What is Multimedia?

Multimedia is everything you can hear or see: texts, books, pictures, music, sounds, CDs,
videos, DVDs, Records, Films, and more.

File Formats
Sounds:

Audio Interchange File Format (*.aiff, *.aif, *.aifc)

Motion Pictures Expert Group Layer-3 (*.mp3, *.m3u)

Musical Instrument Digital Interface (*.midi, *.mid, *.rmi)

Unix Environment (*.au, *.snd)

Microsoft Wave (*.wav)

Microsoft Streaming Format (*.asf, *.asx)

CDAudio (*.cda)

Multimedia Sound Formats

Sound can be stored in many different formats:

The MIDI Format

The MIDI (Musical Instrument Digital Interface) is a format for sending music
information between electronic music devices like synthesizers and PC sound cards.
The MIDI format was developed in 1982 by the music industry. The MIDI format is very
flexible and can be used for everything from very simple to real professional music
making.
MIDI files do not contain sampled sound, but a set of digital musical instructions
(musical notes) that can be interpreted by your PC's sound card.
The downside of MIDI is that it cannot record sounds (only notes). Or, to put it another
way: It cannot store songs, only tunes.

The upside of the MIDI format is that since it contains only instructions (notes), MIDI
files can be extremely small. The example above is only 23K in size but it plays for
nearly 5 minutes.
The MIDI format is supported by many different software systems over a large range of
platforms. MIDI files are supported by all the most popular Internet browsers.
Sounds stored in the MIDI format have the extension .mid or .midi.
The RealAudio Format
The RealAudio format was developed for the Internet by Real Media. The format also
supports video.
The format allows streaming of audio (on-line music, Internet radio) with low
bandwidths. Because of the low bandwidth priority, quality is often reduced.
Sounds stored in the RealAudio format have the extension .rm or .ram.
The AU Format
The AU format is supported by many different software systems over a large range of
platforms.
Sounds stored in the AU format have the extension .au.
The AIFF Format
The AIFF (Audio Interchange File Format) was developed by Apple.
AIFF files are not cross-platform and the format is not supported by all web browsers.
Sounds stored in the AIFF format have the extension .aif or .aiff.
The SND Format
The SND (Sound) was developed by Apple.
SND files are not cross-platform and the format is not supported by all web browsers.
Sounds stored in the SND format have the extension .snd.

The WAVE Format

The WAVE (waveform) format is developed by IBM and Microsoft.
It is supported by all computers running Windows, and by all the most popular web
browsers.
Sounds stored in the WAVE format have the extension .wav.
The MP3 Format (MPEG)
MP3 files are actually MPEG files. But the MPEG format was originally developed for
video by the Moving Pictures Experts Group. We can say that MP3 files are the sound
part of the MPEG video format.
MP3 is one of the most popular sound formats for music recording. The MP3 encoding
system combines good compression (small files) with high quality. Expect all your future
software systems to support it.
Sounds stored in the MP3 format have the extension .mp3, or .mpga (for MPG Audio).

Video Format: (Movies)

QuickTime files (*.mov, or *.qt) created with versions one and two of
QuickTime. (Versions three and four are not supported.)

Audio Video Interleave (*.avi)

Motion Picture Experts Group (*.mpg, *. mpeg, *.m1v, *.mp2, *.mpa,

*.mpe)

Microsoft Streaming Format (*.asf, *.asx)

Animated GIF (*.gif) While animated GIFs are not movies, they come
close enough for discussion in this document. (Note: Only PPT2000
supports animated gifs.)

Multimedia Video Formats

Video can be stored in many different formats.

The AVI Format
The AVI (Audio Video Interleave) format was developed by Microsoft.
The AVI format is supported by all computers running Windows, and by all the most
popular web browsers. It is a very common format on the Internet, but not always
possible to play on non-Windows computers.
Videos stored in the AVI format have the extension .avi.
The Windows Media Format
The Windows Media format is developed by Microsoft.
Windows Media is a common format on the Internet, but Windows Media movies cannot
be played on non-Windows computer without an extra (free) component installed. Some
later Windows Media movies cannot play at all on non-Windows computers because no
player is available.
Videos stored in the Windows Media format have the extension .wmv.
The MPEG Format
The MPEG (Moving Pictures Expert Group) format is the most popular format on the
Internet. It is cross-platform, and supported by all the most popular web browsers.
Videos stored in the MPEG format have the extension .mpg or .mpeg.
The QuickTime Format
The QuickTime format is developed by Apple.
QuickTime is a common format on the Internet, but QuickTime movies cannot be played
on a Windows computer without an extra (free) component installed.
Videos stored in the QuickTime format have the extension .mov.
The RealVideo Format
The RealVideo format was developed for the Internet by Real Media.
The format allows streaming of video (on-line video, Internet TV) with low bandwidths.
Because of the low bandwidth priority, quality is often reduced.
Videos stored in the RealVideo format have the extension .rm or .ram.
The Shockwave (Flash) Format
The Shockwave format was developed by Macromedia.
The Shockwave format requires an extra component to play. This component comes
preinstalled with the latest versions of Netscape and Internet Explorer.
Videos stored in the Shockwave format have the extension .swf.

