 |
 |
|
||
 |
 |
|||
Ocean colour: why we see colours
Colour in the outside world Colour as we see them Colour models
You are here: why we see colours
White light is not just white; it is made up of a whole range of different colours - the colours you see in a rainbow. You can also see them if you shine a beam of white light through a clear glass prism.
Two explanations for colour
There are two explanations for why we see colours.
- Firstly colour is caused by light interacting with the world around us.
- Secondly it is a result of the way our eyes detect light, and how our brains interpret what we see.
The colours we see are a result of both, so to understand colour, we need to look at each in more detail.
Absorption Transmission Scattering Reflection
Click on the text above to change the animation and see how photons interact with objects in the world around us.
Grass absorbs red and blue wavelengths, leaving green, and the grass looks green to us. The spectrum shows the colour reflected from grass Any other colour you see can also be plotted as a spectrum.
|
|
| |
|
|
Left: Sunlight contains all the visible wavelengths and therefore reveal all colours well. Right: Traditional lamp light contains less blue, so everything gets a warm glow.
|
|
| |
|
|
Left: Mercury light is has bright bands that activate all three types of cones, so you can see colours almost as well as in daylight. Right: Sodium street lights contain only orange. Other colours look orange, brown or black.
Colours in the outside world
The colours you see correspond to different wavelengths of visible light. Violet is the shortest and red the longest of the visible wavelengths. Light from the sun contains all these wavelengths, but looks white to us.
When a stream of photons (light) falls on the surface of an object (an apple, a window, you, me, ) several things may happen. The photons may
react with atoms in the object and disappear - absorption
pass through to the other side - transmission
bounce off in a new direction - reflection or scattering
Absorption is what gives the object its colour. Everything around us absorbs light, but different things absorb different wavelengths. You see the wavelengths that are not absorbed.
Some objects emit (give off) their own light. This can happen because they are hot enough to glow or because they can change electric or chemical energy into light.
Colours change with incident light
The colour of an object also depends on the illumination - the type of light that is available to fall on an object and be reflected towards your eyes. (The light falling on an object is known as the 'incident light').
At first this may sound strange, but if you think about it for a moment it becomes obvious: if there is no blue light available, there are no blue photons reaching your eyes. So even if a ball is actually blue, in yellow lamplight it will look black, because there are no blue photons bouncing off the ball to enter your eyes.
Our eyes and brains are good at compensating for changes in incident light, adjusting to bright or dim light without us even noticing. But if certain colours are missing from the light altogether, we cannot see them, even if our brains are better at image processing than the most sophisticated image processing software. The photos on the right illustrate this.
Next time you spend time in warm, cosy lamplight, or in the orange light from sodium street lamps - try to see how this light changes the colours around you.
|
|
When light enters the eye, it passes through the cornea, the pupil and the lens, which focuses it (upside down) onto the retina at the back of the eye.
The rods are very light sensitive and respond even at night. The cones react only to high light intensities, and only to a limited wavelength range.
Absorption spectra of rods (black) and the three different types of cone in the human eye. As you see the red cones respond most strongly to orange rather than red. At true red wavelengths only the red cones respond. At yellow, both green and red cones respond.
All colours we see are worked out from the relative response of the three cones - red, green and blue. As a result we can recreate an impression of all the wavelengths and colours using only three wavelengths of light - red, green and blue. This is how colour TV and computor monitors work.
Colours as we see them
Light that enters the eye falls on the retina at the back of the eye. The retina contains two types of light sensitive receptors - the rods and the cones. Rods are responsible for night vision, and are found in the peripheral retina. Cones are responsible for colour vision. They are concentrated in the central part of the retina and work only in daylight.
Normal human colour vision has three types of cones: Red, green and blue. Each type of cone is most sensitive to a specific wavelength of visible light. However, the sensitivity of the cones overlap, so a particular wavelength of light may stimulate two types of cone.
When a cone is stimulated, it sends a signal along the optical nerve to the brain. Different wavelengths of light stimulate the cones in different combinations, and the brain interprets these signals as colours.
For example:
|
Colour models
We now know a few facts about colour and colour perception, which can help us understand how colour is produced on computer screens and TV, in photography, and in painting and printing.
The colour of light depends on the wavelength of that light.
The colour of an object is determined by the wavelengths that are NOT absorbed by the object.
We see colours as a result of the light reaching our eyes from objects around us falling on cones in the retina, which are receptive to red, green and blue wavelengths respectively.
){kind=link}
){kind=link}
The additive (RGB) colour model adds light to darkness (black). Click on the text links under the figure to see how the three primary colours are used to build an image.
The additive colour model
In the additive colour model we start with no light (black). We then add light at three different wavelengths, each activating one type of cone in the retina: red, green or blue. Two of these 'primary' colours activate two types of cone, and the mixture of light looks like light of a new colour.
For example: Strong light at 550nm (green) and 620nm (red) looks the same as light at 575 nm (yellow). It's still light at 550 and 620 nm, of course, but to us these two kinds of yellow look the same to us.
The additive colour model is also known as the RGB model after its three primary colours, Red, Green and Blue. Video cameras and computer displays use this model.
Cyan
Magenta
Yellow
Black
Reset
The subtractive (CMYK) colour model subtracts colours from strong light containing all colours (white).
Click on the text-links under the figure to see how an image is built up by adding ink in the three primary CMYK colours and black.
The subtractive colour model
In the subtractive colour model you start with a clean sheet (white) and create the image by covering it with different coloured inks or paints. Each colour removes (subtracts) some wavelengths of light from the many reflected by the white sheet.
For example: Cyan absorbs red, so covering an area with cyan ink removes red. When you look at it, only the green and blue cones in your retina react.
The subtractive colour model is also known as the CMYK model after its three primary colours, Cyan, Magenta and Yellow. To add shading and make the black truly black with need a fourth ink: blacK. The CMYK model is the model used for dying, painting and printing.
You may come across other colour models such as HSI (hue, saturation, intensity) and YIQ (luminance-inphase-quadrature). Our links contain more information about these.
Why do we need colour models?
Colour is a difficult thing to describe accurately. We have only a handful of colour words, but as I'm sure you have noticed, words like 'blue' can contain many different shades. So how do we distuingish them in a clear way? - With a colour model.
In a model each colour has a numerical colour code, which can be used to reproduce exactly the colour you intended. With colour codes we can pass instructions to computers and other machines. The models make modern colour printing, computer graphics and image processing possible.
Left: The colour cube. By plotting red, green and blue along three different axes (R,G,B) we can specify a large number of colours exactly. Black (0,0,0) has no light of any colour, while white (1,1,1) contains full strength light of all three. Diagonally from black to white lie all the different shades of grey; with equal amounts of red, green and blue light. Colours that contain only red light lie along the 'R' axis. Similarly with green and blue. Opposite corners of the cube give you complementary colours. Thus blue (0,0,1) complements yellow (1,1,0). Light of complementary colours add up to white, so red (1,0,0) complements cyan (0,1,1), making white (1,1,1).
Note: If you like digital photography and have tried to adjust your photos with image processing software, or if you have tried your hand at web authoring, you may have noticed that colours can be specified by six numbers or letters. These represent 2 digits for red, 2 for green and 2 for blue, in that order. Each is a hexadecimal number (base 16) specifying a strength between no light (00) and full light (FF). For example #FF0000 is bright, pure red (1,0,0 on the colour cube). Similarly, #00FF00 is bright, pure green (0,1,0). And what colour is specified by #0000FF?
You guessed it - it's bright, pure blue.
|
Last update: 28 November 2008 |
Contact: o4s@noc.soton.ac.uk |