Lec06, Color Space, v1.08.ppt -...

Multimedia SystemsMultimedia Systems

Color SpaceColor Space

Course PresentationCourse Presentation

Mahdi Amiri

March 2014

Sharif University of Technology

Physics of ColorLightLight

Light or visible light is the portion of electromagnetic radiation that

is visible to the human eye.

The wavelength λ of a sinusoidal waveform

traveling at constant speed ν is given by

In the case of electromagnetic radiation, ν is

the speed of light, about 3×108 m/s.

Page 1 Multimedia Systems, Mahdi Amiri, Color Space

Color is Human SensationHuman EyeHuman Eye

Through cone and rod receptors in the retina

Three kinds of cones: Long (L, Red), Medium (M, Green), Short (S, BlueS, Blue)

Rod receptor is mostly for luminance perception(useful for night vision)


Human EyeFoveaFovea

The fovea is responsible for sharp central vision

The human fovea has a diameter of about 1.0 mm with a high

concentration of cone photoreceptors.

Cone cells are densely packed in the fovea, but gradually

become sparser towards the periphery of the retina.

A rod cell is sensitive enough to respond to a single

photon of light and is about 100 times more sensitive to a

single photon than cones.

Page 3

Spatial DistributionSpatial Distribution

Multimedia Systems, Mahdi Amiri, Color Space

Human EyePhotoreceptor cellPhotoreceptor cell

Normalized typical human cone (and rod) absorbances

(not responses) to different wavelengths of light

IdeaIdea

Page 4

IdeaIdea

We can have different We can have different

colors by mixing colors by mixing

primary color primary color

componentscomponents

of “light” or “pigment”of “light” or “pigment”


Additive Color MixingMixing of LightMixing of Light

Usually red, green and blue

RGB color model

Application

LCD, LED, plasma and CRT

Page 5

LCD, LED, plasma and CRT

(picture tube) color video displays

Examine TV display with a Examine TV display with a

sufficiently strong magnifiersufficiently strong magnifier


RGBLCD TV CloseLCD TV Close--upup

RGB subRGB sub--pixels in an LCD TVpixels in an LCD TV

(on the right: an orange(on the right: an orange

and a blue color; and a blue color;

on the left: a closeon the left: a close--up)up)


RGBColor ComponentsColor Components


Subtractive Color MixingMixing of PigmentMixing of Pigment

Primary: Cyan, Magenta, Yellow

CMYK color model

Application

Printers

Page 8

Printers

Pigments Pigments

absorb lightabsorb light


CMYKColor ComponentsColor Components


CMYKWhy Including the Black?Why Including the Black?

Separated with Separated with

maximum black, maximum black,

to minimize ink to minimize ink

use.use.

Page 10

Separated for printing with Separated for printing with

process cyan, magenta, and process cyan, magenta, and

yellow inks.yellow inks.

A color photograph of A color photograph of

the Teton Range.the Teton Range.


CMYKCloseClose--up of a Printup of a Print

This closeThis close--up of printed halftone rasters up of printed halftone rasters

show that magenta on top of yellow show that magenta on top of yellow

appears as orange/red, and cyan on top appears as orange/red, and cyan on top

of yellow appears as greenof yellow appears as green

Page 11

of yellow appears as greenof yellow appears as green


Color Space ModelsCIE 1931 XYZ color spaceCIE 1931 XYZ color space

One of the first mathematically defined color

spaces

Created by the International Commission on

Illumination (CIE)

Page 12

Illumination (CIE)

The CIE XYZ color space was derived from a

series of experiments done in the late 1920s by W.

David Wright and John Guild.


Color Space ModelsCIE ExperimentsCIE Experiments

The observer would alter the brightness The observer would alter the brightness

of each of the three primary beams until of each of the three primary beams until

a match to the test color was observed.a match to the test color was observed.

Page 13

The Y primary is intentionally defined to have a The Y primary is intentionally defined to have a

colorcolor--matching function that exactly matches the matching function that exactly matches the

luminousluminous--efficiency function of the human eye.efficiency function of the human eye.

CIE XYZ primariesCIE XYZ primaries


Color Space ModelsCIE 1931 XYZ color spaceCIE 1931 XYZ color space

The Result of the Experiments: The cone of visible colors

Page 14

The X + Y + Z = 1 plane is shown as a triangleThe X + Y + Z = 1 plane is shown as a triangle


Color Space ModelsCIE xyY Chromaticity DiagramCIE xyY Chromaticity Diagram

Page 15

Y in Y in xyYxyY: measure of the brightness or : measure of the brightness or

luminance of a color.luminance of a color.


