In chrominance subsampling, what information is primarily averaged?

In chrominance subsampling, the primary focus is on the chrominance information from pixel pairs. This technique is based on the human visual system's sensitivity to luminance (brightness) as opposed to chrominance (color). By averaging the chrominance information, which may involve combining the color information of adjacent pixels, it allows for a reduction in data size while maintaining the perceived visual quality of the image.

This process typically entails encoding fewer chrominance samples than luminance samples since humans are less adept at discerning color detail compared to light intensity details. As a result, chrominance subsampling significantly optimizes storage and bandwidth without substantially impacting image quality, especially in formats used for streaming and broadcasting.

The other choices relate to aspects of video data that are not primarily the focus of chrominance subsampling. For instance, luminance information of pixels is not averaged but rather retained in greater detail. Total pixel data encompasses both luminance and chrominance information, which is not the specific target of subsampling. Lastly, raw digital video files refer to unprocessed video data that has not undergone any compression or subsampling and thus does not pertain directly to the concept of chrominance subsampling itself.