Tutorial (PixInsight):
Narrowband Bicolour Palette Combinations
Narrowband imaging carries with it several advantages over LRGB or One Shot Colour imaging. Firstly, narrowband filters select out a very distinct part of the spectrum, which corresponds to a particular emission line. This allows astrophotographers to image right through a lot of light pollution (not necessarily white LED broad spectrum light pollution, however). Secondly, stars appear significantly less bloated and nebulosity appears richly detailed. These narrowband images can be used to either enhance LRGB or One Shot Colour images, or to produce images in their own right.
As most of the visible Universe is Hydrogen, the emission line for Hydrogen-Alpha is particularly prominent. Astrophotographers producing monochrome narrowband images always choose Hydrogen-Alpha as a result. In the most part however, those pursuing narrowband imaging tend to capture images through multiple narrowband filters. These include Hydrogen-Alpha, Hydrogen-Beta, Nitrogen-II, Sulphur-II and Oxygen-III. The most popular of these are Hydrogen-Alpha, Sulphur-II and Oxygen-III, as these can be used to produce colour images of the likes produced by the Hubble Space Telescope, using the popular Hubble Palette. This tutorial however, is dedicated to those wishing to produce colour images of the Bicolour Palette - colour images involving only two narrowband filters. These are normally chosen as Hydrogen-Alpha and Oxygen-III, though the tutorial can be used for any combination of two filters. Moreover, section 4 can be used to produce colour images out of any number of narrowband filters as we blend different filters for different colour channels.
Assumed for this tutorial:
Please feel free to ask questions or leave comments via the comments section on the bottom of this page.
As most of the visible Universe is Hydrogen, the emission line for Hydrogen-Alpha is particularly prominent. Astrophotographers producing monochrome narrowband images always choose Hydrogen-Alpha as a result. In the most part however, those pursuing narrowband imaging tend to capture images through multiple narrowband filters. These include Hydrogen-Alpha, Hydrogen-Beta, Nitrogen-II, Sulphur-II and Oxygen-III. The most popular of these are Hydrogen-Alpha, Sulphur-II and Oxygen-III, as these can be used to produce colour images of the likes produced by the Hubble Space Telescope, using the popular Hubble Palette. This tutorial however, is dedicated to those wishing to produce colour images of the Bicolour Palette - colour images involving only two narrowband filters. These are normally chosen as Hydrogen-Alpha and Oxygen-III, though the tutorial can be used for any combination of two filters. Moreover, section 4 can be used to produce colour images out of any number of narrowband filters as we blend different filters for different colour channels.
Assumed for this tutorial:
- Knowledge of operating PixInsight, related to dealing with images and processes (read this, sections 3 and 4).
- Your images have already been pre-processed fully (read this).
- Your images have all been registered with each other, have had the black edges resulting from pre-processing removed using DynamicCrop and any background gradients present removed using DynamicBackgroundExtraction (read this, sections 1, 2 and 3).
Please feel free to ask questions or leave comments via the comments section on the bottom of this page.
1. Initial Preparations for the Narrowband Images
It is typical for astrophotographers to want to colour-combine images during their linear state, as this is common practice for LRGB images. Large differences between image brightness as captured through different emission line filters however, make it difficult for linear images to blend into different colour palettes. Typically, the images appear washed in shades of whatever colour channel is predominantly occupied by Hydrogen-Alpha. In order for different filter images to have a more or less equal contribution to the final colour blend, one can stretch these images to non-linear before blending together.
Though your images may differ, it is common to apply some noise reduction to images in their linear state. The choice of whether or not to do this, or how aggressively to do it, depends on the level of noise in your images. Please note that this is beyond the scope of this tutorial and is covered amply by another tutorial specially written on the subject of noise reduction. My personal noise reduction routine of choice for the images we are about to post-process was using MultiscaleLinearTransform (as described in the tutorial on noise reduction) with stretched clone copies of the images themselves acting as masks.
Though your images may differ, it is common to apply some noise reduction to images in their linear state. The choice of whether or not to do this, or how aggressively to do it, depends on the level of noise in your images. Please note that this is beyond the scope of this tutorial and is covered amply by another tutorial specially written on the subject of noise reduction. My personal noise reduction routine of choice for the images we are about to post-process was using MultiscaleLinearTransform (as described in the tutorial on noise reduction) with stretched clone copies of the images themselves acting as masks.
