Shade FAQ – Incessantly Requested Questions Shade

2022-08-15 07:48:56

a.ok.a. Color FAQ – Incessantly Requested Questions Color

This doc clarifies features of color specification and picture coding
which can be vital to pc graphics, picture processing, video, and the
switch of digital photos to print.

I assume that you’re acquainted with depth, luminance (CIE Y), lightness
(CIE L*), and the nonlinear relationship between CRT voltage and depth
(gamma). To be taught extra about these subjects, learn the companion
Frequently Asked Questions about Gamma.

This doc is obtainable on the Web:
http:// www.poynton.com/notes/colour_and_gamma/

A typeset-quality model of this doc is obtainable:
(Acrobat PDF format).

I retain copyright to this observe. You’ve permission to make use of it, however you
might not publish it.

A translation
to Belorussian is obtainable,
because of Paul Bukhovko.

One other supply of knowledge regarding colour is the doc Frequently asked questions
about Colour Physics, by Steve
Westland <coa23@cc.keele.ac.uk>.

0. Where does this document live?
1. What is colour?
2. What is intensity?
3. What is luminance?
4. What is lightness?
5. What is hue?
6. What is saturation?
7. How is colour specified?
8. Should I use a colour specification system for image data?
9. What weighting of red, green and blue corresponds to brightness?
10. Can blue be assigned fewer bits than red or green?
11. What is “luma”?
12. What are CIE XYZ components?
13. Does my scanner use the CIE spectral curves?
14. What are CIE x and y chromaticity coordinates?
15. What is white?
16. What is colour temperature?
17. How can I characterize red, green and blue?
18. How do I transform between CIE XYZ and a particular set of RGB primaries?
19. Is RGB always device-dependent?
20. How do I transform data from one set of RGB primaries to another?
21. Should I use RGB or XYZ for image synthesis?
22. What is subtractive colour?
23. Why did my grade three teacher tell me that the primaries are red, yellow and blue?
24. Is CMY just one-minus-RGB?
25. Why does offset printing use black ink in addition to CMY?
26. What are colour differences?
27. How do I obtain colour difference components from tristimulus values?
28. How do I encode Y’PBPR components?
29. How do I encode Y’CBCR components from R’G’B’ in [0, +1]?
30. How do I encode Y’CBCR components from computer R’G’B’ ?
31. How do I encode Y’CBCR components from studio video?
32. How do I decode R’G’B’ from PhotoYCC[tm]?
33. Will you tell me how to decode Y’ UV and Y’ IQ?
34. How should I test my encoders and decoders?
35. What is perceptual uniformity?
36. What are HSB and HLS?
37. What is true colour?
38. What is indexed colour?
39. I want to visualize a scalar function of two variables. Should I use RGB values corresponding to the colours of the rainbow?
40. What is dithering?
41. How does halftoning relate to colour?
42. What’s a colour management system?
43. How does a CMS know about particular devices?
44. Is a colour management system useful for colour specification?
45. I’m not a colour expert. What parameters should I use to code my images?
46. References
47. Contributors

This doc resides in Toronto, linked from my
ColorFAQ
page <http://www.poynton.com/ColorFAQ.html>.
Must you want to make a hyperlink to the Color FAQ, hyperlink to that web page, not
on to this one. It will permit me the flexibleness to maneuver it, and
keep away from the nuisance of stale hyperlinks.

When you’ve got an Acrobat Reader, or if you want to print this doc,
I like to recommend that you just receive the Acrobat
PDF version (277356 bytes)
of this observe. You’ll find
information about document formats.

I retain copyright to this observe. You’ve permission to make use of it, however you
might not publish it.

Color is the perceptual results of gentle within the seen area of the
spectrum, having wavelengths within the area of 400 nm to 700
nm, incident upon the retina. Bodily energy (or radiance) is expressed
in a spectral energy distribution (SPD), usually in 31 elements every
representing a ten nm band.

The human retina has three varieties of color photoreceptor cone cells,
which reply to incident radiation with considerably completely different spectral response
curves. A fourth kind of photoreceptor cell, the rod, can also be current
within the retina. Rods are efficient solely at extraordinarily low gentle ranges (colloquially,
night time imaginative and prescient), and though vital for imaginative and prescient play no position in
picture copy.

As a result of there are precisely three varieties of color photoreceptor, three
numerical elements are crucial and ample to explain a color,
offering that acceptable spectral weighting features are used. That is
the priority of the science of colorimetry. In 1931, the Fee
Internationale de L’Éclairage (CIE) adopted normal curves for a
hypothetical Normal Observer. These curves specify how an SPD can
be reworked right into a set of three numbers that specifies a color.

The CIE system is straight away and virtually universally relevant to self-luminous
sources and shows. Nevertheless the colors produced by reflective methods
akin to images, printing or paint are a operate not solely of the colourants
but additionally of the SPD of the ambient illumination. In case your software has
a robust dependence upon the spectrum of the illuminant, you might have to
resort to spectral matching.

Sir Isaac Newton mentioned, “Certainly rays, correctly expressed, will not be
colored.” SPDs exist within the bodily world, however color exists solely
within the eye and the mind.

Depth is a measure over some interval of the electromagnetic spectrum
of the movement of energy. Depth
is what I name a linear-light measure. The usual SI unit for
luminous depth is the candela (cd).

The voltages offered to a CRT monitor management the intensities of the
color elements, however in a nonlinear method. CRT voltages will not be proportional
to depth.

For data on problems with nomenclature and models surrounding depth,
See the article Getting
Intense on Intensity, by James
M. Palmer. You could be considering his
FAQ
on Radiometry and Photometry.

Brightness is outlined by the CIE as the attribute of a visible
sensation in accordance with which an space seems to emit roughly gentle.
As a result of brightness notion could be very complicated, the CIE outlined a extra tractable
amount luminance which is radiant energy weighted by a spectral
sensitivity operate that’s attribute of imaginative and prescient. The luminous effectivity
of the Normal Observer is outlined numerically, is all over the place optimistic,
and peaks at about 555 nm. When an SPD is built-in utilizing this curve
as a weighting operate, the result’s CIE luminance, denoted
Y.

The magnitude of luminance is proportional to bodily energy. In that
sense it’s like depth. However the spectral composition of luminance is
associated to the brightness sensitivity of human imaginative and prescient.

Strictly talking, luminance must be expressed in a unit akin to candelas
per meter squared, however in follow it’s usually normalized to 1 or 100 models
with respect to the luminance of a specified or implied white reference.
For instance, a studio broadcast monitor has a white reference whose luminance
is about 100 cd*m ^-2, and Y = 1 refers
to this worth.

