Video - A Practical Guide to Video and Audio Compression - page 88

Text safe area

Action safe area

Overscan margin

Figure 5-11 Safe areas.

The term NTSC is commonly taken to mean an image that is 720 non-square pixels
wide by 480 visible lines high. This is delivered on a 525-line raster being presented
approximately 30 times per second, again using an interlaced alternate field scheme.
For typical PAL TV transmissions, the picture size will be 720 non-square pixels wide
by 579 visible lines delivered on a 625-line raster. This raster will be refreshed 25 times
per second using an interlaced presentation of two alternate fields.
The SECAM system is structurally very similar to PAL, but the encoding frequencies
and technical aspects of the color model are different. A special SECAM decoder must be
used because a PAL decoder will not work. The signals are converted between the two for-
mats relatively easily since they are physically similar in terms of raster size (625 lines) and
image size (720 579).

Figure 5-12 shows the three most common physical raster sizes.

×

5.8.5 Source Input Format

Source input format ( SIF ) is a term that was originally defined in the MPEG-1 standard to
describe the format of the source video being compressed.
Beware. It is a similar size to that defined by the Common Intermediate Format ( CIF )
but is not the same thing.
CIF is defined in the basic form as a 352

× 288 pixels at a display frequency of 29.97

29.97 fps. Where CIF jumbles attrib-
utes of the U.S. and European TV services, SIF keeps them separate and consistent.
SIF is very easily sub-sampled from TV signals. CIF is most assuredly not, since both
the spatial and the temporal data must be interpolated with a non-integer calculation
depending on the source format of the video.

fps. SIF is defined as 352 × 288 × 25 fps or 352 × 240 ×