Hi there,
Thanks for checking out a 'Explained' blog! In these; my aim is to give you some details on what some common technical terms mean in digital video and film. I hope you find something useful here.
Please note; I do not consider myself an expert on the topics - if you believe I've made a mistake somewhere please do let us know! I'd love for this to be a learning experience for me as much as anyone else.
In this blog we're going to look at Colour Models & Spaces. An area I find particularly complicated. As mentioned, please don’t regard what you read here as fact; this is merely my understanding of it as I can gain from my independent research.
Colour Models & Colour Space/Gamuts are tightly linked to one another, to the point where I believe the terms are often misused for one another. This is hardly surprising considering RGB is a Colour Model, sRGB and Adobe RGB are Colour Spaces which is more or less the same thing as Colour Gamuts. Confusing, right?
So to try clear this up; this is how I see it:
So let’s go through these one by one, starting with Colour Models.
Colour Models & Colour Space/Gamuts are tightly linked to one another, to the point where I believe the terms are often misused for one another. This is hardly surprising considering RGB is a Colour Model, sRGB and Adobe RGB are Colour Spaces which is more or less the same thing as Colour Gamuts. Confusing, right?
So to try clear this up; this is how I see it:
- Colour Space & Gamuts are more or less interchangeable with one another. This represents the specific standard a system is using within a Colour Model. It tells us the range of colours a system has available to it.
- A Colour Model is the mathematical method that defines how numbers can be used to to represent colours.
So let’s go through these one by one, starting with Colour Models.
Colour Models
A Colour Model defines how numerical values can be used to represent colours in a given system. There are two common ways this can be achieved in video. These are RGB and YCbCr.
RGB
In an RGB Colour Model, each pixel is made up of three unique values, Red, Green and Blue. There may also be a fourth value representing an Alpha channel which allows for transparency, particularly useful for graphics and overlays. RGB is commonly used within computer displays, TV's and software as well as many digital cameras. The main issue with RGB is it’s high data redundancy levels, this is where YCbCr comes in.
RGB
In an RGB Colour Model, each pixel is made up of three unique values, Red, Green and Blue. There may also be a fourth value representing an Alpha channel which allows for transparency, particularly useful for graphics and overlays. RGB is commonly used within computer displays, TV's and software as well as many digital cameras. The main issue with RGB is it’s high data redundancy levels, this is where YCbCr comes in.
YCbCr
Most signals originate from a form of RGB at some stage in the process, for example, a camera sensor would capture raw RGB values which could be converted to YCbCr in camera. But why do we have more than one method? Although RGB and YCbCr can generally achieve more or less the same final result; the nature of each makes one more suitable for a certain purpose than the other. For example, it is possible to reduce the amount of redundant data (unneccessary information) in YCbCr far more than RGB. This makes it much more efficient and therefore more suitable for broadcast, disc based formats and streaming. Instead of having channels for each primary colour (Red, Green and Blue), YCbCr has channels for Luminance, difference-Red and difference-Blue. By separating the luminance (essentially brightness levels) from the colour channels, YCbCr allows for a process known as Chroma-Subsampling (LINK) to take place.
![Picture](/uploads/2/4/5/7/24574909/9084270.jpg?611)
With YCbCr we can clearly make out the Luminance channel; this is essentially a greyscale version of the final image; it contains brightness levels and detail. The other channels, however, (difference-red and difference-blue) are far less defined to the point where you can't tell what the image is showing.
Colour Space & Gamuts
So we know what a Colour Model is; it defines what pixel values refer to. But these values are meaningless without a scale, something those numbers can relate to. This is where Colour Space and Gamuts come in. These tell us the range of colours a system has to work with. For more details on this, take a look at Colour Space and Gamuts here (along with Colour Depth/Bit Depth)