Human imaginative and prescient has a nonlinear perceptual response to brightness: a supply
having a luminance solely 18% of a reference luminance seems about half
as brilliant. The perceptual response to luminance known as Lightness.
It’s denoted L* and is outlined by the CIE as a modified dice root
of luminance:

Yn is the luminance of the white reference. When you normalize luminance
to reference white then you needn’t compute the fraction. The CIE definition
applies a linear phase with a slope of 903.3 close to black, for (Y/Yn)
<= 0.008856. The linear phase is unimportant for sensible functions
however in the event you do not use it, just remember to restrict L* at zero. L* has a variety
of 0 to 100, and a “delta L-star” of unity is taken to be roughly
the edge of visibility.

Acknowledged in another way, lightness notion is roughly logarithmic. An observer
can detect an depth distinction between two patches when their intensities
differ by multiple about p.c.

Video methods approximate the lightness response of imaginative and prescient utilizing R’G’B’
alerts which can be every topic to a 0.45 energy operate. That is comparable
to the 1/3 energy operate outlined by L*.

In accordance with the CIE [1], hue is
the attribute of a visible sensation in accordance with which an space seems to
be much like one of many perceived colors, purple, yellow, inexperienced and blue,
or a mixture of two of them. Roughly talking, if the dominant wavelength
of an SPD shifts, the hue of the related color will shift.

Once more from the CIE, saturation is the colourfulness of an space judged
in proportion to its brightness. Saturation runs from impartial grey by means of
pastel to saturated colors. Roughly talking, the extra an SPD is concentrated
at one wavelength, the extra saturated would be the related color. You
can desaturate a color by including gentle that incorporates energy in any respect wavelengths.

The CIE system defines methods to map an SPD to a triple of numerical
elements which can be the mathematical coordinates of color house. Their
operate is analagous to coordinates on a map. Cartographers have completely different
map projections for various features: some map projections protect areas,
others present latitudes and longitudes as straight strains. No single map projection
fills all of the wants of map customers. Equally, no single color system fills
all the wants of color customers.

The methods helpful right this moment for color specification embody CIE XYZ,
CIE xyY, CIE L*u*v* and CIE L*a*b*. Numerical values of
hue and saturation will not be very helpful for color specification, for causes
to be mentioned in section 36.

A color specification system wants to have the ability to signify any color
with excessive precision. Since few colors are dealt with at a time, a specification
system could be computationally complicated. Any system for color specification
have to be intimately associated to the CIE specs.

You may specify a single “spot” color utilizing a color order
system akin to Munsell. Programs like Munsell include swatch books
to allow visible color matches, and have documented strategies of reworking
between coordinates within the system and CIE values. Programs like Munsell are
not helpful for picture information. You may specify an ink color by specifying the
proportions of ordinary (or secret) inks that may be combined to make the color.
That is how PANTONE[tm] works. though widespread,
it is proprietary. No translation to CIE is publicly obtainable.

A digitized color picture is represented as an array of pixels, the place
every pixel incorporates numerical elements that outline a color. Three elements
are crucial and ample for this objective, though in printing it’s
handy to make use of a fourth (black) element.

In idea, the three numerical values for picture coding could possibly be supplied
by a color specification system. However a sensible picture coding system wants
to be computationally environment friendly, can’t afford limitless precision, want
not be intimately associated to the CIE system and customarily must cowl
solely a fairly big selection of colors and never all the colors. So picture
coding makes use of completely different methods than color specification.

The methods helpful for picture coding are linear RGB, nonlinear R’G’B’,
nonlinear CMY, nonlinear CMYK, and derivatives of nonlinear R’G’B’ such
as Y’CBCR. Numerical values of hue and saturation will not be helpful in color
picture coding.

When you manufacture automobiles, you need to match the color of paint on the
door with the color of paint on the fender. A color specification system
shall be crucial. However to convey an image of the automobile, you want picture coding.
You may afford to do fairly a little bit of computation within the first case as a result of
you’ve solely two colored parts, the door and the fender. Within the second
case, the color coding have to be fairly environment friendly as a result of you might have 1,000,000
colored parts or extra.

For a extremely readable brief introduction to color picture coding, see
DeMarsh and Giorgianni [2]. For a terse,
full technical therapy, learn Schreiber [3].

Direct acquisition of luminance requires use of a really particular spectral
weighting. Nevertheless, luminance can be computed as a weighted sum of
purple, inexperienced and blue elements.

If three sources seem purple, inexperienced and blue, and have the identical radiance
within the seen spectrum, then the inexperienced will seem the brightest of the
three as a result of the luminous effectivity operate peaks within the inexperienced area
of the spectrum. The purple will seem much less brilliant, and the blue would be the
darkest of the three. As a consequence of the luminous effectivity operate,
all saturated blue colors are fairly darkish and all saturated yellows are
fairly gentle. If luminance is computed from purple, inexperienced and blue, the coefficients
shall be a operate of the actual purple, inexperienced and blue spectral weighting
features employed, however the inexperienced coefficient shall be fairly giant, the purple
could have an intermediate worth, and the blue coefficient would be the smallest
of the three.

Up to date CRT phosphors are standardized in Rec. 709 [8], to be described in section
17. The weights to compute true CIE luminance from linear purple, inexperienced
and blue (indicated with out prime symbols), for the Rec. 709, are
these:

This computation assumes that the luminance spectral weighting could be
shaped as a linear mixture of the scanner curves, and assumes that the
element alerts signify linear-light. Both or each of those circumstances
could be relaxed to some extent relying on the appliance.

Some pc methods have computed brightness utilizing (R+G+B)/3. This
is at odds with the properties of human imaginative and prescient, as shall be mentioned beneath
What are HSB and HLS? in section 36.

The coefficients 0.299, 0.587 and 0.114 correctly computed luminance for
displays having phosphors that had been up to date on the introduction of
NTSC tv in 1953. They’re nonetheless acceptable for computing video
luma to be mentioned under in section
11. Nevertheless, these coefficients don’t precisely compute luminance
for up to date displays.

Blue has a small contribution to the brightness sensation. Nevertheless, human
imaginative and prescient has terribly good color discrimination functionality in blue
colors. So in the event you give blue fewer bits than purple or inexperienced, you’ll introduce
noticeable contouring in blue areas of your photos.

It’s helpful in a video system to convey a element consultant of
luminance and two different elements consultant of color. It’s important
to convey the element consultant of luminance in such a means that noise
(or quantization) launched in transmission, processing and storage has
a perceptually comparable impact throughout the complete tone scale from black to
white. The best strategy to accomplish these objectives could be to kind a luminance
sign by matrixing RGB, then subjecting luminance to a nonlinear switch
operate much like the L* operate.

There are sensible causes in video to carry out these operations within the
reverse order. First a nonlinear switch operate – gamma correction
– is utilized to every of the linear R, G and B. Then a weighted sum of the
nonlinear elements is computed to kind a sign consultant of luminance.
The ensuing element is said to brightness however just isn’t CIE luminance.
Many video engineers name it luma and provides it the image Y’.
It’s usually carelessly referred to as luminance and given the image
Y. You have to be cautious to find out whether or not a specific creator
assigns a linear or nonlinear interpretation to the time period luminance
and the image Y.