The above two autostretched images correspond to Hydrogen-Alpha (left) and Oxygen-III (right). They have been fully pre-processed, have been registered with each other, have had background gradients removed with DynamicBackgroundExtraction and have been noise reduced with MultiscaleLinearTransform. As aforementioned, in order to get the most out of colour combining these images, it is best they are stretched to non-linear. Since this is the subject of another tutorial, the procedure covering stretching the images to non-linear will not be covered in detail. I generally use either HistogramTransformation or MaskedStretch, or a combination of both (first MaskedStretch and then HistogramTransformation to tweak).
An advantage of using MaskedStretch over HistogramTransformation is that it can more accurately match the background brightness for both narrowband images. This is the case because MaskedStretch allows the user to set a specific background brightness and apply the same to both narrowband images. However, due to weaker emissions in Oxygen-III, the overall nebulosity will be brighter in the Hydrogen-Alpha image. One can then exaggerate the brightness of the Oxygen-III image nebulosity by stretching it further, perhaps with HistogramTransformation or even CurvesTransformation in RGB/K mode.
Below shows the result of stretching both images to non-linear using MaskedStretch with Target background set to the default 0.12500000, Iterations set to the default 100, Clipping fraction set to 0.00000000 and a Background reference selected referring to a small preview box over simply background (no nebulosity), on each image.
An advantage of using MaskedStretch over HistogramTransformation is that it can more accurately match the background brightness for both narrowband images. This is the case because MaskedStretch allows the user to set a specific background brightness and apply the same to both narrowband images. However, due to weaker emissions in Oxygen-III, the overall nebulosity will be brighter in the Hydrogen-Alpha image. One can then exaggerate the brightness of the Oxygen-III image nebulosity by stretching it further, perhaps with HistogramTransformation or even CurvesTransformation in RGB/K mode.
Below shows the result of stretching both images to non-linear using MaskedStretch with Target background set to the default 0.12500000, Iterations set to the default 100, Clipping fraction set to 0.00000000 and a Background reference selected referring to a small preview box over simply background (no nebulosity), on each image.
Having used the same settings in MaskedStretch (besides the Background reference image referring to each image's own background preview box), the background has been raised to the same brightness in both images. The trouble is that naturally, the Oxygen-III image nebulosity is also less bright than Hydrogen-Alpha's and this means it will contribute less to the colour combination. We can increase the overall Oxygen-III image nebulosity brightness with HistogramTransformation, applying a stretch as well as a black-point adjustment, as shown below.
Indeed one can switch between viewing the Hydrogen-Alpha histogram and the Oxygen-III histogram, in order to see how the stretched Oxygen-III histogram will match the current Hydrogen-Alpha histogram. Having the Real-Time Preview window for the Oxygen-III image open next to the Hydrogen-Alpha image also allows the user to visually gauge the brightness between both, for both their background and their nebulosity. At this stage it also helps to have some idea of what colour you would like your individual narrowband images to contribute to the colour-combined image. After all, if you want Oxygen-III to represent Blue, then you would prefer if the Oxygen-III image is a little brighter than the Hydrogen-Alpha image (as Hydrogen-Alpha is generally so over-powering). The Statistics process can be used to see how the images compare, by comparing their Mean and Median pixel values while in 16-bit [0,65535] mode.
Some minor extra adjustments to the images can be made using CurvesTransformation in RGB/K mode. Producing a slight S-curve by creating a point a quarter of the way up the graph and dragging it down a bit, and creating a point three quarters of the way up the graph and dragging it up a bit, gives rise to extra contrast in bright areas and reduced background brightness.
Again, the user can visually compare the Real-Time Preview with the other image as adjustments are made. The Statistics process can also be used to compare the images after adjustments are applied. Either way, we achieve what we wanted - stretched non-linear narrowband images in Hydrogen-Alpha and Oxygen-III that will now be used for produce a number of different Bicolour Palette combinations.
Though not strictly the subject of this tutorial, one could do this comparative stretching of narrowband images for more than two narrowband images. This could be three or even more different narrowband images.
In order to colour-combine our narrowband images, we will use the PixelMath process, which allows all sorts of fine adjustments to combinations. This section ends with setting up PixelMath for what will be different ways to colour-combine the two narrowband images. We open the process and disable Use a single RGB/K expression, select Create new image and select RGB color under Color space. This will allow us to enter separate combinations of images for the Red, Green and Blue channels, and will create a new colour image when the process is applied.
In order to colour-combine our narrowband images, we will use the PixelMath process, which allows all sorts of fine adjustments to combinations. This section ends with setting up PixelMath for what will be different ways to colour-combine the two narrowband images. We open the process and disable Use a single RGB/K expression, select Create new image and select RGB color under Color space. This will allow us to enter separate combinations of images for the Red, Green and Blue channels, and will create a new colour image when the process is applied.
