 |
MPEG-1 Video Encoding
MPEG-1 coding offers high compression and
moderate quality for Video CD and CD-ROM. |
In 1992, the ISO MPEG committee
published the MPEG-1 specification for compressed digital video as ISO 11172. This
standard was designed to allow 74 minutes of digital video to be stored on a compact disc.
The quality was noticeably superior to existing digital video formats and, very
importantly, represented an international standard which could be adopted universally by
the industry. Philips, as one of the MPEG contributing companies, adopted
MPEG-1 for CD-i and it was subsequently adopted for Video CD.
MPEG-1 is intended for non-interlace displays in contrast with MPEG-2
which is for interlaced video display. This means that a PAL or
NTSC video signal will be encoded at SIF resolution, ie half the number
of lines per video frame and half the pixels per line.
MPEG-1 Example
The following diagram illustrates the sequence of frames in an MPEG-1 video
sequence. Note that the display order is different from the order in which
they are read from the disc.
Each macroblock of a B-frame is coded as a change from the
previous or next P-frame or I-frame. The I-frames are repeated at approximately 0.5
second intervals, otherwise the PBBPBB.. pattern is repeated indefinitely.
For the constrained parameter system, used for
Video CD and CD-i, there are a maximum of 396 macroblocks per frame.
Therefore the source PAL or NTSC video signal is encoded at SIF
resolution, ie half the number of lines per video frame and half the
pixels per line compared with CCIR 601.
|