The coefficients that correspond to the “NTSC” purple, inexperienced and
blue CRT phosphors of 1953 are standardized in ITU-R Advice BT.
601-2 (previously CCIR Rec. 601-2). I name it Rec. 601.
To compute nonlinear video luma from nonlinear purple, inexperienced and blue:

The prime symbols on this equation, and in these to observe, denote nonlinear
elements.

The CIE system is predicated on the outline of color as a luminance element
Y, as described above, and two extra elements X and Z. The spectral
weighting curves of X and Z have been standardized by the CIE based mostly on statistics
from experiments involving human observers. XYZ tristimulus values
can describe any color. (RGB tristimulus values shall be described later.)

The magnitudes of the XYZ elements are proportional to bodily power,
however their spectral composition corresponds to the color matching traits
of human imaginative and prescient.

The CIE system is outlined in Publication CIE No 15.2, Colorimetry,
Second Version (1986) [4].

Most likely not. Scanners are most frequently used to scan photos akin to color
images and color offset prints which can be already “information”
of three elements of color data. The same old process of a scanner is
not spectral evaluation however extraction of the values of the three elements
which have already been recorded. Narrowband filters are extra suited to this
process than filters that adhere to the rules of colorimetry.

When you place in your scanner an authentic colored object that has “authentic”
SPDs that aren’t already a file of three elements, likelihood is your
scanner is not going to very report correct RGB values. It’s because most scanners
don’t conform very carefully to CIE requirements.

It’s usually handy to debate “pure” color within the absence
of brightness. The CIE defines a normalization course of to compute “little”
x and y chromaticity coordinates:

A color plots as some extent in an (x, y) chromaticity diagram.
When a narrowband SPD comprising energy at only one wavelength is swept throughout
the vary 400 nm to 700 nm, it traces a shark-fin formed spectral
locus in (x, y) coordinates. The feeling of purple can’t
be produced by a single wavelength: to supply purple requires a mix
of shortwave and longwave gentle. The line of purples on a chromaticity
diagram joins excessive blue to excessive purple. All colors are contained in
the world in (x, y) bounded by the road of purples and the spectral
locus.

A color could be specified by its chromaticity and luminance, within the kind
of an xyY triple. To get better X and Z from chromaticities and luminance,
use these relations:

The bible of color science is Wyszecki and Kinds, Shade Science
[5]. But it surely’s daunting. For Wyszecki’s
personal condensed model, see Shade in Enterprise, Science and Business, Third
Version [6]. It’s directed to the
color {industry}: ink, paint and the like. For an approachable introduction
to the identical idea, accompanied by descriptions of picture copy, learn
R.W.G. Hunt, The Replica of Color [7].

In additive picture copy, the white level is the chromaticity
of the color reproduced by equal purple, inexperienced and blue elements. White
level is a operate of the ratio (or stability) of energy among the many
primaries. In subtractive copy, white is the SPD of the illumination,
multiplied by the SPD of the media. There isn’t any distinctive bodily or perceptual
definition of white, so to realize correct color interchange you will need to
specify the traits of your white.

It’s usually handy for functions of calculation to outline white as
a uniform SPD. This white reference is called the equal-energy illuminant,
or CIE Illuminant E.

A extra real looking reference that approximates daylight has been specified
numerically by the CIE as Illuminant D65. It’s best to use this until you
have a superb motive to make use of one thing else. The print {industry} generally makes use of
D50 and images generally makes use of D55. These signify compromises between
the circumstances of indoor (tungsten) and daylight viewing.

Planck decided that the SPD radiated from a scorching object – a black
physique radiator – is a operate of the temperature to which the article
is heated. Many sources of illumination have, at their core, a heated object,
so it’s usually helpful to characterize an illuminant by specifying the temperature
(in models of kelvin, Okay) of a black physique radiator that seems to have the
identical hue.

Though an illuminant could be specified informally by its color temperature,
a extra full specification is supplied by the chromaticity coordinates
of the SPD of the supply.

Trendy blue CRT phosphors are extra environment friendly with respect to human imaginative and prescient
than purple or inexperienced. In a quest for brightness on the expense of color accuracy,
it is not uncommon for a pc show to have extreme blue content material, about
twice as blue as daylight, with white at about 9300 Okay.

Human imaginative and prescient adapts to white within the viewing setting. A picture considered
in isolation – akin to a slide projected in a darkish room – creates its personal
white reference, and a viewer shall be fairly tolerant of errors within the white
level. But when the identical picture is considered within the presence of an exterior white
reference or a second picture, then variations in white level could be objectionable.

Full adaptation appears to be confined to the vary 5000 Okay to
5500 Okay. For most individuals, D65 has just a little trace of blue. Tungsten illumination,
at about 3200 Okay, all the time seems considerably yellow.

Additive copy is predicated on bodily units that produce all-positive
SPDs for every main. Bodily and mathematically, the spectra add. The
largest vary of colors shall be produced with primaries that seem purple,
inexperienced and blue. Human color imaginative and prescient obeys the precept of superposition,
so the color produced by any additive combination of three main spectra
could be predicted by including the corresponding fractions of the XYZ elements
of the primaries: the colors that may be combined from a specific set of
RGB primaries are utterly decided by the colors of the primaries
by themselves. Subtractive copy is rather more sophisticated: the colors
of mixtures are decided by the primaries and by the colors of their
combos.

An additive RGB system is specified by the chromaticities of its primaries
and its white level. The extent (gamut) of the colors that may be
combined from a given set of RGB primaries is given within the (x, y) chromaticity
diagram by a triangle whose vertices are the chromaticities of the primaries.

In computing there aren’t any normal primaries or white level. When you’ve got
an RGB picture however haven’t any details about its chromaticities, you can not
precisely reproduce the picture.

The NTSC in 1953 specified a set of primaries that had been consultant
of phosphors utilized in color CRTs of that period. However phosphors modified over
the years, primarily in response to market pressures for brighter receivers,
and by the point of the primary the videotape recorder the primaries in use
had been fairly completely different than these “on the books”. So though you
may even see the NTSC main chromaticities documented, they’re of no use right this moment.

Up to date studio displays have barely completely different requirements in North
America, Europe and Japan. However worldwide settlement has been obtained
on primaries for prime definition tv (HDTV), and these primaries
are carefully consultant of latest displays in studio video, computing
and pc graphics. The primaries and the D65 white level of Rec.
709 [8] are:

       R           G           B           white       
x      0.640       0.300       0.150       0.3127      
y      0.330       0.600       0.060       0.3290      
z      0.030       0.100       0.790       0.3582

For a dialogue of nonlinear RGB in pc graphics, see Lindbloom
[9]. For technical particulars on monitor
calibration, seek the advice of Cowan [10].