2. LRGB-Style Combination
One good way of combining Hydrogen-Alpha and Oxygen-III data is in a way that simulates the look of an LRGB image. This exploits the fact that the Hydrogen-Alpha emission line lies in the deep red part of the spectrum and the Oxygen-III emission line lies between the green and blue parts of the spectrum. As a result, the Hydrogen-Alpha image (called HA) is assigned to the Red channel and the Oxygen-III image (called OIII) is assigned to both, the Green and the Blue channels, as such:
R/K: HA
G: OIII
B: OIII
We now click Apply on PixelMath and the new colour image is created:
To get a good idea on how the image actually looks, we need to neutralise the background. For this, we create a small preview box around an area that contains nothing but background - no nebulosity or stars. We then open the BackgroundNeutralization process and select the preview box from the list under Reference image. Since the colour image is non-linear, we also need to increase the Upper limit parameter. Generally 0.2000000 works well but this should basically be set to just above the maximum background pixel value. You can simply use the Statistics process to determine this Maximum within the background preview box, or simply use the Readout Preview hovering the mount around the background preview box and sampling pixels within. Once set, Apply the BackgroundNeutralization process to the colour image.
You may delete the background preview box through the Preview -> Delete All menu after applying BackgroundNeutralization. As is to be expected, the intensity of specific colours visible is dependent on the overall brightness of your narrowband images prior to colour-combination.
3. Synthetic Green Combination
In absence of a third narrowband image, one can also perform a colour-combination that follows Steve Cannistra's famous synthetic green combination. This is doable very easily in PixInsight using the PixelMath process by simply multiplying the Hydrogen-Alpha (called HA) and Oxygen-III (called OIII) images together for the Green channel. This is ideally multiplied by a constant such as 1.5 in order to expand its dynamic range, since the synthetic green image, by default, is quite dark. The Red channel is occupied solely by the Hydrogen-Alpha image and the Blue channel is occupied solely by the Oxygen-III image.
R/K: HA
G: (HA*OIII)*1.5
B: OIII
We now click Apply on PixelMath and the new colour image is created:
To get a good idea on how the image actually looks, we need to neutralise the background. For this, we create a small preview box around an area that contains nothing but background - no nebulosity or stars. We then open the BackgroundNeutralization process and select the preview box from the list under Reference image. Since the colour image is non-linear, we also need to increase the Upper limit parameter. Generally 0.2000000 works well but this should basically be set to just above the maximum background pixel value. You can simply use the Statistics process to determine this Maximum within the background preview box, or simply use the Readout Preview hovering the mount around the background preview box and sampling pixels within. Once set, Apply the BackgroundNeutralization process to the colour image.
You may delete the background preview box through the Preview -> Delete All menu after applying BackgroundNeutralization. As is to be expected, the intensity of specific colours visible is dependent on the overall brightness of your narrowband images prior to colour-combination.
4. Blended Channels Combination
If we throw away all the ideas of emission line wavelengths and number of images used to produce a Red, Green, Blue colour image, we can really get creative. In a blended channels mode, we simply add percentages of different images together to be combined in a single channel. For example, the following shows the effect of assigning Hydrogen-Alpha (called HA) to the Red channel and assigning Oxygen-III (called OIII) to the Blue channel, with the Green channel taking a blend of 40% Hydrogen-Alpha with 60% Oxygen-III.
R/K: HA
G: (0.4*HA)+(0.6*OIII)
B: OIII
We now click Apply on PixelMath and the new colour image is created:
To get a good idea on how the image actually looks, we need to neutralise the background. For this, we create a small preview box around an area that contains nothing but background - no nebulosity or stars. We then open the BackgroundNeutralization process and select the preview box from the list under Reference image. Since the colour image is non-linear, we also need to increase the Upper limit parameter. Generally 0.2000000 works well but this should basically be set to just above the maximum background pixel value. You can simply use the Statistics process to determine this Maximum within the background preview box, or simply use the Readout Preview hovering the mount around the background preview box and sampling pixels within. Once set, Apply the BackgroundNeutralization process to the colour image.
You may delete the background preview box through the Preview -> Delete All menu after applying BackgroundNeutralization. As is to be expected, the intensity of specific colours visible is dependent on the overall brightness of your narrowband images prior to colour-combination.
