Basic Concepts of Computer Vision — Part 2
Computer vision allows machines to perceive, interpret, and understand the visual world just as humans do. From autonomous vehicles navigating busy streets to medical imaging aiding in diagnoses, computer vision is transforming the way we interact with the world.
Some of the basic concepts that are very popular in computer vision are briefly explained in these computer vision series of articles. In this part 2 following topics are discussed:
- Visible Spectrum
- Luminance
- Color Representation
- Gamut
- Out of Gamut colours
- Color Transformations
- Non-perceptive uniformity
- Color Constancy
- Color Temperature
1. Visible Spectrum:
The electromagnetic wavelengths of light that are perceptible to the human eye is known as visible spectrum. It covers the colors of the rainbow, from violet at shorter wavelengths to red at longer wavelengths. The visible spectrum typically ranges from approximately 380 to 750 nanometers (nm).
2. Luminance :
Luminance measures the intensity of light or color. It quantifies the amount of light emitted by an object or area and is an essential aspect of color perception. Luminance can be measured in various units, such as candelas per square meter (cd/m²) or nits in displays.
3. Color Representation:
Color representation is a method of encoding and displaying colors. For example representing colors in RGB (Red, Green, Blue) or HSV (Hue, Saturation, Value). These representations are used in various applications, from digital displays to printing.
4. Gamut:
A gamut is a range of colors that a particular color model or device can display or produce. It defines the limits of the color space, encompassing all the colors that can be produced or displayed. Different devices and color models have different gamuts.
5. Out of Gamut Colors:
Out of gamut colors are colors that fall outside the range or limitations of a particular color space or device. When colors are out of gamut, they cannot be accurately represented or reproduced, which can result in color distortion.
6. Color Transformations:
Color transformations involve converting color representations from one color space to another. This is done for various reasons, such as color correction, matching colors between different devices, or enhancing the perceptual qualities of an image. For example: color transformation from RGB color space to XYZ color space.
7. Non-perceptive Uniformity:
Non-perceptive uniformity refers to the phenomenon where changes in color or brightness are not perceived uniformly by the human eye. In some regions of color space, small changes are more noticeable, while in other regions, more significant changes may go unnoticed. In colour image processing we need to evaluate the colour differences.
8. Color Constancy:
Color constancy is the human visual system’s ability to perceive colors of objects consistently under varying lighting conditions. If two sources of light have different spectrums, the white object should change its colour but the perceived colour of object remains relatively constant under varying illuminating condition.
For example, a red apple will still appear red under different types of lighting, such as natural daylight or indoor incandescent lighting.
9. Color Temperature:
Color temperature is a characteristics of visible light, a way to measure the spectral(hue) of the light. It is determined by comparing its chromaticity with that of an ideal black-body radiator. It is usually expressed in degrees Kelvin (K). Colour temperature of 2700–3600 are recommended for most general indoor lighting.
References
E. R Davies, Computer Vision: Principles, Algorithms, Applications, Learning, Academic Press; 5th edition; 2017
D Forsyth, Computer Vision. A Modern Approach