Image formats

1. Image sources

• Images may be obtained from a variety of sources, including photographs, video,

or computer data.
• Photographs or other two-dimensional materials can be captured using a flatbed
desktop scanner similar to a photocopier. This can produce high-quality images in
 either monochrome or colour. Specialised drum scanners used in print
reproduction produce very high quality images.
• Video pictures can be digitised as freeze frames from any video source using a
suitable frame grabber board. Specialised systems can capture clips of moving
video pictures in real time. Video images are generally lower in quality than those
obtained using a scanner.
• Original images can be generated using a computer 'drawing' or 'paint' package.
This will often be done by an experienced artist, designer, or technical illustrator.
• Computer data can be manipulated in visualisation packages and the resulting
output can be incorporated as an image.

2. Image types

Images can be stored in many formats. Three fundamental types of images can be
considered: vector, bitmap, and moving images.
Vector image
The vector image stores the sequence of graphics instructions required to reproduce the
representation.

Bitmap image
The bitmap image stores the pixels required to reproduce the representation.

Moving images
Moving images are stored as a sequence of pictures used to give the illusion of motion.

Vector and bitmap images will be considered here. Moving images will be considered
separately elsewhere.

Detailed Description
Vector images
Vector images define an image as a set of instructions composed of graphic objects or
primitives. Strictly speaking, vectors are lines, but vector graphics generally include
ellipses, arcs, curves, filled areas and in many cases text. Many vector formats allow
complex bezier curves to be stored as mathematical equations.

Advantage of Vector images:

Vector graphics are well suited to computer handling, because they can be
mathematically manipulated. This enables individual elements to be scaled and stretched.
Vector graphics will always display or print at the best resolution for any output device.
Original vector images can be created in application programs. Such applications are
often called drawing packages and are sometimes said to be object-oriented. Vector
images are used extensively in Computer Aided Design. They are also highly suitable for
certain graphic applications.
Vector images are more efficient than bitmap images in storing certain graphic
representations. In most cases vector files are recorded in symbolic form as text codes.
Although inefficient, this approach is offset by the high efficiency of vector
representation.
Increasingly, typefaces or fonts or are stored as vector information, allowing characters to
be resized or otherwise manipulated without loss.
Vector images can generally be translated into bitmaps. That is essentially what happens
when a vector image is displayed on a conventional computer screen. Often, an image
will be manipulated as a vector model, but ultimately displayed as a bitmap image.
Unlike bitmaps, vector images must be interpreted by a rendering engine to build a
graphic image. This can impose a significant overhead in display times.
In some cases, bitmap images can be automatically traced algorithmically to produce a
vector representation. This allows the image to be resized or otherwise mathematically
manipulated. Generally the success of this technique depends upon the source image.

Certain images, such as simple black and white silhouettes can translate adequately. It is
invariably impossible to convert complex subjects such as those found in photographs.
In certain cases it may be preferable to store an image as a vector graphic. Many line
drawings, silhouettes, logos, diagrams and charts can be more efficiently stored as vector
graphics. This offers the advantage of resolution independence. A vector graphic will
always reproduce at the optimum resolution for an output device and will maintain a
consistent size when reproduced.
Bitmap images
Bitmapped images or bitmaps represent an image as a grid of picture elements or pixels,
each with a value mapping a particular value of colour and intensity.

Any image can be represented in bitmap form, although this is not necessarily always
efficient. Bitmaps are generally employed to represent real world images. These can be
acquired using either a scanner or a video frame grabber.
Existing bitmap images can be modified using photo retouching packages. Image
processing functions can be used to modify such characteristics as sharpness, contrast,
and colour balance.
Original bitmap images can also be created in software applications. These are often
called painting packages. There are certain limitations in creating original images in
bitmap form:

• A bitmap image is stored as a pattern of pixels that can be difficult to modify.

 • A bitmap image is effectively tied to the resolution at which it was created.
• A bitmap image of high resolution can require a great deal of storage.

When creating original artwork images it is often preferable to create images in vector
form, possibly converting to a bitmap image if required.
Mixed types
Vector and bitmap images tend to come together on screen displays and in printed output.
Generally, vector images are rasterized or converted to bitmap images for display on
bitmapped screens or other ultimately bitmapped devices such as laser printers.
For this reason, the software interpreters or engines responsible for handling screen
display and printed output often deal with a combination of vector and bitmap images.
The Microsoft Windows environment supports metafiles, sequences of graphics
instructions that can include both vector and bitmap descriptions. Page description
languages such as Hewlett Packard's PCL or Adobe PostScript support both vector and
bitmap images. This feature can cause confusion in particular when dealing with
PostScript as an image file format.

