography Image Processing Video (YouTube)

 

 

Related Tags

Manual Stacking in Photoshop for Reduced Noise

|Tutorials|4 Comments

Even without a tracking mount, your astrophotography images can benefit from manual stacking in Photoshop. This method involves aligning exposures and combining them into a master composite.  It can make a big impact on your astrophotos by reducing noise and improving the signal-to-noise ratio.

The images used in the example were captured using a camera lens in place of a telescope. A wide-angle lens can be a tremendous way to photograph a large area of the night sky at once, rather than zooming in on a deep sky object.

manual stacking in Photoshop

This night sky photo contains 11 exposures stacked manually in Photoshop

This technique requires no darks frames, no stacking software, and no tracking mount.

The 2 software applications used in this tutorial are Adobe Bridge and Adobe Photoshop. (Lightroom will also work) The main idea is to reduce noise by layering single exposures gradually, effectively canceling out much of the noise from a single frame.  The key is to gradually reduce the opacity of each layer until you reach the top of the image set.

Basic Astrophotography Camera Settings and Tips

Manual stacking is easy with Adobe Photoshop

Let’s say you have taken a 30-second exposure of the starry night sky with your camera on a tripod.  By taking 10-12 shots of the same scene, your final composite version will have much less noise, a smoother background, and more detail.

Watch the video tutorial below:


In my example, the camera happened to be sitting on top of a telescope on a German equatorial mount.  Don’t confuse this fact into thinking this method does not work without a tracking mount.  30-second exposures at wide focal lengths do not contain much star-trailing and are perfect candidates for this treatment.

The reason this technique is referred to as “manual stacking” is that the user layers each exposure manually, rather than loading the frames into Deep Sky Stacker.  The simplicity and effectiveness of this technique is something all astrophotography enthusiasts can benefit from.

Manual Stacking Example

If you are interested in the full explanation of this process, follow the link below.

Astrophotography Tutorial – Align and stack exposures in Photoshop

Capture longer exposures

A small tracking camera mount such as the iOptron SkyTracker Pro allows you to shoot longer exposures than you can on a stationary tripod. The SkyTracker compensates for the rotation of the Earth, meaning that objects in space stay put as your camera collects light.
 
This is a useful piece of astrophotography gear if you plan on photographing the Milky Way. The following images uses 60 x 2-minute exposures at ISO 1600. The individual images were stacked using DeepSkyStacker, but similar results could be produced using the manual stacking process.
 
The Milky Way

What’s New?

My deep-sky imaging has been put on hold as it has been cloudy for nearly 3 weeks straight. This weather is sure to get any backyard astrophotographer down. However, my passion for imaging couldn’t be stronger as I have recently added some new astrophotography equipment to my arsenal.  This time, it was my primary imaging camera that needed replacement.

New Camera: Canon Rebel T3i (600D)

After much consideration, I have decided to finally upgrade my primary imaging camera to a Canon Rebel T3i.  If you follow AstroBackyard on Facebook, you may have heard that this DSLR was professionally modified for astrophotography by removing the IR cut filter.  The Canon 600D features an 18MP CMOS sensor capable of producing images 5184 x 3456 pixels in size.  This is a huge leap from the 12MP 450D sensor I was using.

Modified Canon T3i for astrophotographyPowering the camera for a full night of imaging requires more than a single battery charge.  I purchased an AC adapter to plug the camera into an electrical outlet for unlimited power.  I never had this function with my previous imaging camera, and fumbled around with batteries for far too long.  I also sprung for a remote shutter release cable with a built-in intervalometer.  The great news is that this model is compatible with my Canon Xsi as well.

The AC Adapter for the Canon T3i is available on Amazon

Canon t3i AC adapter

 

I am really excited to start using this camera for deep-sky imaging.  Although tests show that the thermal noise level is comparable with the 450D, the Canon T3i’s increased resolution means larger astrophotos with enhanced detail. Also, the flip-out LCD screen will come in handy when focusing using live view.  This camera was modified by Astro Mod Canada.

Drone Footage?

Did you notice the aerial footage in the latest AstroBackyard video on YouTube?  That video was shot at a friend’s house using a DJI Phantom 4 in his backyard. I absolutely love drone footage and hope to get one of my own in the near future.  There are infrequent situations where that type of footage makes sense for an astrophotography video, but I’ll find a way to work them in!

I’ve also been working away trying to improve the resources on this website.  The photo gallery, equipment, and tutorial sections have seen a number of changes this month.  Thank you to everyone who has taken the time to comment on my blog posts, I really appreciate the kind words.  For the latest information and photos, please follow AstroBackyard on Facebook.

Latest Image Re-Process: The Eagle Nebula

DSLR astrophotography

Related Tags

Forgotten Light Frames

|Nebulae|0 Comments

While digging though some old folders on Adobe Bridge, I stumbled across some unprocessed, 300 second light frames of the Flaming Star Nebula from November 2013!  When you are desperate to get out and image a new target, this is like hitting gold.  

I was originally looking for my raw files of the Pacman Nebula, which I feel is in desperate need a new process. (Those stars look pretty rough)  I found a folder labelled “Flaming Star – 5 Min Lights”.  I never processed this image!  The Flaming Star Nebula is a colorful collection of glowing gas and dust lit up by the bright star AE Aurigae. 

