CSIT 402

GROUP ONE
COLOR IN IMAGE AND VIDEO
DETAILED WRITE-UP

INTRODUCTION
Color is an integral aspect of visual perception and image representation. This report delves into
various facets of color, covering sub-topics such as light and spectra, human vision, spectral
sensitivity of the eye, image formation, camera systems, gamma correction, color-matching
functions, CIE Chromaticity Diagram, color monitor specifications, out-of-gamut colors, white
point correction, color transformations, color models, and more.
To become aware of the simple, yet strangely involved world of color, this report shall consider
the following topics:
 Color Science
 Color Models in Images
 Color Models in Video

Color Science: Color science is an interdisciplinary field that studies how humans perceive and
interpret colors, the physics of color generation, and the techniques used to represent colors
digitally. It involves the understanding of color perception, color models, and color reproduction.
Light & Spectra: Light is electromagnetic radiation that our eyes can perceive. Spectra refers to
the distribution of light energy across different wavelengths. The visible spectrum consists of
colors from violet to red, with each color corresponding to a different wavelength. Visible light,
an electromagnetic wave with wavelengths ranging from 400 to 700 nanometers, is analyzed
using Spectral Power Distribution (SPD) or a spectrum. This curve represents the relative power
of light energy at each wavelength, with the symbol for wavelength (lambda) represented by E
(Measurements are used to sum voltage over a small wavelength range, such as 5 or 10 nm,
using segments joining function values every 10 nm.
Human Vision: Human vision involves the process of light entering the eye, being focused by
the lens onto the retina, and then being converted into electrical signals that the brain interprets
as visual information. The eye functions like a camera, focusing images onto the retina through
rods and cones. Rods produce gray images at low light levels, while cones produce signals at
higher light levels. The brain uses differences in red, green, and blue light to create a high- and
light-level achromatic channel, demonstrating its ability to process algebra.
Spectral Sensitivity of the Eye: The human eye's sensitivity to different wavelengths of light
varies. The three types of cone cells in the retina are sensitive to short, medium, and long
wavelengths, which collectively allow us to perceive a wide range of colors. The human eye has
approximately 6 million cones with varying sensitivity functions in red (R), green (G), and blue
(B). These cones have a 40:20:1 distribution, resulting in an achromatic channel that
approximates the sum of their spectral sensitivity functions. These sensitivity functions, denoted
as vector functions q(λ), are wavelength-dependent and correspond to the components of the
three sensors. The response within each color channel is proportional to the activity of neurons
firing, with any light falling within the nonzero region of the red cone function eliciting a certain
level of response.
Image Formation: Image formation is the process of creating a visual representation of a scene
using cameras or other imaging devices. It involves capturing light from the scene and
converting it into a digital or analog image. The image formation situation is thus as follows:
light from the illuminant with
SPD E(λ) impinges on a surface, with surface spectral reflectance function S(λ), is.
reflected, and then filtered by the eye’s cone functions q(λ). The basic arrangement
is as shown in Fig. 4.5. The function C(λ) is called the color signal and consists of
the product of the illuminant E(λ) times the reflectance S(λ): C(λ) = E(λ) S(λ).
The equations like Eq. (4.2), then, consider the image formation.
model is:
R = # E(λ) S(λ) qR(λ) dλ
G = # E(λ) S(λ) qG(λ) dλ
B = # E(λ) S(λ) qB(λ) dλ.
Camera Systems: Camera systems consist of various components such as lenses, sensors, and
image processing units. They capture and record images by converting light into electronic
signals. Humans follow a comparable approach when creating camera systems, where a high-
quality camera generates three signals per pixel, resembling retinal positions. Analog signals are
transformed into digital form, rounded down to whole numbers, and then saved. When utilizing
an 8-bit precision, the upper limit for any of the color components (R, G, B) is 255, while the
lower limit is 0.
Gamma Corrections: Gamma correction is a nonlinear adjustment applied to images to
compensate for the nonlinear way in which humans perceive brightness. It ensures that images
appear visually consistent across different devices.
Color-Matching Functions: These functions describe the human eye's response to different
wavelengths of light. They are essential for creating color spaces and color models that
accurately represent human perception.
CIE Chromaticity Diagram: The CIE chromaticity diagram is a graphical representation of all
visible colors. It's used to illustrate color relationships and is widely used in color science.
Color Monitor Specifications: Color monitors have specifications related to resolution, color
depth, and color accuracy. These specifications affect how accurately colors are displayed on the
screen.
Out-of-Gamut Colors: Some colors cannot be accurately reproduced on a particular device due
to limitations in its color gamut (range of reproducible colors).
White Point Correction: Adjustments are made to ensure that the white point of an image
appears neutral and consistent across different devices.
XYZ to RGB Transform: The conversion from CIE 1931 XYZ color space to RGB color
space, which is used to represent colors on displays.
Lab (CIE LAB) Color Model*: A color model designed to be perceptually uniform, meaning
equal perceptual differences in values correspond to equal visual differences.
More Color Coordinate Schemes: Various color coordinate systems like LCH, HSV, and HSL
offer alternative ways to represent and manipulate colors. There are several other coordinate
schemes in use to describe color as humans perceive it, with some confusion in the field as to
whether gamma correction should or should not be applied. Here we are describing device-
independent color—based on XYZ and correlated to what humans see. However, generally, users
make free use of RGB or R′, G′, B′.
RGB Color Model for Displays: A color model based on the additive mixing of red, green, and
blue light to produce a wide range of colors on electronic displays. In computer-generated
images, we save integers in the frame buffer that correspond to intensity. To ensure accurate
representation, a gamma correction lookup table (LUT) is placed between the frame buffer and
the display. If gamma correction is implemented on floating-point values prior to being
converted to integers and stored in the frame buffer, it's possible to use just 8 bits per channel
and prevent undesired contouring artifacts.
Multi-Sensor Cameras: Cameras with multiple sensors capture different spectral bands,
allowing them to gather more information about a scene.
Camera-Dependent Color: The color captured by a camera can be influenced by factors like
sensor sensitivity, white balance, and lens characteristics.
Subtractive Color: CMY Color Model: Used in printing, the CMY model involves subtracting
cyan, magenta, and yellow pigments from white to create different colors.
Transformation from RGB to CMY: The conversion process from the RGB color model to the
CMY color model.
Under-Color Removal: CMYK System: In printing, under-color removal adjusts the amount of
cyan, magenta, and yellow in shadow areas to produce more accurate colors.
Printer Gamuts: The range of colors that a printer can reproduce. It's often smaller than the
gamut of displays.
Multi-Ink Printers: Printers with more than the standard CMYK inks can produce a wider
range of colors.
Video Color Transforms Conversions between color spaces to ensure consistent color
representation in the video.
YUV Color Model: Used in video processing, it separates brightness (luma) from color
information (chroma) to compress video data.
YIQ Color Model: Another video-specific color model used for analog NTSC television
transmission.
YCbCr Color Model: A digital color model used for video compression, like YUV but scaled
and offset.
CONCLUSION
In summary, color in image and video is a cornerstone of multimedia, shaping how audiences
perceive, engage with, and respond to visual content. Its nuanced use, combined with
technological advancements and attention to accessibility, continues to drive innovation in
multimedia production, ensuring that color remains a central and dynamic element of this ever-
evolving field.