Image File formats :

Vector image file format:
Given below are brief details of some popular vector image file formats. The details
given are no intended to be comprehensive and are for general guidance only.

PostScript

Adobe PostScript is the de-facto standard for desktop publishing and produces consistent
results on output devices ranging from laser printers to phototypesetters.Originally
intended for printers and other output devices, PostScript is a page description language
which supports both vector and bitmap images and has more in common with a true
programming language rather than a file format. PostScript offers enormous flexibility
and the highest quality output, but this involves certain restrictions.

Computer Graphic Metafile

The CGM format was created by the American National Standards Institute as a universal
metafile specification. Although originally intended to include bitmaps, these are
generally not supported. The CGM file format is a useful intermediate format for
transferring vector images between different computer platforms and application
packages. It should be noted, however, that CGM does not support some features of
vector graphics and these will therefore be lost.
Windows Metafile

The WMF (Windows Metafile) is intended to support the vector drawing capabilities of
the Microsoft Windows interface. Although unique to this environment, it is almost
universally supported by Windows applications. The Windows Metafile serves as a
medium of exchange between various Windows applications and it can represent
anything that can be drawn directly through the Windows environment. Designed for
speed rather than compactness, the WMF format often produces very large files.

The Windows environment does not support certain features of vector images, notably
bezier curves or fountain fills. Such attributes will therefore be substituted. Bezier curves
are generally replaced by using short straight line approximations, which have the
advantage of faster display, although this can result in large file sizes. For this reason,
WMF files are rarely used as a primary file format, although they are a reliable export
path when the destination is a Windows application.

Macintosh PICT

The PICT QuickDraw picture format is a metafile format commonly used on the Apple
Macintosh platform for storing vector and or bitmap images. It is a native format of the
Macintosh Clipboard. It is additionally supported by some PC and UNIX applications to
provide some compatibility.

AutoCAD DXF

The AutoCAD DXF Drawing eXchange Format is supported by several applications

dealing with vector graphic models that can be manipulated in three-dimensional space.
The files are stored as numbers and commands in text form.

HPGL

The HPGL Hewlett Packard Graphics Language was designed to support pen plotters.
Since all HPGL code ultimately drives the same range of plotters, it is not liable to the
variations and inconsistencies that affect some file formats.
Bitmap images
File formats
A bitmap image simply contains a matrix of pixels, together with information about their
colour value. There are only so many ways in which this information can be represented
in binary form. Unfortunately, it varies from computer to computer. To decode a bitmap
it is therefore necessary to know the order in which the bits and bytes are stored. This
information is often coded in the file format.
Given below are brief details of some popular bitmap image file formats. The details
given are not intended to be comprehensive and are for general guidance only.

TIF

The TIFF Tagged Image File Format was jointly developed by Aldus and Microsoft for
use in desktop publishing and related graphics applications. The format deals with black-
and-white, greyscale, and colour images, including true colour images. It is well
supported on PC, Mac and Unix platforms and is highly suitable for the storage and
transfer of bitmap images and is often one of the best choices of bitmap format for data
exchange.

GIF

The CompuServe GIF (Graphics Interchange Format) is a popular bitmap image file
format employing data compression developed specifically for the storage and
transmission of images on the CompuServe network. The format includes LZW
compression and produces good lossless compression ratios.

PCX

The Zsoft PCX format is among the most commonly supported in the PC environment.
Originally developed as a proprietary format for use with the Zsoft Paintbrush program, it
established a de facto standard that became public and is very widely supported. It is
therefore a useful interchange format among PC applications.

BMP/DIB/RLE

The Microsoft Windows 3.x environment on the PC employs a special family of formats
for storage and display of bitmap images of 1, 4, 8 or 24 bits per pixel. Although the
bitmap file structures themselves are essentially similar, different filename extensions are
employed according to context. The file types used are BMP, DIB, and RLE, respectively
referring to BitMaP, Device Independent Bitmap, and Run Length Encoded bitmap..

TGA

The Targa TGA bitmap image format was developed by Truevision for their Targa and
Vista video hardware products on the PC and MAC. It is an industry standard although
not as widely supported as PCX or TIFF formats. The format provides support for run
length encoded compression.

EPS

The Adobe Encapsulated PostScript file format .

PCD

Kodak's PCD Photo CD format for digital photography is an important current

development in the field of bitmap images.
Photo CDs can be read by a computer with a suitable CD-ROM drive and appropriate
software.