RGB values in a specific set of primaries could be reworked to and
from CIE XYZ by a three-by-three matrix rework. These transforms
contain tristimulus values, that’s, units of three linear-light
elements that conform to the CIE color matching features. CIE
XYZ is a particular case of tristimulus values. In XYZ, any color is represented
by a optimistic set of values.

Particulars could be present in SMPTE RP 177-1993
[11].

To remodel from CIE XYZ into Rec. 709 RGB (with its D65
white level), use this rework:

This matrix has some unfavorable coefficients: XYZ colors which can be out
of gamut for a specific RGB rework to RGB the place a number of RGB
elements is unfavorable or higher than unity.

Here is the inverse rework. As a result of white is normalized to unity, the
center row sums to unity:

To get better main chromaticities from such a matrix, compute little
x and y for every RGB column vector. To get better the white level,
rework RGB=[1, 1, 1] to XYZ, then compute x and y.

Video requirements specify summary R’G’B’ methods which can be carefully matched
to the traits of actual displays. Bodily units that produce additive
color contain tolerances and uncertainties, however when you have a monitor that
conforms to Rec. 709 inside some tolerance, you possibly can think about the monitor
to be device-independent.

The significance of Rec. 709 as an interchange normal in studio
video, broadcast tv and excessive definition tv, and the perceptual
foundation of the usual, assures that its parameters shall be used even by
units akin to flat-panel shows that do not need the identical physics as
CRTs.

RGB values in a system using one set of primaries could be reworked
into one other set by a three-by-three linear-light matrix rework. Typically
these matrices are normalized for a white level luminance of unity. For
particulars, see Tv Engineering Handbook [12].

For instance, right here is the rework from SMPTE
240M (or SMPTE RP 145) RGB to Rec. 709:

All of those phrases are near both zero or one. In a case like this,
if the rework is computed within the nonlinear (gamma-corrected) R’G’B’ area
the ensuing errors shall be insignificant.

Here is one other instance. To remodel EBU 3213 RGB to Rec.
709:

Reworking amongst RGB methods might result in an out of gamut RGB
consequence the place a number of RGB elements is unfavorable or higher than unity.

As soon as gentle is on its strategy to the attention, any tristimulus-based system will
work. However the interplay of sunshine and objects entails spectra, not tristimulus
values. In artificial pc graphics, the calculations are literally simulating
sampled SPDs, even when solely three elements are used. Particulars regarding
the resultant errors are present in Corridor [13].

Subtractive methods contain colored dyes or filters that take in energy
from chosen areas of the spectrum. The three filters are positioned in tandem.
A dye that seems cyan absobs longwave (purple) gentle. By controlling the
quantity of cyan dye (or ink), you modulate the quantity of purple within the picture.

In bodily phrases the spectral transmission curves of the colourants
multiply, so this technique of color copy ought to actually be referred to as
“multiplicative”. Photographers and printers have for many years
measured transmission in base-10 logarithmic density models, the place
transmission of unity corresponds to a density of 0, transmission of 0.1
corresponds to a density of 1, transmission of 0.01 corresponds to a density
of two and so forth. When a printer or photographer computes the impact of filters
in tandem, he subtracts density values as an alternative of multiplying transmission
values, so he calls the system subtractive.

To realize a variety of colors in a subtractive system requires filters
that seem colored cyan, yellow and magenta (CMY). Cyan in tandem with
magenta produces blue, cyan with yellow produces inexperienced, and magenta with
yellow produces purple. Smadar Nehab suggests this reminiscence support:

Additive primaries are on the high, subtractive on the backside. On the
left, magenta and yellow filters mix to supply purple. On the correct, purple
and inexperienced sources add to supply yellow.

To get a variety of colors in an additive system, the primaries should
seem purple, inexperienced and blue (RGB). In a subtractive system the primaries
should seem yellow, cyan and magenta (CMY). It’s sophisticated to foretell
the colors produced when mixing paints, however roughly talking, paints combine
additively to the extent that they’re opaque (like oil paints), and subtractively
to the extent that they’re clear (like watercolours). This query
additionally pertains to color names: your grade three “purple” was most likely
just a little on the magenta facet, and “blue” was most likely fairly cyan.
For a dialogue of paint mixing from a pc graphics perspective, seek the advice of
Haase [14].

In a theoretical subtractive system, CMY filters might have spectral
absorption curves with no overlap. The color copy of the system
would correspond precisely to additive color copy utilizing the purple,
inexperienced and blue primaries that resulted from pairs of filters together.

Sensible photographic dyes and offset printing inks have spectral absorption
curves that overlap considerably. Most magenta dyes take in mediumwave (inexperienced)
gentle as anticipated, however by the way take in about half that quantity of shortwave
(blue) gentle. If copy of a color, say brown, requires absorption
of all shortwave gentle then the incidental absorption from the magenta dye
just isn’t seen. However for different colors, the “one minus RGB” method
produces mixtures with a lot much less blue than anticipated, and subsequently produce
photos which have a yellow solid within the mid tones. Comparable however much less extreme
interactions are evident for the opposite pairs of sensible inks and dyes.

As a result of spectral overlap among the many colourants, changing CMY utilizing
the “one-minus-RGB” technique works for functions akin to enterprise
graphics the place correct color needn’t be preserved, however the technique fails
to supply acceptable color photos.

Multiplicative combination in a CMY system is mathematically nonlinear, and
the impact of the undesirable absorptions can’t be simply analyzed or compensated.
The colors that may be combined from a specific set of CMY primaries can’t
be decided from the colors of the primaries themselves, however are additionally
a operate of the colors of the units of combos of the primaries.

Print and photographic copy can also be sophisticated by nonlinearities
within the response of the three (or 4) channels. In offset printing, the
bodily and optical processes of dot acquire introduce nonlinearity
that’s roughly corresponding to gamma correction in video. In a typical system
used for print, a black code of 128 (on a scale of 0 to 255) produces a
reflectance of about 0.26, not the 0.5 that you’d anticipate from a linear
system. Computations can’t be meaningfully carried out on CMY elements
with out taking nonlinearity into consideration.

For an in depth dialogue of transferring colorimetric picture information to
print media, see Stone [15].

Printing black by overlaying cyan, yellow and magenta ink in offset printing
has three main issues. First, colored ink is dear. Changing colored
ink by black ink – which is primarily carbon – makes financial sense. Second,
printing three ink layers causes the printed paper to develop into fairly moist.
If three inks could be changed by one, the ink will dry extra rapidly, the
press could be run sooner, and the job shall be inexpensive. Third, if black
is printed by combining three inks, and mechanical tolerances trigger the
three inks to be printed barely out of register, then black edges will
undergo colored tinges. Imaginative and prescient is most demanding of spatial element in black
and white areas. Printing black with a single ink minimizes the visibility
of registration errors.