The fact that what separates you from a colour image is simply tweaking a couple of numbers, or adding some more images together, means you can really get creative with experimentation. You could try boosting Hydrogen-Alpha and reducing Oxygen-III, or vice versa. If we leave the Bicolour Palette behind, if you have Sulphur-II data, you could add this to Hydrogen-Alpha in the Red channel. If you have a fourth narrowband image (say, Hydrogen-Beta or Nitrogen-II), you could also add this into a specific colour channel. Below shows an example of blending colour channels with three filters (Sulphur-II, Hydrogen-Alpha and Oxygen-III):
The fact that what separates you from a colour image is simply tweaking a couple of numbers, or adding some more images together, means you can really get creative with experimentation. You could try boosting Hydrogen-Alpha and reducing Oxygen-III, or vice versa. If we leave the Bicolour Palette behind, if you have Sulphur-II data, you could add this to Hydrogen-Alpha in the Red channel. If you have a fourth narrowband image (say, Hydrogen-Beta or Nitrogen-II), you could also add this into a specific colour channel. Below shows an example of blending colour channels with three filters (Sulphur-II, Hydrogen-Alpha and Oxygen-III):
R/K: (0.5*SII)+(0.5*HA)
G: (0.2*HA)+(0.8*OIII)
B: OIII
One thing to keep in mind is that the total of the addition should not exceed 100% as otherwise you will be white-clipping data (over-saturating very bright areas). For example, if you are adding Hydrogen-Alpha to Oxygen-III for the Green channel, as shown above, 40% and 60% work well but refrain from the likes of 40% and 80% as the total there would be 120%. The equivalent of 40% and 80% is 34% and 66% as the 1:2 ratio is roughly maintained, while keeping the total at 100%.
5. Notes on Colour
Always keep in mind when producing colour images from pure narrowband data that there are no rules to stick to. The colour image you produce is your own artistic and scientific representation of the line emissions captured. This is why one can feel total freedom when experimenting with different percentage combinations of different narrowband data within colour channels, or with how different narrowband data is actually combined together (such as using the synthetic green method).
Another thing to keep in mind is that the idea of colour-calibration with pure narrowband images is meaningless. We apply the use of the BackgroundNeutralization process because we wish the background to have a neutral tone to it, as in any type of image, but the ColorCalibration process for example, is not used at any point. The basis for this is quite simply as above - pure narrowband images are false colour images by their very nature.
Remember that how much you stretch your individual narrowband images to non-linear, defines how much contribution that data will provide into the colour-combined image (also dependent on your percentage combination values, if applicable). If your Oxygen-III image is extremely bright and will populate the Blue channel, then you can expect the image to have a bright blue tone to it, within the areas that there is dominant Oxygen-III data present. As a result, how much one stretches images to non-linear in relation to other images, is tweaked in tandem with PixelMath expressions for colour-combining to produce a desirable result. Following colour-combination, further tweaks can be made to an image's colour palette and saturation, but touching up colour in images is the subject of another tutorial.
Finally, please note that once the narrowband images have been colour-combined and produce a desirable result, whether or not you choose to alter the colour palette and/or saturation afterwards, further post-processing may be necessary to finalise the image. For example, you may wish to stretch the colour image further using the likes of HistogramTransformation or CurvesTransformation. Other contrast-enhancing techniques can also be employed to further bring out and sharpen fine details. These techniques are all the subject of other tutorials.
Another thing to keep in mind is that the idea of colour-calibration with pure narrowband images is meaningless. We apply the use of the BackgroundNeutralization process because we wish the background to have a neutral tone to it, as in any type of image, but the ColorCalibration process for example, is not used at any point. The basis for this is quite simply as above - pure narrowband images are false colour images by their very nature.
Remember that how much you stretch your individual narrowband images to non-linear, defines how much contribution that data will provide into the colour-combined image (also dependent on your percentage combination values, if applicable). If your Oxygen-III image is extremely bright and will populate the Blue channel, then you can expect the image to have a bright blue tone to it, within the areas that there is dominant Oxygen-III data present. As a result, how much one stretches images to non-linear in relation to other images, is tweaked in tandem with PixelMath expressions for colour-combining to produce a desirable result. Following colour-combination, further tweaks can be made to an image's colour palette and saturation, but touching up colour in images is the subject of another tutorial.
Finally, please note that once the narrowband images have been colour-combined and produce a desirable result, whether or not you choose to alter the colour palette and/or saturation afterwards, further post-processing may be necessary to finalise the image. For example, you may wish to stretch the colour image further using the likes of HistogramTransformation or CurvesTransformation. Other contrast-enhancing techniques can also be employed to further bring out and sharpen fine details. These techniques are all the subject of other tutorials.
Comment Box is loading comments...