Light and color theory form the foundation of digital imaging and computer vision. Understanding the electromagnetic spectrum, visible light range, and how humans perceive color is crucial for developing accurate color reproduction techniques and image processing algorithms.

Color models and spaces define how colors are represented and organized in digital systems. From RGB and CMYK to more advanced spaces like Lab and LCH, these frameworks enable consistent color reproduction across devices and influence the design of color-based computer vision applications.

Electromagnetic spectrum

• Encompasses all types of electromagnetic radiation, ranging from radio waves to gamma rays
• Crucial for understanding how different wavelengths of light interact with digital imaging sensors and affect color reproduction
• Forms the foundation for color theory and its applications in computer vision and image processing

Visible light range

• Occupies a small portion of the electromagnetic spectrum, approximately 380-700 nanometers
• Corresponds to the colors humans can perceive, from violet to red
• Divided into distinct color bands (violet, blue, green, yellow, orange, red)
• Impacts how digital cameras and image sensors capture and interpret color information

Infrared and ultraviolet

• Infrared radiation lies beyond the red end of the visible spectrum, with wavelengths longer than 700 nm
• Ultraviolet radiation occurs below the violet end, with wavelengths shorter than 380 nm
• Both types of radiation can be detected by specialized cameras and sensors
• Used in various computer vision applications (night vision, material analysis, forensics)
• Require specific filtering techniques to prevent interference with visible light imaging

Color perception

• Involves complex processes in the human visual system and brain
• Influences the design of color models and spaces used in digital imaging
• Crucial for developing accurate color reproduction techniques in computer vision systems

Human visual system

• Consists of the eyes, optic nerves, and visual cortex in the brain
• Retina contains two types of photoreceptors: rods (sensitive to light intensity) and cones (responsible for color vision)
• Three types of cones: short (S), medium (M), and long (L) wavelength sensitive
• Processes visual information through parallel pathways for different aspects of vision (color, motion, form)

Trichromatic theory

• Proposed by Young and Helmholtz in the 19th century
• States that human color vision relies on three types of cone cells
• Each cone type responds differently to various wavelengths of light
• Forms the basis for RGB color models used in digital displays and cameras
• Explains color mixing phenomena and the ability to create a wide range of colors from three primaries

Opponent process theory

• Developed by Ewald Hering to explain certain color perception phenomena
• Proposes three opponent color pairs: red-green, blue-yellow, and black-white
• Accounts for the inability to perceive "reddish-green" or "yellowish-blue" colors
• Explains afterimages and color contrast effects
• Influences the design of some color spaces (Lab, LCH) used in image processing

Color models

• Mathematical representations of color used in digital imaging and computer graphics
• Enable consistent color reproduction across different devices and platforms
• Crucial for accurate color manipulation and analysis in computer vision applications

RGB color model

• Additive color model based on the trichromatic theory of vision
• Uses red, green, and blue primary colors to create a wide range of hues
• Each color represented by values ranging from 0 to 255 (8-bit) or 0 to 1 (floating-point)
• Widely used in digital displays, cameras, and image processing software
• Allows for easy color manipulation through channel separation and blending

CMYK color model

• Subtractive color model used primarily in printing and physical color reproduction
• Based on cyan, magenta, yellow, and black inks
• Each color value represents the amount of ink coverage (0-100%)
• Requires conversion from RGB for digital images to be printed
• Limited color gamut compared to RGB, affecting color accuracy in print reproduction

HSV and HSL models

• Represent colors using hue, saturation, and value/lightness components
• Hue: the color itself (0-360 degrees on a color wheel)
• Saturation: color intensity or purity (0-100%)
• Value/Lightness: brightness or darkness of the color (0-100%)
• More intuitive for human perception and color selection
• Useful for color-based segmentation and object detection in computer vision

Color spaces

• Define the range and organization of colors within a specific color model
• Crucial for ensuring consistent color reproduction across different devices and software
• Impact the accuracy and efficiency of color-based computer vision algorithms

sRGB vs Adobe RGB

• sRGB: standard color space for most consumer displays and digital cameras
• Offers a smaller color gamut but ensures consistent color reproduction across devices
• Adobe RGB: wider color gamut, particularly in the green-cyan region
• Used in professional photography and high-end printing applications
• Requires color management to ensure accurate display on sRGB devices

CIE color spaces

• Developed by the International Commission on Illumination (CIE)
• CIE XYZ: based on human color perception experiments
• Serves as a device-independent reference color space
• CIE xyY: derived from XYZ, separates chromaticity (xy) from luminance (Y)
• Used for color gamut mapping and colorimetric calculations

Lab and LCH color spaces

• Lab: perceptually uniform color space with L (lightness), a (red-green), and b (blue-yellow) components
• Closely aligns with human color perception and opponent process theory
• LCH: cylindrical representation of Lab, using lightness, chroma, and hue
• Useful for color difference calculations and color harmony analysis
• Employed in advanced color correction and grading techniques

Color depth

• Determines the number of distinct colors that can be represented in a digital image
• Affects image quality, file size, and processing requirements in computer vision applications

Bit depth

• Refers to the number of bits used to represent each color channel
• Common bit depths: 8-bit (256 levels per channel), 10-bit (1024 levels), 12-bit (4096 levels)
• Higher bit depths allow for smoother color gradients and more precise color information
• Impacts the ability to perform detailed color analysis and manipulation in image processing

Dynamic range

• Represents the range of luminance values that can be captured or displayed
• Measured in stops or f-stops, with each stop doubling the amount of light
• High Dynamic Range (HDR) imaging captures a wider range of luminance than standard images
• Affects the ability to preserve detail in very bright or dark areas of an image
• Crucial for computer vision applications in challenging lighting conditions

Color temperature

• Describes the apparent color of a light source
• Influences the overall color balance of captured images
• Important for color correction and white balance adjustments in image processing

Kelvin scale

• Measures color temperature in degrees Kelvin (K)
• Lower temperatures (2000-3000K) appear warm (reddish)
• Higher temperatures (6000-7000K) appear cool (bluish)
• Daylight typically ranges from 5000-6500K
• Used to characterize light sources and adjust white balance in digital imaging

White balance

• Process of removing unrealistic color casts caused by different light sources
• Aims to render white objects as truly white in the final image
• Automatic white balance uses algorithms to detect and correct color temperature
• Manual white balance allows for creative control over color mood
• Critical for accurate color reproduction in computer vision and image analysis