Different printing processes might or might not be topic to comparable constraints.

This time period is ambiguous. In its first sense, color distinction
refers to numerical variations between color specs. The notion
of color variations in XYZ or RGB is extremely nonuniform. The examine of perceptual
uniformity considerations numerical variations that correspond to color
variations on the threshold of perceptibility (simply noticeable variations,
or JNDs).

In its second sense, color distinction refers to color elements
the place brightness is “eliminated”. Imaginative and prescient has poor response to spatial
element in colored areas of the identical luminance, in comparison with its response
to luminance spatial element. If information capability is at a premium it’s advantageous
to transmit luminance with full element and to kind two color distinction
elements every having no contribution from luminance. The 2 color elements
can then have spatial element eliminated by filtering, and could be transmitted
with considerably much less data capability than luminance.

As an alternative of utilizing a real luminance element to signify brightness,
it’s ubiquitous for sensible causes to make use of a luma sign that
is computed nonlinearly as outlined above ( What’s luma? ).

The best strategy to “take away” brightness data to kind
two color channels is to subtract it. The luma element already incorporates
a big fraction of the inexperienced data from the picture, so it’s normal
to kind the opposite two elements by subtracting luma from nonlinear blue
(to kind B’-Y’ ) and by subtracting luma from nonlinear purple (to kind
R’-Y’ ). These are referred to as chroma.

Numerous scale elements are utilized to (B’-Y’ ) and (R’-Y’ )
for various functions. The Y’PBPR scale elements are optimized for
element analog video. The Y’CBCR scaling is acceptable for element
digital video akin to studio video and MPEG. (JPEG is described within the
literature as utilizing Y’CBCR; nonetheless, no footroom or headroom are used
for both luma or chroma. Luma is scaled to an tour of 255. CB and CR
run -128 to +127; code +128 is clipped.) Kodak’s PhotoYCC[tm] makes use of scale elements optimized for the gamut of movie colors. Y’UV scaling
is acceptable as an intermediate step within the formation of composite NTSC
or PAL video alerts, however just isn’t acceptable when the elements are stored
separate. The Y’UV nomenclature is now used quite loosely, and it typically
denotes any scaling of (B’-Y’ ) and (R’-Y’ ). Y’IQ coding
is out of date.

The subscripts in CBCR and PBPR are sometimes written in decrease case. I discover
this to compromise readability, so with out introducing any ambiguity I write
them in uppercase. Authors with nice consideration to element typically “prime”
these portions to point their nonlinear nature, however as a result of no sensible
picture coding system employs linear color variations I think about it protected
to omit the primes.

Right here is the block diagram for luma/color distinction encoding and decoding:

From linear XYZ – or linear R1 G1 B1 whose chromaticity coordinates are
completely different from the interchange normal – apply a 3 x 3 matrix
rework to acquire linear RGB in accordance with the interchange primaries. Apply
a nonlinear switch operate (“gamma correction”) to every of
the elements to get nonlinear R’G’B’. Apply a 3 x 3 matrix to
receive color distinction elements akin to Y’PBPR , Y’CBCR or PhotoYCC.
If crucial, apply a color subsampling filter to acquire subsampled color
distinction elements. To decode, invert the above process: run by means of
the block diagram right-to-left utilizing the inverse operations. In case your monitor
conforms to the interchange primaries, decoding needn’t explicitly use
a switch operate or the tristimulus 3 x 3.

The block diagram emphasizes that 3 x 3 matrix transforms are
used for 2 distinctly completely different duties. When somebody palms you a 3 x 3,
you need to ask for which process it’s meant.

Though the next matrices might in idea be used for tristimulus
alerts, it’s ubiquitous to make use of them with gamma-corrected alerts.

To encode Y’PBPR , begin with the essential Y’, (B’-Y’ ) and (R’-Y’
) relationships:

Eq 1

Y’PBPR elements have unity tour, the place Y’ ranges [0..+1] and every
of PB and PR ranges [-0.5..+0.5]. The (B’-Y’ ) and (R’-Y’ )
rows must be scaled by
and . To encode from R’G’B’
the place reference black is 0 and reference white is +1:

Eq 2

The primary row includes the luma coefficients; these sum to unity. The
second and third rows every sum to zero, a necessity for color distinction
elements. The +0.5 entries replicate the utmost tour of PB and PR
of +0.5, for the blue and purple primaries [0, 0, 1] and [1,
0, 0].

The inverse, decoding matrix is that this:

Rec. 601 specifies eight-bit coding the place Y’ has an tour of
219 and an offset of +16. This coding locations black at code 16 and white
at code 235, reserving the extremes of the vary for sign processing headroom
and footroom. CB and CR have excursions of +/-112 and offset of +128,
for a variety of 16 by means of 240 inclusive.

To compute Y’CBCR from R’G’B’ within the vary [0..+1], scale the rows of
the matrix of Eq 2 by the elements 219, 224 and 224, comparable to the
excursions of every of the elements:

Eq 3

Summing the primary row of the matrix yields 219, the luma tour from
black to white. The 2 entries of 112 replicate the optimistic CBCR extrema
of the blue and purple primaries.

Clamp all three elements to the vary 1 by means of 254 inclusive, since
Rec. 601 reserves codes 0 and 255 for synchronization alerts.

To get better R’G’B’ within the vary [0..+1] from Y’CBCR, use the inverse of
Eq 3 above:

This seems overwhelming, however the Y’CBCR elements are integers in eight
bits and the reconstructed R’G’B’ are scaled all the way down to the vary [0..+1].

In computing it’s typical to make use of eight-bit coding with black at
code 0 and white at 255. To encode Y’CBCR from R’G’B’ within the vary [0..255],
utilizing eight-bit binary arithmetic, scale the Y’CBCR matrix of Eq 3 by 256/255:

To decode R’G’B’ within the vary [0..255] from Rec. 601 Y’CBCR, utilizing
eight-bit binary arithmetic:

Eq 4

The multiplications by 1/256 could be achieved by shifting. A few of
the coefficients, when scaled by 1/256, are bigger than unity. These coefficients
will want greater than eight multiplier bits.

For implementation in binary arithmetic the matrix coefficients have
to be rounded. Whenever you spherical, take care to protect the row sums of [1,
0, 0].

The matrix of Eq 4 will decode normal Y’CBCR elements to RGB elements
within the vary [0..255], topic to roundoff error. You will need to take care to
keep away from overflow because of roundoff error. However you will need to defend in opposition to overflow
in any case, as a result of studio video alerts use the extremes of the coding
vary to deal with sign overshoot and undershoot, and these would require
clipping when decoded to an RGB vary that has no headroom or footroom.

