TABLE OF CONTENTS This chapter describes color coding systems that are used to convey image data derived from additive (RGB) primaries. I outline nonlinear R?G?B?, explain the formation of luma, denoted Y?, as a weighted sum of these nonlinear signals, and introduce the color difference (chroma) components [ B?-Y?, R?-Y?], [C B, C R], and [ P B, P R].
The design of a video coding system is necessarily rooted in detailed knowledge of human color perception. However, once this knowledge is embodied in a coding system, what remains is physics, mathematics, and signal processing. This chapter concerns only the latter domains.
A monochrome video system ideally senses relative luminance, described on page 205. Luminance is then transformed by the gamma correction circuitry of the camera, as described in Gamma in video, on page 261, into a signal that takes into account the properties of lightness perception. At the receiver, the CRT itself imposes the required inverse transfer function.
A color image is sensed in three components, red, green, and blue, according to Additive reproduction (RGB), on page 234. To minimize the visibility of noise or quantization, the RGB components should be coded nonlinearly.
Red, green, and blue tristimulus (linear light) primary components, as detailed in Color science for video, on page 233, can be considered to be the coordinates of a three-dimensional color space. Coordinate values between zero and...
