In my last (and first) article here on MyInkBlog, I dove into some of the key differences that exist between Clipping Masks and Layer Masks in Photoshop. We touched on several different areas, but near the beginning of the article I wrote:
we can start by using any one of a number of different extraction methods (in this case, I used channels, which is another article all on its own).
This is that article. Actually, it’s the first of two articles. It seems that there was quite a bit of interest in knowing more about channels and how I used them to extract the woman from the background in that previous article. Well, I’m more than happy to comply with that request. In this first article, we are going to cover some of the basic concepts behind channels. The follow-up article will be more of a tutorial-like piece, in which I will work through a channels-based extraction.
So let’s get started. For the purposes of this discussion, we’re going be using this simple image of an apple, which can be downloaded for free over at Dreamstime.
Now, we’ll start with some of the true fundamentals. First, you can access your channels by opening up the Channels Palette (Window » Channels from in the menu). This is where we will be doing most of our work, both in this article and in the follow up tutorial.
In this case, we can see four channels – RGB, Red, Green and Blue. By the way they are laid out in the palette, they actually look a lot like layers. Don’t be confused, though. That is strictly a matter of interface. Channels and Layers are entirely different entities.
You may also notice that the channels seem to reflect the current colour mode of the document – which is RGB. This is indeed true, and leads us to another important point. The Channels Palette will change based on the colour mode of the document. This is because each colour mode uses a different method for defining colour, and at their most basic level channels are simply a programatic way of displaying the interaction of basic primary colours, which combine to create a much larger range of full colour.
The Big Two: RGB and CMYK
I’m sure that many of you are familiar with basic concepts of how colours are mixed, but since this is one of the fundamental building blocks of how channels work, let’s quickly review our two basic colour structures.
First, we have RGB, which is the colour mode of all things like your monitor, television and various handheld devises. RGB makes use of concepts of additive colour, combining tiny bits of red, green and blue lights in varying degrees in order to create a full and rich colour gamut.
On the other hand, we have CMYK, which is what is used for print. With this kind of colour, a printer or press lays down tiny dots of cyan, magenta, yellow and black inks, which blend together to create the illusion of colour. While pixels on a monitor actually emit coloured light, the tiny dots on a page absorb coloured lights, reflecting only the colour that you see. As more ink is mixed together, more light is absorbed. That’s why it is called subtractive colour – because less and less gets reflected as you add more ink.
I should also note that when you’re working in CMYK your screen is still displaying in RGB – Photoshop is simply approximating CMYK colour for display. If you want to learn a bit more of the science, read about additive colour and subtractive colour on Wikipedia.
For our purposes, the important thing to know is that each of these colour models require different sets of colour information, which are reflected in their types of channels. Let’s examine this a bit more closely.
First, open up the image and then pop open the Channels palette. When downloaded from the internet, the image should be in RGB mode by default, so you should see four different channels. Actually, there are three channels (Red, Green and Blue), and then an automatic, and uneditable channel (called RGB). Each channel visualizes the three different streams of colours that compose the document, which we can see by taking another look at our previous screenshot:
The Photoshop default is for each channel is actually represented by a simple greyscale image (though we’ll look at how to change that). For example, here is the red channel for our apple:
It may seem odd that the apple itself actually seems washed out, rather than strong and vibrant, which you might expect given that the fruit itself is also red. There is a good reason for this. Within an RGB channel, any white pixel will be at its maximum brightness for that colour – red in this case. In other words, it will emit the maximum amount of red light. Conversely, wherever it is black, the pixel will not emit any light of that colour. So, because the apple is red, pixels that make up that part of the image are actually quite light.
Conversely, if we convert the apple image to CMYK, we will see something quite different. We now have five channels, one for cyan, magenta, yellow and black, as well as the automatic combined channel.
In this case the channels work somewhat differently, because we are working in a colour space that is intended to mimic the workings of ink on paper. As such, you will find that wherever a pixel is white, it is meant to represent blank, unprinted paper. Wherever a pixel is black, it is meant to represent an area where the paper is covered with the maximum saturation for that particular colour.
Let’s look at the magenta channel as an example. To create red with ink, we can actually combine magenta with yellow. So, if we printed our apple image, we would expect there to be a lot of magenta ink used. This is accurately reflected in the magenta channel.
Notice that the areas of the deepest red are the most saturated with black, whereas areas that contain relatively little red (such as the reflections and the yellow area near the top of the apple) remain much whiter.