Studio R’G’B’ alerts use the identical 219 tour because the luma element
of Y’CBCR.To encode Y’CBCR from R’G’B’ within the vary [0..219], utilizing eight-bit
binary arithmetic, scale the Y’CBCR encoding matrix of Eq 3 above by 256/219.
Right here is the encoding rework for studio video:

To decode R’G’B’ within the vary [0..219] from Y’CBCR, utilizing eight-bit binary
arithmetic:

The entries of 256 on this matrix point out that the corresponding element
can merely be added; there isn’t any want for a multiplication operation. This
matrix incorporates entries bigger than 256; the corresponding multipliers will
want functionality for 9 bits.

The matrices on this part conform to Rec. 601 and apply straight
to standard 525/59.94 and 625/50 video. It’s not but determined whether or not
rising HDTV requirements will use the identical matrices, or undertake a brand new set of
matrices having completely different luma coefficients. In my opinion it could be unlucky
if completely different matrices had been adopted, as a result of then picture coding and decoding
would rely upon whether or not the image was small (typical video) or giant
(HDTV).

In digital video, Rec. 601 standardizes subsampling denoted 4:2:2,
the place CB and CR elements are subsampled horizontally by an element of two
with respect to luma. JPEG and MPEG conventionally subsample by an element
of two within the vertical dimension as properly, denoted 4:2:0.

Color distinction coding is standardized in Rec. 601. For particulars
on color distinction coding as utilized in video, seek the advice of Watkinson [16].

Kodak’s PhotoYCC makes use of the Rec. 709 primaries, white level and switch
operate. Reference white codes to luma 189; this preserves movie highlights.
The color distinction coding is asymmetrical, to embody movie gamut. You
are unlikely to come across any uncooked picture information in PhotoYCC kind as a result of YCC
is carefully related to the PhotoCD[tm] system whose compression strategies
are proprietary. However simply in case, the next equation is comparable
to in that it produces R’G’B’ within the vary [0..+1] from integer YCC. If
you need to return R’G’B’ in a distinct vary, or implement the equation
in eight-bit integer arithmetic, use the methods within the part above.

Decoded R’G’B’ elements from PhotoYCC can exceed unity or go under
zero. PhotoYCC extends the Rec. 709 switch operate above unity,
and displays it round zero, to accommodate vast excursions of R’G’B’. To
decode to CRT primaries, clip R’G’B’ to the vary zero to 1.

No, I will not! Y’ UV and Y’ IQ have scale elements acceptable to composite
NTSC and PAL. They haven’t any place in element digital video! You should not
code into these methods, and if somebody palms you a picture claiming it is
Y’ UV, likelihood is it is truly Y’CBCR, it is received the mistaken scale elements,
or it is linear-light.

Nicely OK, simply this as soon as. To remodel Y’, (B’-Y’) and (R’-Y’) elements
from Eq 1 to Y’UV, scale (B’-Y’) by 0.492111 to get U and scale R’-Y’
by 0.877283 to get V. The elements are chosen to restrict composite NTSC or
PAL amplitude for all authorized R’G’B’ values:

To remodel to Y’ IQ to Y’ UV, carry out a 33 diploma rotation and an trade
of color distinction axes:

To check your encoding and decoding, make sure that colourbars are dealt with
accurately. A colourbar sign includes a binary RGB sequence ordered for
lowering luma: white, yellow, cyan, inexperienced, magenta, purple, blue and black.

To make sure that your scale elements are appropriate and that clipping just isn’t
being invoked, take a look at 75% bars, a colourbar sequence having 75%-amplitude
bars as an alternative of 100%.

A system is perceptually uniform if a small perturbation to a
element worth is roughly equally perceptible throughout the vary of
that worth. The amount management in your radio is designed to be perceptually
uniform: rotating the knob ten levels produces roughly the identical perceptual
increment in quantity wherever throughout the vary of the management. If the management
had been bodily linear, the logarithmic nature of human loudness notion
would place all the perceptual “motion” of the management on the
backside of its vary.

The XYZ and RGB methods are removed from exhibiting perceptual uniformity.
Discovering a change of XYZ into a fairly perceptually-uniform house
consumed a decade or extra on the CIE and ultimately no single system might
be agreed. So the CIE standardized two methods, L*u*v* and L*a*b*,
typically written CIELUV and CIELAB.
(The u and v are unrelated to video U and V.) Each L*u*v* and L*a*b*
enhance the 80:1 or so perceptual nonuniformity of XYZ to about 6:1. Each
demand an excessive amount of computation to accommodate real-time show, though each
have been efficiently utilized to picture coding for printing.

Computation of CIE L*u*v* entails intermediate u’ and v ‘ portions,
the place the prime denotes the successor to the out of date 1960 CIE u and v system:

First compute un’ and vn’ on your reference white Xn , Yn and Zn . Then
compute u’ and v ` – and L* as mentioned earlier – on your colors. Lastly,
compute:

L*a*b* is computed as follows, for (X/Xn, Y/Yn, Z/Zn ) >
0.01:

These equations are nice for a couple of spot colors, however no enjoyable for 1,000,000
pixels. Though it was not particularly optimized for this objective, the
nonlinear R’G’B’ coding utilized in video is sort of perceptually uniform, and
has the benefit of being quick sufficient for interactive functions.

HSB and HLS had been developed to specify numerical Hue, Saturation and Brightness
(or Hue, Lightness and Saturation) in an age when customers needed to specify colors
numerically. The same old formulations of HSB and HLS are flawed with respect
to the properties of color imaginative and prescient. Now that customers can select colors visually,
or select colors associated to different media (akin to PANTONE),
or use perceptually-based methods like L*u*v* and L*a*b*, HSB and HLS ought to
be deserted.

Listed here are a few of issues of HSB and HLS. In color choice the place “lightness”
runs from zero to 100, a lightness of fifty ought to look like half as brilliant
as a lightness of 100. However the standard formulations of HSB and HLS make no
reference to the linearity or nonlinearity of the underlying RGB, and make
no reference to the lightness notion of human imaginative and prescient.

The same old formulation of HSB and HLS compute so-called “lightness”
or “brightness” as (R+G+B)/3. This computation conflicts badly
with the properties of color imaginative and prescient, because it computes yellow to be about
six occasions extra intense than blue with the identical “lightness” worth
(say L=50).

HSB and HSL will not be helpful for picture computation due to the discontinuity
of hue at 360 levels. You can’t carry out arithmetic mixtures of colors
expressed in polar coordinates.

Practically all formulations of HSB and HLS contain completely different computations
round 60 diploma segments of the hue circle. These calculations introduce
seen discontinuities in color house.

Though the declare is made that HSB and HLS are “system impartial”,
the ever present formulations are based mostly on RGB elements whose chromaticities
and white level are unspecified. Consequently, HSB and HLS are ineffective for
conveyance of correct color data.

If you actually need to specify hue and saturation by numerical values,
quite than HSB and HSL you need to use polar coordinate model of u* and
v*: h*uv for hue angle and c*uv for chroma.

