Kromasubsampling

Kromasubsampling is a digital imaging technique used to reduce color information by sampling chrominance channels at a lower resolution than luminance. It leverages the observation that human vision is more sensitive to brightness details than to color details, allowing data savings with limited perceptual impact.

The method operates in color spaces that separate luminance and chrominance, such as YCbCr. In practice, the chroma channels (Cb and Cr) are decimated according to a sampling pattern, for example one chroma sample per several luma samples in horizontal and/or vertical directions. During display or decoding, the subsampled chroma is upsampled through filtering to reconstruct a full-color image, with quality dependent on the chosen pattern and filters.

Common chroma subsampling formats associated with kromasubsampling include 4:4:4, 4:2:2, and 4:2:0, which differ in how many chroma samples are kept relative to luminance. Kromasubsampling may refer to a family of schemes or a specific implementation that prioritizes certain trade-offs between compression efficiency, color fidelity, and computational complexity, while maintaining the core principle of reducing chroma resolution to save bandwidth and storage.

Benefits of kromasubsampling include reduced bandwidth and storage requirements, enabling higher frame rates or resolutions within fixed resources. Potential drawbacks involve color artifacts such as edge color fringing, mosquito noise, or color bleeding, particularly on high-frequency color content or sharp transitions.

Applications include video streaming, broadcasting, and imaging pipelines where chroma subsampling is commonly employed to optimize performance and cost. See also: chroma subsampling.