That’s really the basics of what channels are. They are simply a way of splitting apart the different colour information in a document, giving you the ability to make tweaks and adjustments on a colour by colour basis. A quick word of warning, though: using channels to make extensive colour modifications on a heavily layered document can be a bit tricky. Making a change in a channel will still only effect the current layer, so I would recommend doing some experimenting.
In fact, I always recommend experimenting! I’ve found that it’s the absolute best way to learn.
Seeing Channels in Colour
Personally, I prefer working with the default greyscale rendering of the channels. However, if you’re finding it difficult to wrap your head around how the colours are actually working, Photoshop has a little interface option to help make it easier for you. Select Photoshop » Preferences » Interface from the menu (PC: Edit » Preferences » Interface). In the dialogue box you should see a little check box with the option â€œShow Channels In Colourâ€. By default this is turned off, but try turning it on. In RGB mode, the channels palette and red channel should look something like this:
In CMYK mode, the channels palette and magenta channel should look something like this:
This little trick can be super helpful when it comes to visualizing the colour behind unique channels. However, when it comes to using them for extractions, it can make things a little bit more difficult. So, in the next article, I will have this option turned off again
The Other Colour Modes
While RGB and CMYK are the two most common colour modes in my experience, they are certainly not the only ones, and the other colour modes also change the way the channels work. The following is a brief summary of how some of the other colour modes affect channels.
This colour space is significantly different from either RGB or CMYK because it is not device dependent. In other words, it is not designed to mix colours, either of ink or light. Instead, it is designed more to emulate human perception of colour, by actually separating the lightness values from the colour values.
There are basically three channels. The first is the Lightness channel, which controls the relative lightness or darkness of a particular pixel. The other two channels (called a and b) contain actual colour information, by determining the relative saturation.
By far the most complex of all the colour modes (at least in my opinion), trying to explain the entire Lab colour model would be an entire article onto itself! It might even take several articles. For the most part, though, it appears to be used primarily for post-production work in photography, so it could be helpful in terms of establishing a proper colour balance. In terms of actual extraction, though, it’s probably not going to be all that useful.
As the name would suggest, this one definitely makes use of channels. Basically, though, it amounts to an unmanaged CMYK. Sort of. If you convert to multi-channel right out of RGB, the channels will be converted into CMY, with no black (K). However, if you convert out of CMYK, you will keep all four channels.
The primary difference, though, is that you can actually change the colour of a channel. So, instead of having a cyan channel, you could have a channel dedicated to a unique Pantone colour. This can make for some interesting effects and can be beneficial when preparing a Photoshop design for printing with alternate colours.
From an extraction perspective, it doesn’t really offer anything that is not already available in CMYK mode.
An index image is basically just that, an index of colours which are defined across a grid, which represents the canvas itself. Each pixel is essentially given a simple and static colour value from the index of 256 colours. This is the kind of technology that is used in image formats such as GIF.
It is useful for maintaining small file sizes, but makes no real use of channels. If you look in the Channels Palette, all you will see is the actual index itself. This means that indexed mode is completely useless when it comes to channel based extractions.
This colour mode is exactly what it sounds like, and really isn’t a colour mode at all, since it contains no colour. In this mode, the image is basically reduced to values of lighness and darkness, as represented by white, black and a full range of greys.
It can be useful in some circumstances, such as preparing a design for simple one colour print, but is not really all that useful for extraction purposes.
This is the most basic of all colour modes, and is comprised entirely of black or white pixels. There is not even any shades of grey. Again, it has only uses a single black and white channel and has no real use for running extractions.
Duotone (and the related tritone and quadtone) is kind of like a fake colour mode, insofar as the main image remains primary a greyscale, with highlights, midtones and shadows. You set a colour for each â€œtoneâ€ you choose and then adjust that colour’s levels to dictate how you want it to effect the various tones in your image.
It’s a very different way of working with colour, and is primarily used for adding interesting colour treatments to a black and white photograph – such as creating a customized sepia effect. Since it only has a single greyscale channel, it really isn’t much good when it comes to extractions.
Well that’s a basic run down of channels. There’s a lot more we could talk about, but that should cover most of the basics that you will need to know for creating extractions. We’ll get into the nitty gritty in the Part 2 tutorial!
For those of you who just can’t wait though, here’s a little preview. To create a selection from a channel, all you have to do is Command-click (PC: Ctrl-click) on the preview for any given channel. A selection will be made based on the black, white and grey pixels for that channel (with white being fully selected and black being unselected).
Pretty simple right? Well yes – and no. It would be incredibly rare for a default colour channel to be ready for a perfect extraction. It usually takes a bit of tweaking and modifying, and that’s exactly what we’ll get into in Part 2. Be sure to subscribe so that you don’t miss it!