The tough part about this process will be the limited exposure time.  1 hour of data is really not ideal for a quality astrophotography image.  I find that out the hard way below:

IC 405 – The Flaming Star Nebula

IC 405 - Flaming Star Nebula

Photo Details

Photographed on: November 29, 2013

Telescope: Explore Scientific ED80 with WO Flat III 0.8x FR/FF
Guiding: Meade DSI Pro II and PHD Guiding
Guide Scope: Orion Mini 50mm
Camera: Canon EOS 450D (Stock)
ISO: 1600
Exposure: 1 hour (12 x 300s)
Processing Software: Deep Sky Stacker, Photoshop CC
Support Files: 15 darks

Guided with PHD Guiding
Stacked in Deep Sky Stacker
Processed in Adobe Photoshop CC

This image was acquired using Canon EOS Utilities, and not BackyardEOS as I use now.  This was photo was also shot before I modified my Canon 450D for astrophotography.

Now you might be thinking “how could you spend hours imaging a nebula and forget to process it?”  It’s simple – life is busy!  I likely had a busy week following the the imaging session, and began I new session before I even looked at the precious data collected on that cold November night.  I don’t see any dark frames to support the image.  

This may have been another reason I held off.  I bet that I wanted to take 5-minute darks of the same temperature before stacking, but never got around to it.  This could be a problem.

But first, let’s get this cleared up

This is budget Astrophotography.  Most of my gear was purchased used from online forums and astronomy classifieds.  The total value of the equipment used to photograph this nebula was purchased for under $3,000.  It’s not top-of-the line gear by any stretch of the imagination.  My astrophotography image processing skills were self-taught.  I am no scientist, that’s for sure. Just like you, I have a strong desire to capture beautiful images of the night sky.  I always appreciate constructive criticism, and enjoy helping others learn through my mistakes.

Stacking without Dark Frames?

First of all, I’ll have to use dark frames from a different night to stack with the Flaming Star light frames.  This means that it is very important to match the temperature of my light frames from that night of imaging.

I have done a poor job of creating a master dark library, so finding matching dark’s may be tough.  I usually try to record the temperature of my dark frames in the file folder, for this very situation.  There are external software applications available that can help create a dark frame library, such as Dark Library.  

I remember using this years ago, but their website appears to be down right now.  I will use the 5 minute dark frames from my Pacman Nebula image taken earlier that month, labelled 4 degrees.

Another option is to just stack the light frames without any dark’s.  I’ll try both and compare the two.

Here is the version stacked with no dark frames:

Deep Sky Stacker with No Darks

Here is the version using dark frames from a previous night:

Deep Sky Stacker with Darks

As you can see, stacking with the dark frames produced a better result.  Even though the temperature of dark frames did not match perfectly, the dark frames removed some of the dead pixels and noise from the image.  Notice the red streak of dead pixels on the “no-darks” version.  All of these imperfections would become intensified after processing!  

I performed a few basic edits to the examples above to have a better look at the differences. (Levels, Gradient Xterminator, and Curves)  Now that we have registered and stacked our 1 hour’s worth of data, let’s start stretching the data in Photoshop.

How to take proper Dark Frames for Deep Sky Stacker

The answer to this and more in the FAQ section

Processing the Image in Photoshop

If you have followed any of my astrophotography tutorials on my website, or video tutorials on YouTube, you already know the basics of my processing workflow.  This process has evolved over the years as I learn new tricks.  However, processing the Flaming Star Nebula was particularly tough because of the limited exposure time on the subject.  

Add in the fact that this nebula is quite faint, with many bright stars surrounding it, and you’ve got an astrophotography challenge for even the most experienced astrophotographer.

 

Quick Astrophotography Tip

Try to frame your deep-sky object in an interesting way.  Include nearby star clusters, nebulae or galaxies.  For inspiration, search for your target on APOD, and see how the professionals have framed the object.  This may spark your creativity to photograph an existing target in a different way.

HLVG – Green Noise Remover

The entire image had a noticeable green cast over it, perhaps because of the extreme amount of noise, or the miss-matched dark frames.  I ran Deep Sky Colors HLVG on medium, which helped a lot.  HLVG was created by Rogelio Bernal Andreo of RBA Premium Astrophotography. 

It is a chromatic noise reduction tool that attempts to remove green noise and the green casts this noise may cause in your astrophotography image.  It is based on PixInsight’s SCNR Average Neutral algorithm.  If you don’t already have this useful filter for Photoshop, I highly recommend it, it’s free!  You can download the plugin here:

Hasta La Vista, Green!

HLVG Filter for astrophotography

Results and Thoughts

I must admit, this post became a bit of a nightmare.  I began to document my processing steps one by one, taking screenshots of progress along the way.  I wanted to provide a detailed tutorial of how I turned this forgotten data into a masterpiece, despite having no associated dark frames, and only an hour’s worth of exposure time.  As I experimented using different methods of noise reduction, and various orders of operations, I became very discouraged with my final image results.  

I spent hours taking different roads with all of my trusted astrophotography tools at my disposal, and the results continued to be unimpressive.  By adjusting the curves enough to show any substantial detail on the nebula, I introduced a frightening amount of noise into the background space.  No amount of noise reduction could remove it, without turning the entire image into a blurry mess.

I just couldn’t bring myself to post a tutorial with the end result turning out like it did.  So I scrapped the idea, and settled for a forgettable image of the Flaming Star Nebula.  Surely this gorgeous nebula that spans 5 light years across deserves more than that.

Astrophotography Processing Tutorial

My unused processing tutorial screenshots

At the end of the day:

No amount of processing can make up for lack of exposure time!

I guess you could say I was doomed from the start.  I am not going to spend any more time on this image until I am able to capture at least another 2 hours of data on it.  I hope you can learn from my experiences in astrophotography, in both victories and failures.  But I guess that’s why you’re here 🙂  Please follow AstroBackyard on Facebook for the latest updates.

Related Tags