Color Space ModelsColor GamutColor Gamut

The range of color representation of a display device

Page 16

A typical CRT gamutA typical CRT gamut


Color SpaceHSVHSV

Hue: Quantity that distinguishes color family,

say red from yellow, green from blue.

Saturation (Chroma): Color intensity (strong to

weak). Intensity of distinctive hue, or degree of

Page 17

weak). Intensity of distinctive hue, or degree of

color sensation from that of white or grey.

Value (Luminance): Light color or dark color.


Color SpaceHSVHSV

Photoshop uses this model to get more control over color


Color SpaceHSV vs. RGBHSV vs. RGB

Example color control

From a relatively colorful orange �� a less saturated orange

RGB: we would need to readjust all three R, G and B sliders

HSV: we can just readjust the ‘Saturation’ related slider

Page 19

HSV: we can just readjust the ‘Saturation’ related slider

Sample Application of HSV color space: Object detection based on color filtering.


Color SpaceYUVYUV

Y: Luminance

U and V: Chrominance

Taking human perception into account, Taking human perception into account,

allowing reduced bandwidth for allowing reduced bandwidth for

YUV was invented when engineers

wanted color television in a black-

and-white infrastructure.

Page 20

Example of UExample of U--V color planeV color plane

allowing reduced bandwidth for allowing reduced bandwidth for

chrominance componentschrominance components

(Since the human eye is less sensitive

to chrominance than luminance)


No need for color space conversion No need for color space conversion

when our image processing technique when our image processing technique

only uses luminance component.only uses luminance component.

YUV's corresponding color space in

ITU standards is YCbCr

U corresponds to the Cb = B – Y

V corresponds to the Cr = R – Y

YUVColor ComponentsColor Components


Color SpaceSubsampling in YUV or YCbCrSubsampling in YUV or YCbCr


Color SpaceDifference Between Four SchemesDifference Between Four Schemes

This image shows the difference between four This image shows the difference between four subsamplingsubsampling

schemes. Note how similar the color images appear. The schemes. Note how similar the color images appear. The

lower row shows the resolution of the color information.lower row shows the resolution of the color information.


RGB, HSV, YUVRGB, HSV, YUV


RGB, CMY, YUVRGB, CMY, YUV


Color Space ConversionRGB and YUVRGB and YUV

RGB to YUVRGB to YUV

Page 26

YUV to RGBYUV to RGB


Color Space ConversionRGB and CMYKRGB and CMYK

CMY to CMYKCMY to CMYKCMY to RGBCMY to RGB

Page 27

CMYK to CMYCMYK to CMYRGB to CMYRGB to CMY


Color SpaceGamma CorrectionGamma Correction

Ou

tpu

t li

gh

t

(bri

gh

tnes

s)


Input signal level

Ou

tpu

t li

gh

t

(bri

gh

tnes

s)

Low gamma (e.g. 1.5)

Low

gamma

Looks “milky”, flat

and lacks punch

With visible details in

the black areas and

Source: Piers Clerk, ISF Calibrator,

www.homecinemaengineering.comGamma control in HDTVs

(May be found in advanced

picture settings menu)

Page 29Multimedia Systems, Mahdi Amiri, Color Space

Normal gamma (e.g. 2.2)

High gamma (e.g. 3.0)

High

gammaHas loads of punch;

but no detail in the

black areas

the black areas and

white areas.

Color SpaceGamma CorrectionGamma Correction

Difference between a scale with linearly-increasing encoded luminance signal (linear input)

Page 30

Difference between a scale with linearly-increasing encoded luminance signal (linear input)

and a scale with linearly-increasing intensity (i.e., gamma-corrected) scale (linear output).


Visual perceptionVisit Mind Lab WebsiteVisit Mind Lab Website

Page 31

http://jvsc.jst.go.jp/find/mindlab/english/index.htmlhttp://jvsc.jst.go.jp/find/mindlab/english/index.html

How is our consciousness connected to the world?How is our consciousness connected to the world?

Explore the unconscious functions of the brain with visual Explore the unconscious functions of the brain with visual

illusions and mysterious perceptual phenomenaillusions and mysterious perceptual phenomena


Thank You

Multimedia SystemsMultimedia Systems

Color SpaceColor Space

Page 32

Thank You

1. http://ce.sharif.edu/~m_amiri/

2. http://www.dml.ir/

FIND OUT MORE AT...

Next Session: Image INext Session: Image I


Lec06, Color Space, v1.08.ppt -...

Documents

Transcript of Lec06, Color Space, v1.08.ppt -...