Specifications

Security 2009
KNOW-HOW: Compression Methods
S14
KNOW-HOW · KNOW-HOW · KNOW-HOW · KNOW-HOW · KNOW-HOW · KNOW-HOW · KNOW-HOW
Compression Methods and Recording Time
Large amounts of data are created when analog video signals are con-
verted into digital data. If they would be recorded in an uncompressed
way, even a large hard disk would only last a few hours.
A sequence of 30 minutes with 25 pictures/sec. of a high quality
would quickly have 40 or more Gbytes of raw data. In order to re-
duce this amount of data without greatly reducing the quality of the
recording, high-performance compression methods have been de-
veloped over the last few years.
The differences and main advantages of the different methods are
roughly described below. Further information on the individual
methods
can be found on the Internet.
JPG is a compression method for individual pictures (still images). The
compression rate works out at a factor 10-20. The JPG format is very
popular. A picture which is stored in this format can be recalled by al-
most any kind of computer system. It is also ideally suited for sending
it as an e-mail attachment.
MJPG (Motion JPG) is a method based on JPG where individual pic-
tures are compressed as a sequence. The advantage of this method is
that each individual picture can be taken from the sequence as a still
image.
MPEG-2. This method was developed for the compression of TV pic-
tures. The new TV standards DVB-T, DVB-S, DVB-C, and also the video
DVD are using this method. The compression rate is clearly higher than
at the MJPG. The MPEG-2 is of an advantage where only smaller areas
of a picture change, because only the changes from the previous pic-
ture are saved. A complete picture is only saved every 10-15 pictures
for reference.
But this is exactly the limitation. From a sequence which has been
compressed by MPEG-2, pictures may not be taken accurately. When
the MPEG-2 files are exported, a special programme is required for the
reproduction because the standard viewers do not recognise these files.
This programme usually has to be paid for.
MPEG-4 is the further development of the MPEG-2. MPEG-4 resolves
the picture more effectively and can thus compress a sequence
quicker and maybe smaller. Special hardware is usually used for the
MPEG-4 compression, because a higher efficiency of the electronic
equipment is required.
Reducing the amount of data does not only save storage space on
the hard disks, the reduction for a faster network transmission is more
and more important. This is particularly important for narrow-band
connections, e.g. Internet connections.
How does the MPEG-4 affect the recording time?
The MPEG-4 method has a crucial affect on the possible recording time.
At a MJPG recording, the required capacity could be determined quite
accurately, because each picture had the same size which only de-
pended on the recording quality. This completely changes with MPEG-4.
The amount of data which has to be recorded depends on the amount
of change in the picture contents from picture to picture. These
changes may be caused by different kind of reasons.
The two following sequences are to illustrate this with a typical example.
A black car is seen moving up a driveway. (see picture 1 below)
Sequence 1 shows a recording by MJPG method. Each picture is com-
pressed individually and stored. Thus, each individual picture can be
taken from this sequence. A storage space for 6 full picture contents will
be required.
Sequence 2 shows a recording by MPEG-4 method (see picture 2
below). The first picture is stored completely as a reference picture.
This reference picture will be called ”1frame”. After that, only the
changes in the picture will be filtered and stored. Thus, only the rela-
tively small section of the picture will be stored where the vehicle is
moving in. Only after 10 such pictures, another reference picture
(1frame) is stored completely. This is one of the reasons, why storage
capacity is also used, even if no changes have occured in the picture.
Picture 1
Picture 2