IMG

The IMG format was developed by Digital Research as a bitmap format for use in the
GEM windowing environment. Supporting 1, 4 or 8 bits per pixel in monochrome

MAC (PNTG)

The MAC or PNTG bilevel bitmap format was the native bitmap image file format for
MacPaint on Apple Macintosh Computer. It is widely supported in the Mac environment.

RAS

The Sun Microsystems RAS Rasterfile is the native bitmap image format for Sun
workstations.

SGI

Silicon Graphics Inc. has a proprietary bitmap format used by much of their workstation
software.
Data compression
Large bitmaps require a large amount of storage. This makes heavy demands on
transmission and display systems. For this reason, various forms of data compression can
be used to minimise the storage requirements.
Data compression techniques work on the basis that there are redundant, repeating
elements in any data set. These repeating elements are identified and encoded to achieve
compression. The degree to which image data can be compressed is image dependent.
Images that contain lots of repeated information can be compressed more efficiently than
images containing lots of different information.
There are two fundamental types of image compression. Lossless compression allows the
image to be reconstituted exactly. Lossy compression discards data that is less important.
Both types of compression may be combined to yield high compression ratios.

Lossless data compression schemes generally involve the substitution of shorter codes for
common patterns in the image data, thus reducing the amount of storage space required.
This can work well for bilevel bitmaps or simple graphic images. However, such
conventional data-compression methods are not very effective for greyscale or colour
images. Higher resolution and pixel depth images do not generally display simple
arithmetic relationships or redundancies among nearby pixels. In the worst case, complex
images can result in negative compression, requiring more storage once data compression
has been applied.
The simplest form of data compression is known as Run Length Encoding or RLE. In
RLE a run of a certain value is represented by the length of the run, followed by the
value. Such a simple scheme works well for images with large areas of uniform value.

More complex algorithms include Huffman encoding, Lempel-Ziv Welch or LZW

compression, and Arithmetic compression. These use statistical methods to replace
common values with shorter codes, so achieving higher compression ratios of up to 3:1,
although 10:1 is possible for some images. Arithmetic compression comes close to the
theoretical limit for lossless data compression, although it has not been widely adopted
due to patent protection.

Lossy compression
While data compression involves coding data in such a way that it can be perfectly
reconstructed, lossy compression involves discarding the least significant data. Most
systems allow a threshold to be set so that the degree of loss and hence compression can
be varied.
Lossy compression is an irreversible operation in which information is lost. It is
therefore important to preserve an archive copy of the original file for future use. In
some circumstances, where the loss is not too great, the compressed file may be suitable
for archival purposes. Note that repeatedly reading and writing an image to and from a
file using a lossy compression system will result in progressive degradation

JPEG

Soon to be a dominant format for storage of photographic digital images, the published
JPEG system defined by the Joint Photographic Experts Group is an ISO standard for
compressing grey-scale or full colour digital images. Intended for hardware
implementation, JPEG is also supported in software on the PC, Mac, and workstation
platforms.
JPEG compression is proposed for inclusion in ISO Office Document Architecture,
CCITT colour facsimile, European ETSI videotex, and in version 6.0 of the TIFF bitmap
format. The standard defines the compression algorithm, but there are currently several
different file formats for storing JPEG compressed images. As a result many early
implementations are mutually incompatible.
The compression system makes use of the Discrete Cosine Transform. There is also a
lossless version that does not make use of the DCT, although it has little in common with
regular JPEG except for the name.
Most implementations start by converting the RGB images into a luminance/chrominance
colour space, that is, a greyscale base image plus two channels of colour difference
information, essentially encoding brightness, blueness and redness. Then a Discrete
Cosine Transformation is applied to convert the raster data into rate-of-change
information. Quantization truncates the results of the DCT coding to a smaller range of
values. This is the step that makes JPEG lossy. The quantization coefficients determine
just how much data is lost and hence the degree of compression and ultimate image
quality. Finally, the results of quantization are compressed using either Huffman or
arithmetic coding to produce the final output. Arithmetic coding is slightly more
compact, but due to patent restrictions, most current implementations use the public
domain Huffman scheme.
The lossy compression does not deal particularly well with images with a small number
of colours and sharp edges, such as cartoons. It is also limited in handling edge detail in
monochrome images. The system is therefore inappropriate for storing images intended
for machine analysis or for images where accuracy is critical such as applications in
medicine or fine art.
The level of loss and hence compression can be controlled, allowing a compromise
between image quality and file size. Usable compression ratios of up to 10:1 or 20:1 can
be achieved with minimal degradation. Effectively therefore, JPEG can usually construct
a full-colour picture nearly indistinguishable from the original, effectively using only
about 1 bit per pixel for storage.
At present there are various implementations of JPEG compression and there are a
number of different incompatible file formats.