True color is the supply of three separate elements for
additive purple, inexperienced and blue copy. True color methods usually present
eight bits for every of the three elements, so true color is typically
known as 24-bit color.

A real color system normally interposes a lookup desk between every element
of the framestore and every channel to the show. This makes it attainable
to make use of a real color system with both linear or nonlinear coding. Within the
X Window System, true color refers to fastened lookup tables, and direct
color refers to lookup tables which can be beneath the management of software
software program.

Listed color (or pseudocolour), is the supply of a
comparatively small quantity, say 256, of discrete colors in a colormap
or palette. The framebuffer shops, at every pixel, the index quantity
of a color. On the output of the framebuffer, a lookup desk makes use of the index
to retrieve purple, inexperienced and blue elements which can be then despatched to the show.

The colors within the map could also be fastened systematically on the design of a
system. For instance, 216 index entries an eight-bit listed color system
could be partitioned systematically right into a 6 x 6 x 6 “dice”
to implement what quantities to a direct color system the place every of purple, inexperienced
and blue has a worth that’s an integer within the vary zero to 5.

An RGB picture could be transformed to a predetermined colormap by selecting,
for every pixel within the picture, the colormap index comparable to the “closest”
RGB triple. With a scientific colormap akin to a 6 x 6 x 6
colourcube that is easy. For an arbitrary colormap, the colormap
needs to be searched in search of entries which can be “shut” to the
requested color. “Closeness” must be decided in accordance with
the perceptibility of color variations. Utilizing color methods akin to CIE
L*u*v* or L*a*b* is computationally prohibitive, however in follow it’s sufficient
to make use of a Euclidean distance metric in R’G’B’ elements coded nonlinearly
in accordance with video follow.

A direct color picture could be transformed to listed color with an image-dependent
colormap by a means of color quantization that searches by means of all of
the triples used within the picture, and chooses the palette for the picture based mostly
on the colors which can be in some sense most “vital”. Once more,
the selections must be made in accordance with the perceptibility of color variations.
Adobe Photoshop[tm] can carry out this conversion. UNIX[tm] customers can make use of
the pbm package deal.

In case your system accommodates arbitrary colormaps, when the map related
with the picture in a specific window is loaded into the {hardware} colormap,
the maps related to different home windows could also be disturbed. In window system
such because the X Window System[tm] operating on a multitasking working
system akin to UNIX, even shifting the cursor between two home windows with completely different
maps could cause annoying colormap flashing.

An eight-bit listed color system requires much less information to signify a
image than a twenty-four bit truecolour system. However this information discount
comes at a excessive value. The truecolour system can signify every of its three
elements in accordance with the rules of sampled steady alerts. This
makes it attainable to perform, with good high quality, operations akin to resizing
the picture. In listed color these operations introduce extreme artifacts
as a result of the underlying illustration lacks the properties of a steady
illustration, even when transformed again to RGB.

In graphic file codecs akin to GIF of TIFF, an listed color picture
is accompanied by its colormap. Typically such a colormap has RGB entries
which can be gamma corrected: the colormap’s RGB codes are meant to be offered
on to a CRT, with out additional gamma correction.

Whenever you take a look at a rainbow you don’t see a clean gradation of colors.
As an alternative, some bands seem fairly slender, and others are fairly broad. Perceptibility
of hue variation close to 540 nm is half that of both 500 nm or
600 nm. When you use the rainbow’s colors to signify information, the visibility
of variations amongst your information values will rely upon the place they lie within the
spectrum.

In case you are utilizing color to assist within the visible detection of patterns, you
ought to use colors chosen in accordance with the rules of perceptual uniformity.
This an open analysis downside, however basing your system on CIE L*a*b* or L*u*v*,
or on nonlinear video-like RGB, could be a superb begin.

A show system might have solely a small variety of decisions of greyscale
values or color values at every system pixel. Nevertheless if the viewer is sufficiently
distant from the show, the worth of neighboring pixels could be set so
that the viewer’s eye integrates a number of pixels to realize an obvious enchancment
within the variety of ranges or colors that may be reproduced.

Pc shows are typically considered from distances the place the system
pixels subtend a quite giant angle on the viewer’s eye, relative to his
visible acuity. Making use of dither to a traditional pc show usually
introduces objectionable artifacts. Nevertheless, cautious software of dither
could be efficient. For instance, human imaginative and prescient has poor acuity for blue spatial
element however good color discrimination functionality in blue. Blue could be dithered
throughout two-by-two pixel arrays to supply 4 occasions the variety of blue
ranges, with no perceptible penalty at regular viewing distances.

The processes of offset printing and traditional laser printing are
intrinsically bilevel: a specific location on the web page is both
lined with ink or not. Nevertheless, every of those units can reproduce closely-spaced
dots of variable measurement. An array of small dots produces the notion of
gentle grey, and an array of huge dots produces darkish grey. This course of
known as halftoning or screening. In a way that is dithering,
however with system dots so small that acceptable photos could be produced at
cheap viewing distances.

Halftone dots are normally positioned in an everyday grid, though stochastic
screening has not too long ago been launched that modulates the spacing of
the dots quite than their measurement.

In color printing it’s typical to make use of cyan, magenta, yellow and
black grids which have precisely the identical dot pitch however completely different carefully-chosen
display screen angles. The not too long ago launched strategy of Flamenco screening
makes use of the identical display screen angles for all screens, however its registration necessities
are extra stringent than typical offset printing.

Agfa’s booklet [17] is a wonderful
introduction to sensible considerations of printing. And it is in color! The
normal reference to halftoning algorithms is Ulichney [18], however that work doesn’t element the nonlinearities
present in sensible printing methods. For particulars about screening for color
copy, seek the advice of Fink [19]. Seek the advice of
Incessantly Requested Questions on Gamma for an introduction to the
switch operate of offset printing.

Software program and {hardware} for scanner, monitor and printer calibration have
had restricted success in coping with the inaccuracies of color dealing with
in desktop computing. These options cope with particular pairs of units
however can’t handle the end-to-end system. Sure software builders
have added color transformation functionality to their functions, however the
majority of software builders have inadequate experience and inadequate
assets to put money into correct color.

A color administration system (CMS) is a layer of software program resident
on a pc that negotiates color copy between the appliance
and color units. It cooperates with the working system and the graphics
library elements of the platform software program. Color administration methods carry out
the color transformations essential to trade correct color between
numerous units, in varied color coding methods together with RGB, CMYK and
CIE L*a*b*.

The CMS makes obtainable to the appliance a set of services whereby
the appliance can decide what color units and what color areas
can be found. When the appliance needs to entry a specific system,
it requests that the color supervisor carry out a mathematical rework
from one house to a different. The color areas concerned could be device-independent
summary color areas akin to CIE XYZ, CIE L*a*b* or
calibrated RGB. Alternatively a color house could be related to a specific
system. Within the second case the Color supervisor wants entry to characterization
information for the system, and maybe additionally to calibration information that displays
the state of the actual occasion of the system.

