Chroma subsampling

The color subsampling (English chroma subsampling [ kɹoʊmə sʌb ˌ sæmplɪŋ ], color subsampling ) is in the range of color models, a method that chrominance ( color information ) to store a relation to the luminance ( brightness information ) reduced sampling rate. Motivation is to reduce the memory space or in communication engineering systems, the reduction of the transmission bandwidth without subjective- optical quality loss.

Applications are in the video technology, digital image formats such as JPEG and in the field of color television technology.

General

The color sub-sampling makes use of the physiology of the human eye advantage that the color information, in contrast to the brightness information can only perceive with reduced resolution. As can be seen in the illustration, the two color components are reduced in the last two displays in resolution, without thereby impairing the impression of the color image in the first figure. The image sharpness is essentially contained in the luminance information. Generally provide all of the color model with the luminance information separated by the color difference information, the possibility of color subsampling, such as YCbCr or CIELab.

In the field of analog video technology, the color information is transmitted with low bandwidth, digital video technology and digital image processing directly corresponds to the area of ​​a reduction of the sampling and the spatial distance between the sampling points of the color information. Although the color subsampling primarily relates to digital systems and has nothing to do with the cause of the color television technology, comes a part of the names of the analog color television technology, especially from the NTSC color television system.

Designations

There are various forms of the color sub-sampling, which describes the sub-sampling of the color signals in relation to the brightness signal. It usually following notation is used with three numbers:

This notation is based on a pixel block of A pixels wide and 2 pixels high and describes in relation to the value of A, the sample contained therein for the color information.

• The first digit of A is here today usually set to 4, since traditionally all color subsampling rates were low powers of two (even if 4:2:2 could be specified as well as 2:1:1, this is not common in practice). Originally A is the factor of the sampling rate of the luminance signal Y 'is a multiple of the NTSC video band width of 3.375 MHz, Y'CbCr the color model. Typically, the sampling rate is four times the video bandwidth, which is the pixel clock obtained at 13.5 MHz in the digital video standard ITU-R BT 601. Rarely, the number 3 is used as the leading digit, otherwise would have to be used at sample rates of one-third fractures in the values ​​of B and C.

• The second point B is the sample rate of the two color channels Cb and Cr in the top row of pixels in relation to A and initially to 3.375 MHz. She has in the ITU -R BT 601 represents the value of two, representing the color sub-sampling

• The third digit C describes the same value as the second digit, but for the bottom row of pixels of the pixel block.

• Occasionally, a fourth number is specified in the connection in the naming scheme that indicates the sampling rate ratio of a co-assigned alpha channel.

The following are conventional color samples:

White circles represent the sampling positions for the luminance, light and dark gray circles, the positions for the two chrominance signals.

4:4:4 ( 1x1, 1x1, 1x1)

In this case, no color sub-sampling is performed. Applications of 4:4:4 are in the range of high-quality digital image processing and the RGB color space. In the RGB color space is because of the lack of separation between brightness and color information, basically no color subsampling possible.

4:2:2 ( 2x1, 1x1, 1x1)

This color subsampling is originally arose from the analog color television standard NTSC and 601 is used in the standard ITU -R BT for digital video signals. A distinction is made between the horizontal and vertical color scanning: the scanning in the horizontal direction is only half the size in the vertical direction. By the usual in the analog television standard NTSC or PAL interlaced the color resolution is the same in both directions. The Digital Betacam - tape format uses the 4:2:2 method.

4:2:0 ( 2x2, 1x1, 1x1)

The format of 4:2:0 is used in digital images in the JPEG standard or digital video in the MPEG standard and has an identical sampling in both spatial directions. Here come as shown in the above figures, slightly offset sampling of the color signal before: If JPEG is sampled centric, with MPEG in the same vertical orientation as the brightness signal.

Although the reference to the NTSC video bandwidth, JPEG and MPEG does not make sense, the term scheme of color sub-sampling was maintained with digits. The PAL - DV tape format ( DV25 PAL) uses the 4:2:0 method.

4:1:1 ( 4x1, 1x1, 1x1)

Similar to 4:2:2, although the color channels in the horizontal direction with only 1/4 of the sampling rate of the luminance signal is sampled. The NTSC DV tape format ( DV25 NTSC) uses the 4:1:1 method.

More color samples

In addition, there are still some rarely applied color samples. In the area of the DVCAM ( DV SDL) is a color subsampling, called 3:1:1 application, which was coined by Sony. However, the pixel clock is not 3 times the NTSC bandwidth but 12 MHz.

In the field of High Definition Television (HDTV ) on the HD -SDI interface according to the standard SMPTE 292M comes before 22:11:11. However, the term is used only rarely, and is derived from the pixel clock of 74.25 MHz used in HDTV.