CPSC 453 – Computer Graphics Colour

CPSC 453 – Computer GraphicsColourSheelagh CarpendaleColour colour is a complex subject understanding of it is incomplete and comes from many differentsubject areas the colour of an object depends on psychophysics, physiology, psychology, art and design the object itself, the light source, the surrounding colour, and the way it isperceived it is heavily used, part of all cultures and as such has many differentpre-assigned meaningsWhy use colour? to make the display more attractiveto emphasize and draw attentionto increase readabilityto encode /colortheory.html1

Colour Light occupies wavelengths from 350to 780 nm Can be characterized by a functioninfraredC(λ)ultraviolet350 780displays do not do thisStill if we see a colour in the real worldwe would like to match it on the displayColour the human retina has 3 kindsof cones.each type of cone respondsas a function of thewavelength of the incidentlightpeaks for each curve are: Blue: 440nm,Green 545nmRed 580nm (red). Note(last two actually peak in theyellow part of the spectrum).perception of colour is anentirely arbitrary creation ofour nervous system, and isnot contained in thewavelengths2

Colour match any colour with a mixture of 3 lights (primaries) – reasonableapproach but have reduced a continuous function to 3 numbers Additive colour combining different intensities gives different coloursC T1 R T2 G T3 BT1 T2 T 3 1.0 gives white, T1 T 2 T3 0.0 gives black,Shades of grey T1 T2 T 3Colour CRT, printer, film, paint, human eye – all operate differently inregards to primariesPaint Colours tints and shades red, yellow, blue all three primaries givebrown RGB Colour additive colours computer monitors red, green, and blue absence all three colours givesblack, all three gives white.CMYK Colour Colour Subtraction printing cyan, magenta, yellow absence all three coloursgives white, all three givesblack.Translations between maybe mathematically accurate without beingvisually accurate3

Perceiving Colour An active organizing process Many illusions – experimentsfrom psychology (if interestedInformation Visualization: Perception forDesign. Colin Ware) Colour, as we perceive, it is notjust a matter of measuringwavelength We perceive difference – notabsolutesThe eye is not a photometer Colour and context4

Chromatic colour Munsell system – set of standard (fromhuman observers) samples in a 3Dspace ofHue, Saturation, Value (HSV) Complimentary colours – add togetherto give whiteColour Colour Gamut – the range of colours that can be createdby mixing a given set of primary colours5

Aside - Opponent Process Theory Late 19th C. Ewald Hering6 elementary coloursarranged perceptually as opponent pairs along 3 axes: black – whitered – greenblue – yellowcornerstone of modern colour theorywell established physiological basisColour In OpenGL APIRGB model – conceptually 3 buffers – R, G, and BA given number of bits per buffer8 per buffer or 24 in totalgives 224 colours or 16 M (where M is 1024) R, G and B are specified separatelyglColor3f(1.0, 0.0, 0.0); Actually GL uses a fourth colour specification RGBA AlphatransparencyglClearColor(1.0, 1.0, 1.0, 1.0);6

Colour – indexed colour Is it possible to have a reasonable range of colours with a smallframe buffer?Indexed Colour Alternative method of associating pixel values with colour Colour Look Up Table (LUT) Example: setcolour (colour-index, red-value, blue-value, greenvalue);Palette: set of possible colours that the system is capable ofdisplaying 7

Colour: number of palettes k: number of planes in the frame buffer (bit) m: width of each entry in the LUT per colour (bits)(2m reds, 2m greens, 2m blues) 23m possible colours (palettes)2k colors at one timeUser constructs LUT – using m bits for each RGBEntries in the LUT can specified by index Advantages Less cost of memory More flexible Largely used when frame buffer cannot support fullcolourColour ReferencesInformation Visualization, Perception for Design; Colin Ware. Morgan KaufmannPublishers.An Introduction to Natural Vision; S. A. J. Winder. PhD Thesis: “From Cones to Contours:A Parallel Simulation of Neural Mechanisms in the Primate Vision System”, can.gc.ca/ccrs/eduref/sradar/chap2/c2p2 g2e.html8

1 CPSC 453 – Computer Graphics Colour Sheelagh Carpendale Colour colour is a complex subject unders