Refined color administration methods are commercially obtainable from
Kodak, Electronics for Imaging (EFI) and Agfa. Apple’s ColorSync[tm] offers
an interface between a Mac software program and color administration capabilities
both built-in to ColorSync or supplied by a plug-in. Solar has introduced
that Kodak’s CMS shall be shipped with the following model of Solaris.

The fundamental CMS companies supplied with desktop working methods are seemingly
to be sufficient for workplace customers, however are unlikely to fulfill high-end customers
akin to in prepress. The entire introduced methods have provisions for plug-in
color administration modules (CMMs) that may present refined rework
equipment. Superior color administration modules shall be commercially obtainable
from third events.

A CMS wants entry to data that characterizes the color copy
capabilities of explicit units. The set of characterization information for
a tool known as a system profile. Business settlement has been
reached on the format of system profiles, though particulars haven’t but
been publicly disseminated. Apple has introduced that the forthcoming ColorSync
model 2.0 will adhere to this settlement. Distributors of color peripherals
will quickly present industry-standard profiles with their units, they usually
must make, purchase or lease characterization companies.

When you’ve got a tool that has not been characterised by its producer,
Agfa’s FotoTune[tm] software program – a part of Agfa’s FotoFlow[tm] color supervisor
– can create system profiles.

Not but. However color administration system interfaces sooner or later are seemingly
to incorporate the flexibility to accommodate business proprietary color specification
methods akin to PANTONE[tm] and COLORCURVE[tm].
These distributors are seemingly to supply their color specification methods in
shrink-wrapped kind to plug into color managers. On this means, customers will
have assured color accuracy amongst functions and peripherals, and
software distributors will now not must pay to license these methods
individually.

Use the CIE D65 white level (6504 Okay) in the event you can.

Use the Rec. 709 main chromaticities. Your monitor might be
already fairly near this. Rec. 709 has worldwide settlement,
affords wonderful efficiency, and is the premise for HDTV improvement so it is
future-proof.

If you should function in linear gentle, so be it. In any other case, for finest
perceptual efficiency and most ease of interchange with digital video,
use the Rec. 709 switch operate, with its 0.45-power legislation. When you
want Mac compatibility you’ll have to undergo a penalty in perceptual efficiency.
Elevate tristimulus values to the 1/1.8-power earlier than presenting them to QuickDraw.

To code luma, use the Rec. 601 luma coefficients 0.299, 0.587 and
0.114. Use Rec. 601 digital video coding with black at 16 and white
at 235.

Use prime symbols ( ‘ ) to indicate your entire nonlinear elements!

PhotoCD makes use of all the previous measures. PhotoCD codes color variations
asymmetrically, in accordance with movie gamut. Except you’ve a requirement for
movie gamut, you need to code into color variations utilizing Y ` CBCR coding
with Rec. 601 studio video (16..235/128+/-112) tour.

Tag your picture information with the first and white chromaticity, switch
operate and luma coefficients that you’re utilizing. TIFF 6.0 tags have
been outlined for these parameters. It will allow clever readers,
right this moment or sooner or later, to find out the parameters of your coded picture
and provide the very best outcomes.

[1] Publication CIE No 17.4, International
Lighting Vocabulary. Central Bureau of the Fee Internationale
de L’Éclairage, Vienna, Austria.

[2] LeRoy E. DeMarsh and Edward J. Giorgianni,
“Shade Science for Imaging Programs”, Physics At this time, September
1989, 44-52.

[3] W.F. Schreiber, Fundamentals of Digital
Imaging Programs, Second Version (Springer-Verlag, 1991).

[4] Publication CIE No 15.2, Colorimetry,
Second Version (1986), Central Bureau of the Fee Internationale
de L’Éclairage, Vienna, Austria.

[5] Günter Wyszecki and W.S. Kinds, Shade
Science: Ideas and Strategies, Quantitative Knowledge and Formulae, Second Version
(John Wiley & Sons, New York, 1982).

[6] D.B. Judd and Günter Wyszecki, Shade
in Enterprise, Science and Business, Third Version (John Wiley, New York,
1975).

[7] R.W.G. Hunt, The Replica of Color
in Images, Printing and Tv, Fifth Version (Fountain
Press, Tolworth, England, 1995).

[8] ITU-R Advice BT.709, Primary
Parameter Values for the HDTV Normal for the Studio and for Worldwide
Programme Change (1990), [formerly CCIR Rec. 709], ITU, 1211
Geneva 20, Switzerland.

[9] Bruce J. Lindbloom, “Correct Shade
Replica for Pc Graphics Functions”, Pc Graphics,
Vol. 23, No. 3 (July 1989), 117-126 (proceedings
of SIGGRAPH ’89).

[10] William B. Cowan, “An Cheap
Scheme for Calibration of a Color Monitor when it comes to CIE Normal Coordinates”,
Pc Graphics, Vol. 17, No. 3 (July
1983), 315-321.

[11] SMPTE RP 177-1993,
Derivation of Primary Tv Shade Equations.

[12] Tv Engineering Handbook, That includes
HDTV Programs, Revised Version by Okay. Blair Benson, revised by Jerry C.
Whitaker (McGraw-Hill, 1992). This supersedes the Second Version.

[13] Roy Corridor, Illumination and Shade in
Pc Generated Imagery (Springer-Verlag, 1989).

[14] Chet S. Haase and Gary W. Meyer, “Modelling
Pigmented Supplies for Life like Picture Synthesis”, ACM Transactions
on Graphics, Vol. 11, No. 4, 1992, p. 305.

[15] Maureen C. Stone, William B. Cowan and
John C. Beatty, “Shade Gamut Mapping and the Printing of Digital Shade
Pictures”, ACM Transactions on Graphics, Vol. 7,
No. 3, October 1988.

[16] John Watkinson, An Introduction to
Digital Video (Focal Press, Sevenoaks, Kent, England, 1994).

[17] Agfa Company, An introduction
to Digital Shade Prepress, Volumes 1 and a pair of (1990), Prepress Training
Assets, P.O. Field 7917 Mt.Prospect, IL 60056-7917. 800-395-7007.

[18] Robert Ulichney, Digital Halftoning
(MIT Press, Cambridge, MA, 1988).

[19] Peter Fink, PostScript Screening:
Adobe Correct Screens (Adobe Press, 1992).

Due to Norbert Gerfelder, Alan Roberts, and Fred Remley for his or her proofreading
and modifying. I discovered about color from LeRoy DeMarsh, Ed Giorgianni, Junji
Kumada, and Invoice Cowan. Thanks!

This observe incorporates some errors: in the event you discover any, please let me know.

I welcome recommendations for additions and enhancements.

Poynton –
Color page
2006-11-28a

Source Link