Video on the web
Conventional video, such
as those used current TV and camcorders, are encoded in analog signals.
The TV signal format used
in North America is called NTSC. European countries used PAL format.
The Red Green Blue (RGB)
signals in the video were encoded in YIQ or YUV
format, where Y encoded the brightness, I and Q in NTSC (U and V
in PAL) encoded the color information. The reason is to take advantage
of human perception. We are more sensitive to brightness than color.
Among colors, we are more sensitive to orange than blue.
In video system, the YIQ
signals can be transmitted over separate channels to reduce interference
and preserve high fidelity. These are called
component
signals.
The YIQ signal can be also
be encode in an analog signal over one channel for saving bandwidth and
resources. This called composite signal.
Most camcorders or VCR's
have three RCA type connectors (jacks) with yellow one conveying composite
signal, red one conveying right audio channel, and white one conveying
left audio channel.
Some camcorders or advanced
VCR's have round connector called S Video jack with separate pins conveying
the component signals.
The NTSC video signal contains
30 frames (still pictures) per seconds.
Each frame is scanned in by the camera with 525 horizontal lines (actually
slightly slanted). The lines with the odd numbers were sent first.
The lines with even numbers were sent later. They formed the two fields
of the frames. Therefore there are 60 fields per seconds. The 60
number is related to the AC power frequency.
The PAL video signal contains
25 frames per seconds, and 50 fields per seconds. In Europe, their
AC power frequency is 50 Hz.
To display these analog video
data, they must be digitized by a video captured card. Since digitized
video produces hugh amount of data, it is typically compressed through
one of the following methods:
-
reduce the resolutions (e.g.,
from 320x240 pixels to 240x180 pixels)
-
reduce the frame rates (e.g.,
from 30 frames to 15 frames)
-
use transformation techniques
such as Discrete Cosine Transform (DCT) to reduce redundant information
within a frame (intraframe)
-
use interpolation techniques
to reduce redundant information between frame (interframe)
-
use huffman coding to present
frequent patterns with fewer bits.
Microsoft's
Audio-Video Interleave (.avi), Apple's
QuickTime (.mov), MPEG (.mpg), and RealNetworks'
RealVideo (.rm) are the four digital video formats supported by the
web browsers through the plug-in software or external viewer applications.
How
to capture video using Intel Smart Video Capture card and Asymetrix's DVP
software.
Including capture.avi video
file in the web page:
MPEG (Moving Picture Experts
Group) Video File Format
-
MPEG (pronounced EM-peg)-1,
developed by the Moving Picture
Experts Group, is an International Standards Organization (ISO) standard.
-
MPEG
Resources home page.
-
MPEG-1 generates about 1.5Mbps
data and is for storage and retrieval of moving pictures and audio
on storage media
-
MPEG-2 generates higher quality
video (broadcast/satellite TV) at above 4 Mbps.
-
MPEG-4 intends for low speed
video, multimedia applications.
-
MPEG-7 is being developed as
the content representation standard for information search.