CPSC 533 C Information Visualization Lightness Brightness Contrast

CPSC 533 C Information Visualization Lightness Brightness & Contrast Constancy Ming Huan Lee Feb. 3, 2003

Quantity of Light u Luminance “the measured amount of light coming from some region of space” u Brightness “the perceived amount of light coming from a source” (self-luminous) u Lightness “the perceived reflectance of a surface” CPSC 532 E 2

Merriam-Webster u Luminance “the luminous intensity of a surface in a given direction per unit of projected area” u Brightness “the attribute of light-source colours by which emitted light is ordered continuously from light to dark in correlation with its intensity” u Lightness “the attribute of object colours by which the object appears to reflect or transmit more or less of the incident light” CPSC 532 E 3 www. m-w. com

Agenda u Eyes: Receptive Fields & “DOG” Model u Simultaneous Contrast & Errors in Reading Maps & in Computer Graphics u Lightness Constancy: Adaptation & Contrast u Computer Monitor u Conclusion CPSC 532 E 4

Our Eyes u light-sensing receptors n u no info about amount of light (light meter) signal of “differences” (contrast mechanism) n relative amount of light (change meter) u n n CPSC 532 E differ from neighbouring / change in the past instant nonlinear, no absolute values (but not inaccurate) reflectance of surfaces 5

Human Vision u perceive objects from surface properties, not from the quality and quantity of illumination n CPSC 532 E lightness is unaffected by illumination lightness constancy 6

Receptive Fields the receptive field of a cell is the visual area over which a cell responds to light u stimulate receptive field u n n CPSC 532 E on-centre: emits pulses at a greater rate off-centre: emits pulses at a lower rate (inhibited) 7

“DOG” Model u Difference of Gaussians (DOG) n u DOG receptive field explain a variety of brightness contrast effects n CPSC 532 E the firing rate of the cell is the difference between two Gaussians (centre and surround) Hermann Grid: more inhibition between squares, hence brighter at the intersections 8

Simultaneous Brightness Contrast u the general effect whereby a grey patch placed on a dark background looks lighter than the same grey patch on a light background cross-cross & snake CPSC 532 E 9

Simultaneous Contrast & Errors u contrast n effects vs. visualization errors of judgment while reading quantitative (value) info encoded using gray scale u ex. in the worst case, found substantial errors that averaged 20% of the entire scale Gravity Map of North Atlantic large errors can occur when values are read using the key CPSC 532 E 10

Gray Scale no perfect gray scale u providing only rough approximations u n because the visual field is radically changed by many factors perceived illumination u specular reflection from glossy surfaces u local contrast effects u n contrast crispening u CPSC 532 E 11 differences are perceived as larger when samples are similar to the background color

Computer Graphics computer graphics images (not like real world) consist of simple luminous light patches u much of the detailed (high level) texture and shadow info (like real world) is missing u leads to errors and distortions in info displays u consequences of contrast effects u n n tend to show up the deficiencies in the common shading algorithms used in computer graphics lead to perception of illusory patterns smooth surfaces displayed using polygons (for simplicity and to speed up computer graphics rendering process) u visual system enhances boundaries at edges of polygons (stand out) 12 u CPSC 532 E

Lightness Constancy is unaffected by illumination lightness constancy u two important mechanisms in lightness constancy, which help the visual system to factor out the effects of the amount and color of the illumination u lightness n n CPSC 532 E adaptation contrast 13

Adaptation u the n n role the changing sensitivity of the receptors and neurons in the eye helps factor out the overall level of illumination allow visual system to adjust overall sensitivity to the ambient light level u example n CPSC 532 E briefly blinded when coming into a darkened room out of bright sunlight 14

(a) white u medium gray help to achieve constancy by signalling differences in light levels, especially at the edge of objects Example n n (a) receives more light, however, it reflects about the same amount of light as (b) simultaneous brightness contrast u CPSC 532 E Contrast the role n u (b) 15 (a) is lighter relative to its background than (b) is, relative to its background

Additional Factors u although both adaptation and contrast can be seen as mechanisms that act in service of lightness constancy, they are not sufficient n n CPSC 532 E brain must take the direction of illumination and surface orientation into account in lightness judgements brain seems to use the lightest object in the scene as a kind of reference white to determine the grey values of all other objects 16

Additional Factors (Cont’d) n the ratio of specular and nonspecular reflection u in the all-dark world, the ratio is much larger than in the all-white world Every point on the black image is brighter than the surroundings. How can we perceive something to be black when it is a bright image? CPSC 532 E 17

Computer Monitor u contrast illusions are much worse in CRT n n no texture (uniform pattern of pixels and phosphor dots) self-luminous (confound lightness constancy) the colour and the brightness of the surround of the monitor can be very important in determining how screen objects appear u the adaptation effect produced by room light u overall contrast is much reduced where the room light falls on the display u LCD? u CPSC 532 E 18

Conclusion nervous system computing difference signals u visualization is not good for representing precise absolute numerical values, but rather for displaying patterns of differences of changes over time, to which the eye and brain are extremely sensitive u contrast effects are severe in computer displays as a consequence of the impoverished nature of those displays, not of any inadequacy of the visual system u CPSC 532 E 19

Color Use Guidelines for Data Representation CPSC 532 E 20

Color Schemes u make the most of the information visualization by using perceptual dimensions of color in ways that parallel the logical structures in the data to allow its organization to be readily perceived n n n CPSC 532 E sequential qualitative and binary diverging spectral two-variable 21

Sequential Schemes u u u suited to ordered data that progress from low to high lightness to represent ordered data with light colors for low data values to dark colors for high data values n ex. a graphic showing pollution level CPSC 532 E 22

Qualitative and Binary Schemes u u do not magnitude differences between legend classes, and hues are use to create the primary difference between classes best suited to representing nominal or categorical data n u ex. government spending for different departments binary schemes n special case for the two categories CPSC 532 E 23

Diverging Schemes u u put equal emphasis on mid-range critical values and extremes at both ends of the data range the critical class or break in the middle of the legend is emphasized with light colors, and low & high extremes are emphasized with dark colors that have contrasting hues n ex. increases and decreases in acid rain over a given time period CPSC 532 E 24

Spectral and Two. Variable Schemes u spectral (rainbow) n n CPSC 532 E often misused as a sequential scheme with poorly used lightness differences the most informative use of a spectral scheme is as a diverging scheme 25 u two-variable n n n sequential-sequential-qualitative binary-diverging sequential-diverging-diverging

Conclusion u matching the organization of the perceptual dimensions of color (hue, lightness, saturation) to the organization of data being represented is one key to gaining insight from data visualization CPSC 532 E 26