References
 Li, Ze-Nian, Drew, Mark S., Liu, Jiangchuan Fundamentals of Multimedia: Color in
Image and Video. 4, pg 81–133 2nd edition (2014)

H. J. Trussell, E. Saber, and M. Vrhel,

“Color image process-
ing,” IEEE Signal Processing
Magazine, v o l . 2 2 , n o . 1 , p p . 1 4 –
22, 200
 D.H. Pritchard, U.S. color television fundamentals a review. IEEE Trans. Consum.
Electron. 23(4), 467–478 (1977)
 G. Wyszecki, W.S. Stiles, Color Science: Concepts and Methods, Quantitative Data and
Formulas, 2nd edn. (Wiley, New York, 2000)
 R.W.G. Hunt, in 2nd Color Imaging Conference Transforms and Transportability of
Color. Color reproduction and color vision modeling. Society for Imaging Science and
 Technology (IS&T)/Society for Information Display (SID) Joint Conference (1993), pp.
1–5
 M.J. Vrhel, R. Gershon, L.S. Iwan, Measurement and analysis of object reflectance
spectra. Color Res. Appl. 19, 4–9 (1994)
 R.W.G. Hunt, The Reproduction of Color, 6th edn. (Fountain Press, England, 2004) 114
4 Color in Image and Video
 J.F. Hughes, A. van Dam, M. McGuire, D.F. Sklar, J.D. Foley, S.K. Feiner, K. Akeley,
Computer Graphics: Principles and Practice, 3rd edn. (Addison-Wesley, Boston, 2013)

H. J. Trussell, E. Saber, and M. Vrhel,

“Color image process-
ing,” IEEE Signal Processing
Magazine, v o l . 2 2 , n o . 1 , p p . 1 4 –
22, 200