Shading

0
0
2391 days ago, 578 views
PowerPoint PPT Presentation
Product, Chapter 4. The exploration of shading visionColour estimation frameworks and standardsOpponent process theoryApplications. . . . The investigation of shading vision. Receptors and trichromacy hypothesis. . Red ?. Blue ?. Green ?. . Shading estimation frameworks and benchmarks. Any shading can be coordinated utilizing a mix of three

Presentation Transcript

Slide 1

Shading CPSC 533C February 3, 2003 Rod McFarland

Slide 2

Ware, Chapter 4 The exploration of shading vision Color estimation frameworks and norms Opponent process hypothesis Applications

Slide 3

The investigation of shading vision Receptors and trichromacy hypothesis Red  Green  Blue 

Slide 4

Color estimation frameworks and measures Any shading can be coordinated utilizing a blend of three "primaries". The primaries are not really red, green, and blue. Any three distinct hues can be utilized. The scope of hues that can be delivered from a given arrangement of primaries is the range .

Slide 5

CIE (Commission Internationale d'éclairage) Primaries decided for numerical properties: don't really compare to hues. These "virtual" hues X, Y, and Z are called tristimulus values . Y is the same as luminance Color guidelines

Slide 6

CIE – Chromaticity is gotten from tristimulus values: Since x+y+z=1, simply utilize x, y qualities and luminance (Y). Chromaticity graph:

Slide 7

Uniform shading space: a portrayal where square with separations in space compare to equivalent separations in recognition Useful for: Specification of shading resiliences Color coding (greatest refinement) Pseudocolour successions to speak to requested information values CIE XYZ shading space is not uniform CIEluv is a change of the chromaticity chart Uniform Color Space - CIEluv

Slide 8

CIEluv does not take care of all issues: Contrast impacts Small shading patches: hard to recognize hues in the yellow-blue heading

Slide 9

Black-white (luminance), red-green, and blue-yellow adversaries Has premise in science and culture Should utilize adversary hues for coding information Opponent process hypothesis

Slide 10

Properties of Color Channels Isoluminant/Equiluminous designs: a shading example whose parts don't vary in luminance Red-green and yellow-blue stations convey just around 1/3 of the detail conveyed by dark white.

Slide 11

Yellow Text on a Blue Background Is genuinely simple to peruse unless the content is isoluminant with the foundation shading. As the luminance of the foundation turns into the same as the luminance of the content, it is exceptionally hard to make out what the content says. To such an extent, that now I can compose pretty much anything I need here and barely anybody would need to invest the push to perceive what it was I had composed.

Slide 12

Other isoluminance impacts Stereoscopic profundity is not noticeable with isoluminant hues Isoluminance in liveliness makes it have all the earmarks of being slower than a similar activity in high contrast Shape and frame are best indicated utilizing luminance:

Slide 13

Color appearance Contrast Saturation Brown low high

Slide 14

Applications Color choice interfaces Color naming Natural Color System (NCS) e.g. 0030-G80Y20 Blackness 00, power 30, green 80, yellow 20 Pantone, Munsell: standard shading chips

Slide 15

Applications Color for naming (ostensible data encoding) Distinctness A quickly recognized shading lies outside the curved polygon characterized by alternate hues in CIE space

Slide 16

Applications Color for naming (2) Unique tints: "all around perceived" tones (red, green, blue, yellow, dark, white) ought to be utilized Contrast with foundation: fringe around articles

Slide 17

Applications Color for marking (3) Color visual impairment: greater part of partially blind individuals can't recognize red-green, yet the vast majority can recognize blue-yellow Number: just 5-10 codes effortlessly recognized

Slide 18

Applications Color for naming (4) Size Color-coded items ought not be little (about ½ degree least size). Littler articles ought to be all the more very soaked, huge shading coded districts ought to have low immersion. Content highlighting ought to be high-luminance, low-immersion. Traditions Common utilization of hues, e.g. red=stop, green=ready, blue=cold…

Slide 19

Applications Color for naming (5) Ware's 12 suggested hues (all together of inclination):

Slide 20

Applications Pseudocolour arrangements for mapping Pseudocolouring is the act of doling out shading to guide values that don't speak to shading Medical imaging Astronomical pictures Mapping nonvisible range data to the unmistakable range (cosmology, infrared pictures) Gray scale best to show surface shape Color best for characterization

Slide 21

Applications Color for mapping (2) For orderable successions, dark white, red-green, blue-yellow, or immersion (dull-distinctive) grouping can be utilized. For nitty gritty information, the arrangement ought to be construct predominantly with respect to luminance. For low letail, chromatic or immersion groupings can be utilized. Uniform shading spaces can be utilized to make shading groupings where break even with perceptual strides compare to equivalent metric strides. Where it is critical to have the capacity to peruse off qualities from a shading guide, an arrangement that spins through many hues is ideal.

Slide 22

Applications Color for mapping (3) A "winding" through shading space (going through a few hues while ceaselessly expanding in luminance) is frequently a decent decision. Tint 0, 50,… 250, 45, 95… Luminance 0, 25, 50… 225

Slide 23

Applications Color for mapping (4) Perception: regardless of the possibility that the arrangement is smooth, individuals tend to see discrete hues, conceivably miscategorizing information. My own division into blue, green, yellow, orange, red, purple: exceptionally nonlinear

Slide 24

Applications Color for mapping (5) Using shading for 3-D data mapping Difficult to peruse precisely May be utilized to distinguish districts Satellite pictures: areas of imperceptible range mapped to red, green, blue stations

Slide 25

Applications Color for multidimensional discrete information 5-D plot utilizing (x, y) position, red, green, blue Possible to recognize bunches Ambiguous: is a point low-red or high-green? Different techniques expected to investigate bunches once distinguished

Slide 26

Rogowitz et al. How Not to Lie with Visualization Visual portrayal of information influences the apparent structure of the information.

Slide 27

Enhancing information interpretation utilizing Color Perceptual effect of a shading is not unsurprising from the red/green/blue segments of the shading Mapping diverse parts of shading to various information is not instinctively decodable by clients. Default shading maps: rainbow Perceptual nonlinearity False shapes Yellow pulls in consideration

Slide 28

Guiding shading map determination Constrain the arrangement of shading maps accessible to the client in view of: Data sort Data spatial recurrence Visualization assignment Other plan decisions made by client

Slide 29

Representing Structure Nominal information Object ought to be noticeably unique however not perceptually requested Ordinal information Distinguishable with perceptual requesting Interval information Equal strides in information compare to equivalent strides in saw size Ratio information Zero point recognizable in shading grouping

Slide 30

Structure Magnitude of a variable at each spatial position Use luminance (dark scale) or immersion

Slide 31

Spatial Frequency high spatial recurrence low immersion based luminance-based

Slide 32

Segmentation Low recurrence – more division steps can be utilized

Slide 33

Highlighting Luminance-based guide can be highlighted utilizing tone varieties. The highlighted districts have a similar luminance esteem as whatever remains of the guide.

Slide 34

PRAVDA Perceptual Rule-Based Architecture for Visualizing Data Accurately Part of IBM's Visualization Data Explorer ( http://www.research.ibm.com/dx/) Provides decisions for shading maps in light of spatial recurrence, information sort, and client chose objective: isomorphic (structure-safeguarding), division, highlighting

Slide 35

PRAVDA

SPONSORS